diff --git "a/data/attention_queue.csv" "b/data/attention_queue.csv" new file mode 100644--- /dev/null +++ "b/data/attention_queue.csv" @@ -0,0 +1,18208 @@ +item_type,number,title,url,primary_bucket,attention_score,attention_reason,maintainer_next_action,updated_at +issue,188812,[MPS] MetalShaderLibrary::getBundledLibrary singleton can crash at process exit,https://github.com/pytorch/pytorch/issues/188812,needs triage,151,"triage signal present; open PR references this issue; high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: flaky-tests, module: macos, module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-04T18:31:40Z +issue,188900,`sparse.mul`: broadcasting a size-1 dimension of the sparse operand silently drops data,https://github.com/pytorch/pytorch/issues/188900,needs triage,139,"triage signal present; open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: sparse",review early; this can change release or regression risk,2026-07-06T17:28:34Z +issue,188323,[Inductor][CPU] dynamic=True convolution lowering crashes with ValueError: Exponent must be non-negative,https://github.com/pytorch/pytorch/issues/188323,needs triage,139,"triage signal present; open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: convolution, module: cpu, module: crash, module: dynamic shapes",review early; this can change release or regression risk,2026-06-30T17:57:26Z +issue,186535,"Windows, gloo: Access violation (0xC0000005) in ProcessGroupGloo::enqueue when calling allreduce on CUDA tensors — GlooAllreduceRegistry has no kCUDA creator",https://github.com/pytorch/pytorch/issues/186535,needs triage,139,"triage signal present; open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: multi-gpu, module: windows",review early; this can change release or regression risk,2026-06-07T10:58:03Z +issue,187912,[CPU] Concurrent `cpublas::brgemm` calls can crash in the AMX path when the underlying oneDNN ukernel is shared,https://github.com/pytorch/pytorch/issues/187912,needs triage,127,"triage signal present; high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: cpu, module: crash, module: intel, module: linear algebra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-06T17:07:04Z +issue,157668,NCCL error caused due to use of NVLS in torch 2.7.1-cu128 on aarch64 gb200 cluster,https://github.com/pytorch/pytorch/issues/157668,needs triage,127,"triage signal present; high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: arm, module: cuda, module: nccl, oncall: distributed",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-09T17:07:16Z +issue,150022,Dynamic Shapes with **kwargs,https://github.com/pytorch/pytorch/issues/150022,needs triage,121,"PyTorch issue lacks triaged label; open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, oncall: export, oncall: pt2",review early; this can change release or regression risk,2025-12-04T07:28:20Z +issue,162422,Runtime failure when running torch.compile() & using GCC 11.5.0 on Neoverse V1,https://github.com/pytorch/pytorch/issues/162422,needs triage,121,"PyTorch issue lacks triaged label; open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: crash, module: third_party, module: vectorization",review early; this can change release or regression risk,2025-11-30T04:27:08Z +issue,162476,heap-buffer-overflow in torch.quantized_max_pool2d via Python API,https://github.com/pytorch/pytorch/issues/162476,needs triage,121,"PyTorch issue lacks triaged label; open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: quantization",review early; this can change release or regression risk,2025-09-11T15:33:20Z +issue,189194,"[ROCm] torch 2.13 wheel: ""Can't detect vectorized ISA for CPU"" in torch.compile smoke test on non-ROCm image (regression vs 2.12.1)",https://github.com/pytorch/pytorch/issues/189194,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: binaries, module: inductor, module: regression, module: rocm",review early; this can change release or regression risk,2026-07-08T22:32:08Z +issue,189150,SIGSEGV / cudaErrorIllegalAddress when replaying a CUDA graph that captured multiple training iterations of two trainers over deeply unrolled recurrent modules,https://github.com/pytorch/pytorch/issues/189150,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: cuda graphs",review early; this can change release or regression risk,2026-07-08T19:25:06Z +issue,189282,"[CPU] operator_benchmark: backward-pass (add, + batchnorm on aarch64) ~200-500x slower since ~Nov 2025 (x86 + aarch64; possible measurement artifact)",https://github.com/pytorch/pytorch/issues/189282,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: cpu, module: performance, module: regression",review early; this can change release or regression risk,2026-07-08T17:39:26Z +issue,189281,[CPU] operator_benchmark: embedding / embeddingbag ~70-130x slower since ~Nov 2025 (x86 + aarch64; possible measurement issue),https://github.com/pytorch/pytorch/issues/189281,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: embedding, module: performance, module: regression",review early; this can change release or regression risk,2026-07-08T17:39:24Z +issue,189239,[Inductor]Unable to apply layout optimization on convolution backward causes training performance regression.,https://github.com/pytorch/pytorch/issues/189239,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: convolution, module: inductor, module: memory format, module: performance",review early; this can change release or regression risk,2026-07-08T06:33:50Z +issue,189144,[XPU][B580] `flex-attn-causal` performance drop with `last_level_cache_size` cache clear between runs,https://github.com/pytorch/pytorch/issues/189144,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: flex attention, module: inductor, module: performance, module: regression",review early; this can change release or regression risk,2026-07-08T02:27:43Z +issue,144965,"RuntimeError ""global alloc not supported yet"" when using TorchScript optimization.",https://github.com/pytorch/pytorch/issues/144965,needs triage,115,triage signal present; high-priority/blocker/regression label present; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2026-07-07T06:52:38Z +issue,189126,"Report a bug, as requested by the error message",https://github.com/pytorch/pytorch/issues/189126,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: memory usage",review early; this can change release or regression risk,2026-07-07T05:17:55Z +issue,189065,"CI for distributed tests on B200 fully red since ~2026-07-01: rank 0 ""CUDA-capable device(s) busy or unavailable"" -> NCCL id timeout -> 1170-min job timeout",https://github.com/pytorch/pytorch/issues/189065,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: ci, oncall: distributed",review early; this can change release or regression risk,2026-07-06T19:18:35Z +issue,170003,batch_isend_irecv with nccl causes illegal memory access depending on P2P ordering,https://github.com/pytorch/pytorch/issues/170003,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: nccl, oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-15T04:21:25Z +issue,174288,[distributed] Batched isend/irecv with NCCL backend hangs on high load,https://github.com/pytorch/pytorch/issues/174288,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: vllm, oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-12T05:49:56Z +issue,162731,DTensor cached op propagation results can be mutated when propagating other ops,https://github.com/pytorch/pytorch/issues/162731,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",review early; this can change release or regression risk,2026-06-11T17:25:13Z +issue,153517,[CI][CUDA][Distributed] test_non_blocking_with_eager_init timeout,https://github.com/pytorch/pytorch/issues/153517,needs triage,115,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; component/oncall labels: module: ci, module: nccl, module: regression, oncall: distributed",review early; this can change release or regression risk,2026-06-10T17:14:11Z +issue,167693,DISABLED test_side_stream_backward_overlap_cuda (__main__.TestAutogradStreamSynchronizationCUDA),https://github.com/pytorch/pytorch/issues/167693,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: flaky-tests, module: rocm",review early; this can change release or regression risk,2026-06-10T08:01:37Z +issue,116423,PyTorch Distributed Elastic Launch Segmentation Fault with Python 3.12,https://github.com/pytorch/pytorch/issues/116423,needs triage,115,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: elastic, oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-07T16:39:19Z +issue,106164,distributed.batch_isend_irecv() crash when send/recv refers to itself,https://github.com/pytorch/pytorch/issues/106164,needs triage,115,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; component/oncall labels: module: c10d, module: crash, oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-07T16:36:35Z +issue,85088,reentrant torch.utils.checkpoint does not work with NamedTuple outputs,https://github.com/pytorch/pytorch/issues/85088,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: activation checkpointing, oncall: distributed, oncall: distributed parallelisms",review early; this can change release or regression risk,2026-06-07T16:35:52Z +issue,71303,[RFC] Cross-Process Performance Analysis: Straggler Detection,https://github.com/pytorch/pytorch/issues/71303,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: c10d, module: ddp, module: distributed_tool, oncall: distributed",review early; this can change release or regression risk,2026-06-06T16:30:33Z +issue,75725,Warning originating in C10 backend does not get translated to Python warning if run from subprocess,https://github.com/pytorch/pytorch/issues/75725,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",review early; this can change release or regression risk,2026-06-06T16:28:36Z +issue,76449,Enhance _verify_param_shape_across_processes,https://github.com/pytorch/pytorch/issues/76449,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: ddp, module: fsdp, oncall: distributed, oncall: distributed parallelisms",review early; this can change release or regression risk,2026-06-06T16:28:02Z +issue,81684,Message exchange failure when perform alltoallv (cpus),https://github.com/pytorch/pytorch/issues/81684,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-05T17:11:38Z +issue,119845,Segmentation fault in dataloader worker sub-process,https://github.com/pytorch/pytorch/issues/119845,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dataloader, module: ddp, oncall: distributed",review early; this can change release or regression risk,2026-06-05T17:06:01Z +issue,183459,Regression in torch.distributed._functional_collectives.all_to_all_single in 2.9.1 -> 2.11,https://github.com/pytorch/pytorch/issues/183459,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-05T16:51:57Z +issue,180088,"[DTensor] RNG tracker does not advance state for CPU tensors on CUDA mesh, causing trunc_normal_ infinite loop",https://github.com/pytorch/pytorch/issues/180088,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",review early; this can change release or regression risk,2026-05-28T14:11:57Z +issue,173921,[Inductor] Significant numerical divergence (4.7% relative error) in Conv2d on CPU with torch.compile,https://github.com/pytorch/pytorch/issues/173921,needs triage,115,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",review early; this can change release or regression risk,2026-05-21T18:17:56Z +issue,179502,[DTensor] sharded view incorrectly passes when redistribution is needed,https://github.com/pytorch/pytorch/issues/179502,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: dtensor, oncall: distributed",review early; this can change release or regression risk,2026-04-06T22:51:39Z +issue,145610,mmap fails on 64k page aarch64 systems for AOTI model loading,https://github.com/pytorch/pytorch/issues/145610,needs triage,115,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; component/oncall labels: module: aotinductor, module: arm, module: crash, oncall: export",review early; this can change release or regression risk,2026-02-17T08:26:10Z +issue,66504,BatchNorm runtimeError: one of the variables needed for gradient computation has been modified by an inplace operation,https://github.com/pytorch/pytorch/issues/66504,needs triage,115,"triage signal present; high-priority/blocker/regression label present; component/oncall labels: module: ddp, oncall: distributed",review early; this can change release or regression risk,2026-02-14T06:22:10Z +pr,188632,Fix Dynamo opaque object staticmethod guards,https://github.com/pytorch/pytorch/pull/188632,ready for maintainer decision,112,"needs-review label present; approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-09T01:33:58Z +pr,186429,Fix ONNX export of pad_sequence with symbolic split lengths,https://github.com/pytorch/pytorch/pull/186429,ready for maintainer decision,112,"needs-review label present; approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2026-07-08T18:59:33Z +pr,186461,Fix SymBool equality handling in symbolic shapes,https://github.com/pytorch/pytorch/pull/186461,ready for maintainer decision,112,needs-review label present; approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-08T18:17:51Z +pr,184716,Fix conv_transpose2d meta output padding validation,https://github.com/pytorch/pytorch/pull/184716,ready for maintainer decision,112,needs-review label present; approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-08T07:16:54Z +pr,185691,Avoid UB in float to signed integer casts,https://github.com/pytorch/pytorch/pull/185691,ready for maintainer decision,112,needs-review label present; approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-07T10:51:55Z +pr,184939,Support SymInt steps for linspace/logspace export,https://github.com/pytorch/pytorch/pull/184939,ready for maintainer decision,112,needs-review label present; approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-18T10:35:58Z +pr,189307,[CI/CD] Copy newer CUPTI headers into manywheel binary-build images,https://github.com/pytorch/pytorch/pull/189307,PR blocked,108,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks; CI lane labels: ciflow/docker,final maintainer merge/release decision,2026-07-09T01:53:06Z +pr,186927,[MPS] Gemv kernels,https://github.com/pytorch/pytorch/pull/186927,PR blocked,108,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks; CI lane labels: ciflow/mps,final maintainer merge/release decision,2026-07-09T00:28:47Z +pr,189312,[CUDA][cuBLAS] Change cuBLAS default workspace size for SM 11.0 to 32 MiB,https://github.com/pytorch/pytorch/pull/189312,PR blocked,108,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-08T23:49:19Z +pr,188301,[inductor] Fix inductor dropping ordering dep between effectful ops with different kernel types,https://github.com/pytorch/pytorch/pull/188301,PR blocked,108,"failed checks or blocked/unstable merge state; approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T18:49:35Z +pr,184648,[Test] Add has_sufficient_memory() hook to DeviceTypeTestBase,https://github.com/pytorch/pytorch/pull/184648,PR blocked,108,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-07T17:58:35Z +pr,188237,Optimize CSR SpMM CPU grain size and inner accumulation,https://github.com/pytorch/pytorch/pull/188237,PR blocked,108,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-07T14:34:05Z +pr,187501,[Stable C Shim] Use error message retrieval shim if available at runtime,https://github.com/pytorch/pytorch/pull/187501,PR blocked,108,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-25T16:14:35Z +pr,185882,[PrivateUse1] Return `False` instead of `None` when PU1 backend is not available,https://github.com/pytorch/pytorch/pull/185882,PR blocked,108,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-23T15:23:08Z +issue,153410,[Feature request] `torch.export` .save/.load could support `safetensors` and/or `weights_only=True`,https://github.com/pytorch/pytorch/issues/153410,needs triage,107,"PyTorch issue lacks triaged label; open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-01-18T09:26:05Z +issue,188892,nvidia-cudnn-cu13==9.20.0.48 bundled by torch is incomplete → CUDNN_STATUS_SUBLIBRARY_VERSION_MISMATCH,https://github.com/pytorch/pytorch/issues/188892,has linked PR,106,"open PR references this issue; high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: binaries, module: convolution, module: crash, module: cudnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-07T09:16:01Z +issue,188970,[MPS] SIGSEGV in MPSStream::copy with triton.cudagraphs (reduce-overhead/max-autotune) on deep fp16 graphs,https://github.com/pytorch/pytorch/issues/188970,has linked PR,106,"open PR references this issue; high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: inductor, module: macos, module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-07T05:43:13Z +pr,135631,[scan] Autograd,https://github.com/pytorch/pytorch/pull/135631,PR blocked,100,failed checks or blocked/unstable merge state; approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2025-02-03T00:37:41Z +pr,188115,[Inductor][TEST] Align `test_main_loop_scaling` with H100 support surface,https://github.com/pytorch/pytorch/pull/188115,PR blocked,98,"failed checks or blocked/unstable merge state; high-priority/blocker/regression label present; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-03T02:51:54Z +issue,149447,DTensor slicing on sharded dimension leads to replication,https://github.com/pytorch/pytorch/issues/149447,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",review early; this can change release or regression risk,2025-12-20T20:52:34Z +issue,164929,get_optimizer_state_dict modifies optimizer state,https://github.com/pytorch/pytorch/issues/164929,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",review early; this can change release or regression risk,2025-12-10T02:42:02Z +issue,156563,Segmentation fault (core dumped) in `torch.profiler.profile`,https://github.com/pytorch/pytorch/issues/156563,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking, oncall: profiler",review early; this can change release or regression risk,2025-11-19T18:16:25Z +issue,122019,`F.max_pool3d`: Segfault due to incorrect `dilation` check in kernel,https://github.com/pytorch/pytorch/issues/122019,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: quantization",review early; this can change release or regression risk,2025-09-26T23:19:29Z +issue,159238,Performance regression: torch.jit.trace() significantly slower on RTX 5090 than RTX 4060 (cu128 nightly),https://github.com/pytorch/pytorch/issues/159238,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, oncall: jit",review early; this can change release or regression risk,2025-08-09T16:07:58Z +issue,121219,Torch profiler corrupted names with Python 3.11,https://github.com/pytorch/pytorch/issues/121219,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,review early; this can change release or regression risk,2025-08-07T04:30:15Z +issue,91165,[FSDP] FSDP with CPU offload consumes `1.65X` more GPU memory when training models with most of the params frozen,https://github.com/pytorch/pytorch/issues/91165,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",review early; this can change release or regression risk,2025-08-04T00:11:21Z +issue,47260,DDP doesn't work with retain_graph = True,https://github.com/pytorch/pytorch/issues/47260,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2025-06-09T11:34:58Z +issue,100253,profiler.export_stacks doesn't return stack trace unless experimental_config is provided,https://github.com/pytorch/pytorch/issues/100253,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, oncall: profiler",review early; this can change release or regression risk,2025-06-08T23:12:36Z +issue,155147,torch.profiler raises Aborted (core dumped) failurer related with GIL (gilstate_tss_set),https://github.com/pytorch/pytorch/issues/155147,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: wsl, oncall: profiler",review early; this can change release or regression risk,2025-06-06T07:45:23Z +issue,155146,Segmentation fault when using torch.profiler,https://github.com/pytorch/pytorch/issues/155146,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: wsl, oncall: profiler",review early; this can change release or regression risk,2025-06-05T22:41:29Z +issue,153941,TorchScript Fails Type Inference When Using .item()-Derived Scalar in Tensor Construction,https://github.com/pytorch/pytorch/issues/153941,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, oncall: jit",review early; this can change release or regression risk,2025-05-20T14:07:31Z +issue,153260,Export shouldn't run TS under the hood.,https://github.com/pytorch/pytorch/issues/153260,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: export, oncall: jit, oncall: pt2",review early; this can change release or regression risk,2025-05-13T17:23:02Z +issue,153145,Segmentation faults with CPU FlexAttention,https://github.com/pytorch/pytorch/issues/153145,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: flex attention, module: higher order operators",review early; this can change release or regression risk,2025-05-09T06:36:24Z +issue,152108,[inductor][cpu] AMP static shape default wrapper AOTInductor performance regression in 2025_04_20 nightly release,https://github.com/pytorch/pytorch/issues/152108,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, oncall: cpu inductor, oncall: pt2",review early; this can change release or regression risk,2025-04-24T23:33:51Z +issue,145088,[torchbench] torch._dynamo.exc.Unsupported: Graph break due to unsupported builtin None.morphologyEx,https://github.com/pytorch/pytorch/issues/145088,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export, oncall: pt2",review early; this can change release or regression risk,2025-03-05T18:26:38Z +issue,146788,Segmentation Fault in `torch.choose_qparams_optimized` with Invalid Parameters,https://github.com/pytorch/pytorch/issues/146788,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: quantization",review early; this can change release or regression risk,2025-02-09T20:17:13Z +issue,124950,[distributed] First NCCL barrier does not respect timeout,https://github.com/pytorch/pytorch/issues/124950,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2025-01-30T00:43:40Z +issue,71855,Regression in multi-node training speed with Transformers + PyTorch,https://github.com/pytorch/pytorch/issues/71855,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: regression, oncall: distributed",review early; this can change release or regression risk,2024-11-27T19:48:40Z +issue,136712,Segmentation fault (core dumped) in `torch.ao.nn.quantized.dynamic.LSTMCell/GRUCell`,https://github.com/pytorch/pytorch/issues/136712,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: quantization",review early; this can change release or regression risk,2024-09-27T01:12:24Z +issue,136728,Aborted (core dumped) in `torch.package.package_exporter.PackageExporter`/`torch.package.PackageExporter`,https://github.com/pytorch/pytorch/issues/136728,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: package/deploy",review early; this can change release or regression risk,2024-09-26T03:56:33Z +issue,135835,A segmentation fault will be raised when using the `torch._C._jit_to_backend`,https://github.com/pytorch/pytorch/issues/135835,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: jit",review early; this can change release or regression risk,2024-09-12T21:00:13Z +issue,81085,RuntimeError: required keyword attribute 'value' is undefined,https://github.com/pytorch/pytorch/issues/81085,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2024-09-09T23:14:38Z +issue,128693,Segmentation fault (core dumped) in `torch.fused_moving_avg_obs_fake_quant`,https://github.com/pytorch/pytorch/issues/128693,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, oncall: quantization",review early; this can change release or regression risk,2024-07-22T17:28:53Z +issue,83606,"test_profiler_experimental_tree_cuda_detailed is too unstable, and as its CUDA only difficult to regen",https://github.com/pytorch/pytorch/issues/83606,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: profiler",review early; this can change release or regression risk,2024-07-16T13:34:20Z +issue,103947,"[torch.fx] Deserialization Error - TypeError: ones() received an invalid combination of arguments - got (tuple, device=Attribute)",https://github.com/pytorch/pytorch/issues/103947,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: fx",review early; this can change release or regression risk,2024-02-05T12:28:08Z +issue,68407,[Runtime Error in ddp_pipeline.py] RuntimeError: unsupported operation: some elements of the input tensor and the written-to tensor refer to a single memory location. Please clone() the tensor before performing the operation.,https://github.com/pytorch/pytorch/issues/68407,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: docs, oncall: distributed",review early; this can change release or regression risk,2024-01-08T22:50:47Z +issue,116351,`torch.quantize_per_channel`: FPE,https://github.com/pytorch/pytorch/issues/116351,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: quantization",review early; this can change release or regression risk,2024-01-05T21:55:02Z +issue,34329,[jit] Hook support tracking,https://github.com/pytorch/pytorch/issues/34329,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2024-01-05T18:39:51Z +issue,113572,Deepcopying an exported model changes numerics for MobileNetV2,https://github.com/pytorch/pytorch/issues/113572,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: export,review early; this can change release or regression risk,2024-01-03T22:07:07Z +issue,115818,Fx graph recompilation along with partitioner produces different node order even though graph operations are same,https://github.com/pytorch/pytorch/issues/115818,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: fx,review early; this can change release or regression risk,2024-01-03T07:58:33Z +issue,115640,"Quantisation results in a ""Illegal instruction (core dumped)""",https://github.com/pytorch/pytorch/issues/115640,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: quantization",review early; this can change release or regression risk,2023-12-21T15:18:12Z +issue,73602,Report DDP efficiency metric to users and guide them for setting up DDP correctly,https://github.com/pytorch/pytorch/issues/73602,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2023-11-07T16:38:55Z +issue,83289,Improvements to ProcessGroupGloo monitored_barrier,https://github.com/pytorch/pytorch/issues/83289,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",review early; this can change release or regression risk,2023-10-19T20:27:58Z +issue,80151,activation checkpointing with non_reentrant implementation memory leaks,https://github.com/pytorch/pytorch/issues/80151,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2023-05-23T04:52:08Z +issue,64345,Profiler UTF-8 decode issue,https://github.com/pytorch/pytorch/issues/64345,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,review early; this can change release or regression risk,2023-02-03T14:07:37Z +issue,86594,JIT model with mean will crash when computing the gradients on cuda,https://github.com/pytorch/pytorch/issues/86594,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: jit",review early; this can change release or regression risk,2022-10-10T18:33:46Z +issue,86595,JIT model with `relu+div+sgn` will crash when computing the gradient,https://github.com/pytorch/pytorch/issues/86595,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: functorch, oncall: jit",review early; this can change release or regression risk,2022-10-10T15:28:27Z +issue,86532,Conv2d will crash by using `jit.trace`,https://github.com/pytorch/pytorch/issues/86532,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: jit",review early; this can change release or regression risk,2022-10-10T15:25:01Z +issue,67646,test_jit_fuser_legacy and test_jit_fuser_te are failing with SIGIOT,https://github.com/pytorch/pytorch/issues/67646,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: jit",review early; this can change release or regression risk,2022-10-08T01:26:04Z +issue,65357,Improve DDP tutorial,https://github.com/pytorch/pytorch/issues/65357,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: docs, oncall: distributed",review early; this can change release or regression risk,2022-07-27T19:47:25Z +issue,81115,torch.package can not be used to serialize `resnet18` from TorchVision-0.12,https://github.com/pytorch/pytorch/issues/81115,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, module: vision, oncall: package/deploy",review early; this can change release or regression risk,2022-07-11T17:20:29Z +issue,80349,Distributed Store `get` doesn't work well with `add`,https://github.com/pytorch/pytorch/issues/80349,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: docs, oncall: distributed",review early; this can change release or regression risk,2022-06-29T14:26:43Z +issue,74115,DistributedDataParallel high peak memory usage with find_unused_parameters=True,https://github.com/pytorch/pytorch/issues/74115,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",review early; this can change release or regression risk,2022-06-08T23:47:29Z +issue,72859,Profiler crashes with ProfilerActivity.CUDA on AWS p4d.24xlarge with A100-SXM4-40GB,https://github.com/pytorch/pytorch/issues/72859,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: profiler",review early; this can change release or regression risk,2022-04-13T21:42:32Z +issue,69179,DISABLED test_builtin_remote_message_dropped_timeout_to_self (__main__.FaultyFaultyAgentRpcTest),https://github.com/pytorch/pytorch/issues/69179,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: rpc, oncall: distributed",review early; this can change release or regression risk,2022-03-25T22:16:25Z +issue,73885,operation not supported crash when initializing RPC tensorpipe,https://github.com/pytorch/pytorch/issues/73885,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: distributed",review early; this can change release or regression risk,2022-03-17T13:18:23Z +issue,58111,Static graph training fails if forward is called multiple times before backward,https://github.com/pytorch/pytorch/issues/58111,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",review early; this can change release or regression risk,2022-03-04T04:50:33Z +issue,67389,Missing doc for torch.distributed functions,https://github.com/pytorch/pytorch/issues/67389,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: distributed",review early; this can change release or regression risk,2022-03-04T04:36:47Z +issue,73352,Segfault on unloading a model,https://github.com/pytorch/pytorch/issues/73352,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: crash, oncall: jit",review early; this can change release or regression risk,2022-02-25T04:38:55Z +issue,63917,Add tests for DDP broadcast_buffers=True,https://github.com/pytorch/pytorch/issues/63917,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: ddp, oncall: distributed",review early; this can change release or regression risk,2021-12-21T19:49:21Z +issue,67541,TestBenchNetwork::test_forward[resnet50_jit-profiling-te] is failing intermittently,https://github.com/pytorch/pytorch/issues/67541,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2021-10-29T15:56:37Z +issue,47218,[JIT] Alias Annotations for Native Schema List Ops Is Subtly Wrong,https://github.com/pytorch/pytorch/issues/47218,needs triage,97,PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2021-10-22T20:44:46Z +issue,66894,test_nccl_barrier_timeout_new_group_non_member fails intermittently,https://github.com/pytorch/pytorch/issues/66894,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",review early; this can change release or regression risk,2021-10-19T20:21:54Z +issue,54524,Store created by dist.new_group doesn't appear to respect timeout,https://github.com/pytorch/pytorch/issues/54524,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2021-10-14T00:42:36Z +issue,39351,Clean up RPC request callback implementation,https://github.com/pytorch/pytorch/issues/39351,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",review early; this can change release or regression risk,2021-10-13T21:33:23Z +issue,64518,TensorPipeDistAutogradTest is frequently failing,https://github.com/pytorch/pytorch/issues/64518,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",review early; this can change release or regression risk,2021-10-12T20:43:42Z +issue,62780,Seeing unit test failures for ProcessGroupShareTensorTest,https://github.com/pytorch/pytorch/issues/62780,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: nccl, oncall: distributed",review early; this can change release or regression risk,2021-08-17T14:25:11Z +issue,58856,NCCL multi-gpu test intermittently failing after NCCL version upgrade,https://github.com/pytorch/pytorch/issues/58856,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: nccl, oncall: distributed",review early; this can change release or regression risk,2021-07-21T19:25:48Z +issue,55905,Attempt to use jited `torch.isnan` hit internal assert,https://github.com/pytorch/pytorch/issues/55905,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: crash, oncall: jit",review early; this can change release or regression risk,2021-04-15T05:29:50Z +issue,40867,[JIT] Infer type of argument foo = None to be Optional[Tensor],https://github.com/pytorch/pytorch/issues/40867,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2021-03-26T07:31:24Z +issue,47999,Torch DDP seems not prefer IPv6 over IPv4?,https://github.com/pytorch/pytorch/issues/47999,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2021-03-15T17:45:33Z +issue,45549,Can't initialize NCCL/GLOO process group if default process group is MPI,https://github.com/pytorch/pytorch/issues/45549,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: nccl, oncall: distributed",review early; this can change release or regression risk,2021-02-11T02:14:42Z +issue,28786,RPC couldn't match torch.ones with requires_grad=True,https://github.com/pytorch/pytorch/issues/28786,needs triage,97,"triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: internals, module: rpc, oncall: distributed",review early; this can change release or regression risk,2020-12-07T18:10:50Z +issue,46166,Error with DistributedDataParallel with specific model,https://github.com/pytorch/pytorch/issues/46166,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2020-12-07T02:00:04Z +issue,34116,autograd with TorchScript does not match finite differences,https://github.com/pytorch/pytorch/issues/34116,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-10-27T17:14:58Z +issue,41289,List[Any] support in TorchScript,https://github.com/pytorch/pytorch/issues/41289,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-10-11T18:45:01Z +issue,43548,JIT tracing incorrectly records some slice bounds,https://github.com/pytorch/pytorch/issues/43548,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-09-17T17:17:29Z +issue,42258,Cannot load certain function from dumped Torchscript file,https://github.com/pytorch/pytorch/issues/42258,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-08-17T21:51:35Z +issue,42317,Instance norm annotation is incorrect,https://github.com/pytorch/pytorch/issues/42317,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-07-30T16:34:37Z +issue,41695,Improve overload resolution order rules for `invokeOperatorFromPython`,https://github.com/pytorch/pytorch/issues/41695,needs triage,97,PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-07-20T21:58:55Z +issue,35998,JITed GRU too slow,https://github.com/pytorch/pytorch/issues/35998,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-06-10T22:11:44Z +issue,37015,Converting to Torch Script: cpp_module does not match nn_module,https://github.com/pytorch/pytorch/issues/37015,needs triage,97,triage signal present; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-04-29T17:35:29Z +issue,7036,[Caffe2] compilation linker error `libtbb.so.2: undefined reference to std::__exception_ptr`,https://github.com/pytorch/pytorch/issues/7036,needs triage,97,"PyTorch issue lacks triaged label; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: crash",review early; this can change release or regression risk,2018-04-27T17:12:50Z +issue,189267,nn.functional.threshold OpInfo reference raises OverflowError on unsigned dtypes with NumPy >= 2.5,https://github.com/pytorch/pytorch/issues/189267,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: nn, module: numpy, module: regression, module: tests",review early; this can change release or regression risk,2026-07-08T16:58:04Z +issue,189034,Limited CI on H100 (smoke) red since 2026-07-03: test_foreach import crash from NumPy 2.x / SciPy ABI mismatch,https://github.com/pytorch/pytorch/issues/189034,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: ci, module: regression",review early; this can change release or regression risk,2026-07-06T17:33:27Z +issue,188792,CPU float unary ops can silently return low-precision results: MKL VML first-call kernel dispatch race,https://github.com/pytorch/pytorch/issues/188792,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: cpu, module: intel, module: mkl",review early; this can change release or regression risk,2026-07-06T17:25:49Z +issue,188492,"swin_base_patch4_window7_224 fails inductor accuracy (bf16 inference, H100) after Triton 3.8 pin bump",https://github.com/pytorch/pytorch/issues/188492,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: bfloat16, module: correctness (silent), module: cuda, module: dependency bug",review early; this can change release or regression risk,2026-07-06T17:23:27Z +issue,188358,[Dynamo] Eager and torch.compile behavior mismatch in F.batch_norm with multidimensional bias,https://github.com/pytorch/pytorch/issues/188358,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-07-06T17:20:16Z +issue,188545,[PT2] Output mismatch between eager and torch.compile for torch.special.bessel_y0 with +inf input on CUDA,https://github.com/pytorch/pytorch/issues/188545,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: cuda, module: edge cases, module: nans and infs",review early; this can change release or regression risk,2026-07-06T17:12:08Z +issue,188541,"[Inductor][CUDA] F.logsigmoid flushes negative float32 subnormal outputs to +0.0, changing sign-sensitive behavior",https://github.com/pytorch/pytorch/issues/188541,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: cuda, module: edge cases, module: inductor",review early; this can change release or regression risk,2026-07-06T17:09:07Z +issue,188033,[BUG] THPStream_richcompare casts arbitrary objects to THPStream* without type check,https://github.com/pytorch/pytorch/issues/188033,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: error checking, module: pybind",review early; this can change release or regression risk,2026-07-06T17:08:05Z +issue,188008,expandable_segments: partial mapAndSetAccess failure → bulk cuMemUnmap of unmapped range → terminate in destructor,https://github.com/pytorch/pytorch/issues/188008,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: cudacachingallocator",review early; this can change release or regression risk,2026-07-06T17:07:36Z +issue,188547,[dynamo] Infinite recursion when returning a Pybind11 enum from a compiled function,https://github.com/pytorch/pytorch/issues/188547,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dynamo, module: pybind, oncall: pt2",review early; this can change release or regression risk,2026-07-06T17:02:39Z +issue,188890,Non-deterministic gradients when using torch.compile and PyTorch >= 2.12,https://github.com/pytorch/pytorch/issues/188890,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: determinism, module: regression, oncall: pt2",review early; this can change release or regression risk,2026-07-06T15:19:14Z +issue,148317,compile SageAttention faing error C2872: “std” for latest torch nightly,https://github.com/pytorch/pytorch/issues/148317,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: regression, module: windows, oncall: pt2",review early; this can change release or regression risk,2026-07-05T04:05:19Z +issue,177116,MPS: catastrophically wrong gradients in backward pass (>32K elements),https://github.com/pytorch/pytorch/issues/177116,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: mps",review early; this can change release or regression risk,2026-07-04T08:42:57Z +issue,188578,[Inductor] Optimize split reduction heuristics on GB200,https://github.com/pytorch/pytorch/issues/188578,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: inductor, module: performance, module: regression, oncall: pt2",review early; this can change release or regression risk,2026-06-30T18:40:02Z +issue,188231,[inductor] `TypeError: A Boolean argument can only be used in Eq and Ne` during CPP codegen,https://github.com/pytorch/pytorch/issues/188231,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-29T19:32:45Z +issue,188198,[MPS] cat op dropped performance in torch==2.10 release,https://github.com/pytorch/pytorch/issues/188198,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: mps, module: performance, module: regression",review early; this can change release or regression risk,2026-06-29T18:21:12Z +issue,187988,[MPS] linear backward crashes (SIGABRT) with a 1-D weight: 'mps.matmul' op contracting dimensions differ,https://github.com/pytorch/pytorch/issues/187988,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: arm, module: autograd, module: crash, module: macos",review early; this can change release or regression risk,2026-06-29T18:01:10Z +issue,188034,[BUG] Reference leak in autograd Function.apply — setup_context return value never released,https://github.com/pytorch/pytorch/issues/188034,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: memory usage, module: pybind",review early; this can change release or regression risk,2026-06-29T17:41:30Z +issue,188023,torch.from_dlpack aborts the Python process for NumPy arrays with negative strides,https://github.com/pytorch/pytorch/issues/188023,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dlpack, module: error checking, module: numpy",review early; this can change release or regression risk,2026-06-29T17:40:54Z +issue,188261,[inductor][XPU] PR #184366 causes AMP fp16 training accuracy regression on XPU,https://github.com/pytorch/pytorch/issues/188261,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: amp (automated mixed precision), module: correctness (silent), module: half, module: inductor",review early; this can change release or regression risk,2026-06-29T17:29:49Z +issue,188113,CK SDPA flash attention kernel asserts on gfx1200/gfx1201 (RDNA4),https://github.com/pytorch/pytorch/issues/188113,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: rocm, module: sdpa",review early; this can change release or regression risk,2026-06-26T23:57:15Z +issue,188069,"torch.compile(mode=""max-autotune"") crashes with too many positional arguments during bmm autotuning when template inputs are deduplicated",https://github.com/pytorch/pytorch/issues/188069,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-26T12:55:56Z +issue,185886,[torch.compile] Clarification/regression: Dynamo cache lookup guard eval is much slower in 2.12 vs 2.7 after `inline_inbuilt_nn_modules` became always True,https://github.com/pytorch/pytorch/issues/185886,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: dynamo, module: guards, module: performance, module: regression",review early; this can change release or regression risk,2026-06-24T08:24:58Z +issue,185484,[Inductor] Forward output mismatch between Eager and Inductor for F.softshrink with bfloat16 and float scalar,https://github.com/pytorch/pytorch/issues/185484,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: bfloat16, module: correctness (silent), module: inductor, module: type promotion",review early; this can change release or regression risk,2026-06-23T17:29:18Z +issue,186084,"[inductor] insert_overlap_deps wraps non-overlap-critical compute with control_deps, blocking fusion and causing silent bf16 accuracy regression",https://github.com/pytorch/pytorch/issues/186084,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-23T17:23:02Z +issue,187415,[Inductor] torch.compile fails on CUDA with bool mask indexing after torch.eye constructor movement: AssertionError in _rename_unbacked_to,https://github.com/pytorch/pytorch/issues/187415,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-22T15:15:37Z +issue,187340,[torch.compile][Inductor][CUDA] torch.nextafter is incorrectly compiled as no-op for non-contiguous float16 tensors,https://github.com/pytorch/pytorch/issues/187340,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-22T15:15:28Z +issue,187567,[MPS] torch.linalg.qr is 25-250× slower than CPU on tall-skinny matrices,https://github.com/pytorch/pytorch/issues/187567,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: linear algebra, module: macos, module: mps, module: performance",review early; this can change release or regression risk,2026-06-22T15:14:47Z +issue,187399,[MPS] torch.baddbmm — INTERNAL ASSERT for size-0 batch dimensions,https://github.com/pytorch/pytorch/issues/187399,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: edge cases, module: linear algebra, module: mps",review early; this can change release or regression risk,2026-06-22T15:14:30Z +issue,187429,"torch.where silently casts out-of-range Python float to +/-Inf in fp16 (inconsistent with full_like / full / fill_, which raise)",https://github.com/pytorch/pytorch/issues/187429,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: edge cases, module: error checking, module: half",review early; this can change release or regression risk,2026-06-22T15:13:08Z +issue,187027,flip after repeat_interleave generates negative (out-of-bounds) index under dynamic shapes,https://github.com/pytorch/pytorch/issues/187027,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: dynamic shapes, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-18T20:46:56Z +issue,187284,`torch.compile` with `inductor` backend silently drops forward-mode AD tangents (returns `None`),https://github.com/pytorch/pytorch/issues/187284,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: correctness (silent), module: dynamo, module: forward ad",review early; this can change release or regression risk,2026-06-15T22:11:36Z +issue,186537,torch 2.12 regression: torch.compile thread-local config - even backward thread has separate config,https://github.com/pytorch/pytorch/issues/186537,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: bc-breaking, module: dynamo, module: multithreading, module: regression",review early; this can change release or regression risk,2026-06-14T18:03:12Z +issue,162358,heap-buffer-overflow in torch.linalg.eig,https://github.com/pytorch/pytorch/issues/162358,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: error checking, module: linear algebra, module: sanitizers",review early; this can change release or regression risk,2026-06-14T16:50:12Z +issue,160018,Triton kernel generated for Flex Attention throws InductorError: RuntimeError: No valid triton configs when seq length < 128,https://github.com/pytorch/pytorch/issues/160018,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: flex attention, module: higher order operators, module: pt2-dispatcher",review early; this can change release or regression risk,2026-06-13T23:19:37Z +issue,175370,`F.embedding_bag` segfaults with float64 weight and empty offsets,https://github.com/pytorch/pytorch/issues/175370,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: crash, module: embedding, module: empty tensor",review early; this can change release or regression risk,2026-06-12T00:34:42Z +issue,183806,`torch.linalg.eig` and `torch.linalg.eigh` 30-130x slower on 2.11.0+cu130 vs 2.5.1+cu124,https://github.com/pytorch/pytorch/issues/183806,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: cuda, module: linear algebra, module: magma, module: performance",review early; this can change release or regression risk,2026-06-11T10:06:06Z +issue,41508,nn.MultiheadAttention causes gradients to become NaN under some use cases,https://github.com/pytorch/pytorch/issues/41508,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: nans and infs, module: nn",review early; this can change release or regression risk,2026-06-09T11:38:12Z +issue,186256,`native_batch_norm` silently produces incorrect 3rd+ order derivatives in training mode with `create_graph=True`,https://github.com/pytorch/pytorch/issues/186256,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: correctness (silent), module: derivatives, module: double backwards",review early; this can change release or regression risk,2026-06-09T11:32:06Z +issue,185715,INTERNAL ASSERT FAILED: NYI SymInt equality in c10/core/Scalar.h when using torch.compile(dynamic=True) and torch.autograd.grad with torch.pow,https://github.com/pytorch/pytorch/issues/185715,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: crash, module: dynamic shapes, oncall: pt2",review early; this can change release or regression risk,2026-06-08T17:11:30Z +issue,186241,TMA descriptor loads use stale `xoffset` in mix-order reduction loop,https://github.com/pytorch/pytorch/issues/186241,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: crash, module: inductor, module: reductions",review early; this can change release or regression risk,2026-06-08T15:57:02Z +issue,186453,"INTERNAL ASSERT FAILED at ""/pytorch/c10/core/AllocatorConfig.h"":124 on python torch import",https://github.com/pytorch/pytorch/issues/186453,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cudacachingallocator",review early; this can change release or regression risk,2026-06-08T14:44:46Z +issue,186233,C++ API: at::range crashes due to division-by-zero in RangeFactories.cpp,https://github.com/pytorch/pytorch/issues/186233,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: cpp, module: crash, module: edge cases, module: tensor creation",review early; this can change release or regression risk,2026-06-08T14:25:45Z +issue,185482,[inductor]`torch.cond` Branch with sliced view and scatter_add crashes with non-tensor input,https://github.com/pytorch/pytorch/issues/185482,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-02T21:02:48Z +issue,185247,Segmentation fault in `torch.linalg.eigvals` triggered by invalid `avg_pool3d` divisor,https://github.com/pytorch/pytorch/issues/185247,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: error checking, module: linear algebra, module: pooling",review early; this can change release or regression risk,2026-06-01T17:18:02Z +issue,185240,CUDACachingAllocator.cpp:995 NVML internal assert fires instead of clean OOMError when cudaMalloc fails on Tegra (Jetson),https://github.com/pytorch/pytorch/issues/185240,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: arm, module: crash, module: cuda, module: cudacachingallocator",review early; this can change release or regression risk,2026-06-01T17:16:46Z +issue,184408,`torch.div(rounding_mode='trunc')` produces off-by-one results under `torch.compile` on CUDA,https://github.com/pytorch/pytorch/issues/184408,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-05-26T18:20:10Z +issue,175154,[inductor/aot_eager] interpolate under torch.compile produces incorrect result when mode is `nearest`,https://github.com/pytorch/pytorch/issues/175154,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: decompositions, module: dispatch, oncall: pt2",review early; this can change release or regression risk,2026-05-26T17:55:02Z +issue,181696,Eval-mode `BatchNorm`/`GroupNorm` produces wrong output under `torch.compile(backend='inductor')`,https://github.com/pytorch/pytorch/issues/181696,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: correctness (silent), module: decompositions, module: norms and normalization",review early; this can change release or regression risk,2026-05-26T17:54:20Z +issue,182649,torch.compile Inductor crashes on onnx2torch-converted Add-Sub-Reshape model with FakeTensor shape input,https://github.com/pytorch/pytorch/issues/182649,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dynamo, module: faketensor, module: pt2-dispatcher",review early; this can change release or regression risk,2026-05-26T17:54:09Z +issue,182651,torch.compile Inductor crashes on onnx2torch-converted ASPP-style Conv-Add-Reshape model with FakeTensor shape input,https://github.com/pytorch/pytorch/issues/182651,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dynamo, oncall: pt2",review early; this can change release or regression risk,2026-05-26T17:54:03Z +issue,182652,torch.compile Inductor crashes on onnx2torch-converted Conv-Reshape model with FakeTensor shape input,https://github.com/pytorch/pytorch/issues/182652,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: faketensor, module: pt2-dispatcher, module: viewing and reshaping",review early; this can change release or regression risk,2026-05-26T17:54:00Z +issue,184876,SIGSEGV when assigning to nn.Parameter.data inside a torch.func transform,https://github.com/pytorch/pytorch/issues/184876,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: functorch",review early; this can change release or regression risk,2026-05-26T17:52:30Z +issue,168126,Large numeric divergence for torch compile vs eager in bf16,https://github.com/pytorch/pytorch/issues/168126,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: decompositions, module: inductor, module: pt2-dispatcher",review early; this can change release or regression risk,2026-05-21T18:19:54Z +issue,171894,`aot_eager` backend does not support cpu pinned memory and device copy in cudagraph,https://github.com/pytorch/pytorch/issues/171894,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: cuda graphs, module: dynamo, oncall: pt2",review early; this can change release or regression risk,2026-05-21T18:18:39Z +issue,184101,`torch.compile` backward crashes for `abs()`/`angle()` on transposed complex tensors,https://github.com/pytorch/pytorch/issues/184101,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: complex, module: crash, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-05-19T18:10:09Z +issue,177118,[Inductor]T5 models performance dropped 10% on inductor mode after pattern match changes,https://github.com/pytorch/pytorch/issues/177118,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: cuda, module: inductor, module: performance, module: regression",review early; this can change release or regression risk,2026-05-19T15:23:39Z +issue,146569,torch.compile with fullgraph=True causes overwritten variable error in versions later than torch==2.5.1,https://github.com/pytorch/pytorch/issues/146569,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: inductor, module: regression, oncall: pt2",review early; this can change release or regression risk,2026-05-18T21:43:34Z +issue,184032,torch.UntypedStorage..new() segfaults via null Allocator dereference at c10/core/StorageImpl.h:84,https://github.com/pytorch/pytorch/issues/184032,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: internals, module: serialization",review early; this can change release or regression risk,2026-05-18T17:20:12Z +issue,161783,torch.compile has runtime overhead on small graphs,https://github.com/pytorch/pytorch/issues/161783,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: dynamo, module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-05-18T00:03:17Z +issue,181690,Heap corruption when compiling models with `nn.LSTM` or `nn.GRU` using `torch.compile(backend='inductor')`,https://github.com/pytorch/pytorch/issues/181690,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: inductor, module: rnn, oncall: pt2",review early; this can change release or regression risk,2026-05-17T02:50:15Z +issue,180740,[Inductor] functorch_dp_cifar10 accuracy regression in dynamic_inductor_torchbench,https://github.com/pytorch/pytorch/issues/180740,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: dynamic shapes, module: inductor, module: norms and normalization, module: regression",review early; this can change release or regression risk,2026-05-17T02:47:59Z +issue,183369,torch.compile crashes on nested_tensor_from_jagged + layer_norm with InternalTorchDynamoError,https://github.com/pytorch/pytorch/issues/183369,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dynamic shapes, module: nestedtensor, oncall: pt2",review early; this can change release or regression risk,2026-05-13T23:05:25Z +issue,182283,INTERNAL ASSERT FAILED at /pytorch/c10/core/Scalar.cpp:42,https://github.com/pytorch/pytorch/issues/182283,has linked PR,94,open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: autograd,review early; this can change release or regression risk,2026-05-11T17:25:44Z +issue,182200,`torch.compile` crashes on `stft`/`fftn`/`rfftn`/`ifftn` backward — meta kernel stride mismatch,https://github.com/pytorch/pytorch/issues/182200,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: crash, module: fft, module: meta tensors",review early; this can change release or regression risk,2026-05-04T16:13:06Z +issue,182217,`torch.compile` crashes on `F.layer_norm(NestedTensor)` — `InternalTorchDynamoError` in guard building,https://github.com/pytorch/pytorch/issues/182217,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dynamic shapes, module: guards, module: nestedtensor",review early; this can change release or regression risk,2026-05-04T16:12:21Z +issue,181510,"torch.nn.utils.weight_norm triggers a UBSan null pointer error for nn.Linear(0, 1) on CPU",https://github.com/pytorch/pytorch/issues/181510,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: crash, module: edge cases, module: empty tensor",review early; this can change release or regression risk,2026-04-28T15:48:29Z +issue,178680,"torch.compile (inductor/aot_eager) produces nondeterministic / incorrect results on model with `add_`, `clamp_`, and `squeeze`",https://github.com/pytorch/pytorch/issues/178680,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-03-31T18:28:00Z +issue,178089,bad-free in torch.sparse.spdiags due to missing bounds check on diagonal offsets,https://github.com/pytorch/pytorch/issues/178089,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: error checking, module: sparse",review early; this can change release or regression risk,2026-03-23T17:19:02Z +issue,139500,"Custom operators registered via decorator slower than ops registered via `torch.Library.{define, impl}`",https://github.com/pytorch/pytorch/issues/139500,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: custom-operators, module: pt2-dispatcher, module: vllm, oncall: pt2",review early; this can change release or regression risk,2026-03-20T14:56:46Z +issue,153044,Unexpected float32 overflow for amp training with torch.compile,https://github.com/pytorch/pytorch/issues/153044,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-03-11T17:33:14Z +issue,155044,"Execution of an `ExportProgram` of a model with `torch.autograd.grad(torch.sqrt(x), x, torch.ones_like(torch.sqrt(x)))` returns a `FakeTensor` instead of a real tensor",https://github.com/pytorch/pytorch/issues/155044,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: oncall: export, oncall: pt2",review early; this can change release or regression risk,2026-03-09T23:26:34Z +issue,124863,Deprecate unsupported types in operator registration,https://github.com/pytorch/pytorch/issues/124863,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-03-09T23:17:25Z +issue,145064,`_pdist_forward` causes segmentation fault for 3D tensor with last dimension of size 0,https://github.com/pytorch/pytorch/issues/145064,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: empty tensor, module: error checking",review early; this can change release or regression risk,2026-03-03T02:33:56Z +issue,175978,The following code snippet causes a segmentation fault when running torch.nn.lstm_cell:,https://github.com/pytorch/pytorch/issues/175978,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: error checking, module: rnn",review early; this can change release or regression risk,2026-03-02T18:10:38Z +issue,175190,[MPS] AvgPool2d/AdaptiveAvgPool2d backward crashes (SIGABRT) on channels_last inputs,https://github.com/pytorch/pytorch/issues/175190,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: crash, module: memory format, module: mps",review early; this can change release or regression risk,2026-02-23T18:23:16Z +issue,142454,Segmentation fault (core dumped) in `torch.max_pool1d`,https://github.com/pytorch/pytorch/issues/142454,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: edge cases, module: error checking, module: nn",review early; this can change release or regression risk,2026-02-18T14:31:53Z +issue,173946,Segmentation fault in aten.lstm.data results if hx shapes are invalid (when running on CPU),https://github.com/pytorch/pytorch/issues/173946,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: crash, module: error checking, module: rnn",review early; this can change release or regression risk,2026-02-10T00:57:48Z +issue,121343,Floating point exception on torch.lcm,https://github.com/pytorch/pytorch/issues/121343,has linked PR,94,open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash,review early; this can change release or regression risk,2026-01-27T08:35:51Z +issue,173318,[docs] `torch.cuda.Stream` link broken on 2.10+,https://github.com/pytorch/pytorch/issues/173318,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: docs, module: regression",review early; this can change release or regression risk,2026-01-26T22:01:55Z +issue,171838,`torch.nn.functional.pixel_shuffle` throws SIGFPE (Floating Point Exception) with large upscale_factor,https://github.com/pytorch/pytorch/issues/171838,has linked PR,94,"open PR references this issue; high-priority/blocker/regression label present; component/oncall labels: module: crash, module: edge cases, module: error checking, module: intel",review early; this can change release or regression risk,2026-01-16T15:27:23Z +issue,65761,torch.utils._pytree -> stable,https://github.com/pytorch/pytorch/issues/65761,has linked PR,94,open PR references this issue; high-priority/blocker/regression label present; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: pytree,route to design/research decision before patch planning,2025-11-07T05:31:33Z +issue,147326,torch.export.export fails when one input is a class inheriting from torch.nn.Module,https://github.com/pytorch/pytorch/issues/147326,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-07-08T02:46:50Z +issue,189121,"[inductor] Single-config reductions extended by _dynamic_scale_rblock are permanently uncacheable — re-benchmark {R0_BLOCK, R0_BLOCK/2} every process; near-tied winners flip perf and numerics",https://github.com/pytorch/pytorch/issues/189121,needs triage,89,"triage signal present; open PR references this issue; component/oncall labels: module: compile-time, module: determinism, module: inductor, module: performance",review linked PR state before touching issue,2026-07-07T05:10:40Z +issue,160399,`torch.compile` causes segmentation fault with simple `ModuleList` model after multiple compile calls,https://github.com/pytorch/pytorch/issues/160399,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2026-07-05T20:06:55Z +issue,188772,Make Fsspec filesystem checkpointing public,https://github.com/pytorch/pytorch/issues/188772,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed,review linked PR state before touching issue,2026-07-02T04:37:17Z +issue,184229,torch.export with prefer_deferred_runtime_asserts_over_guards=True crashes on DTensor-parallelized models (IndexError in set_missing_meta_vals),https://github.com/pytorch/pytorch/issues/184229,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-06-26T20:15:58Z +issue,188193,torch::cuda::nccl::gather/scatter currently implemented as O(world_size) p2p connections on root,https://github.com/pytorch/pytorch/issues/188193,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed infra",review linked PR state before touching issue,2026-06-26T16:46:04Z +issue,185441,[JIT] Null pointer write in SourceRangeDeserializer::deserialize_source() via unchecked text_table indices,https://github.com/pytorch/pytorch/issues/185441,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: jit,review linked PR state before touching issue,2026-06-23T13:12:38Z +issue,172930,torch.export failed while exporting alpamayo model (Qwen3_vl) with symbolic error,https://github.com/pytorch/pytorch/issues/172930,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-06-17T17:12:59Z +issue,187480,`TestQuantizedTensor.test_qtensor_fill_per_channel_nhwc` fails on GB200,https://github.com/pytorch/pytorch/issues/187480,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: quantization,review linked PR state before touching issue,2026-06-16T18:57:47Z +issue,187138,"[DCP] load silently casts checkpoint tensors to the template dtype, truncating precision without any warning",https://github.com/pytorch/pytorch/issues/187138,needs triage,89,"triage signal present; open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",review linked PR state before touching issue,2026-06-15T16:14:43Z +issue,187048,[Bug][Quantization] Inconsistent numerical behavior in _choose_qparams_per_tensor and redundant code in test_quantized_tensor.py,https://github.com/pytorch/pytorch/issues/187048,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: quantization,review linked PR state before touching issue,2026-06-11T14:08:02Z +issue,186938,"ShardedTensor.device hardcodes torch.cuda.current_device(), breaks FSDP checkpoint resume on XPU/non-CUDA builds",https://github.com/pytorch/pytorch/issues/186938,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed,review linked PR state before touching issue,2026-06-10T15:41:19Z +issue,178618,torch.export in PyTorch 2.10 emits INT64_MAX instead of None for aten.slice.Tensor end bounds,https://github.com/pytorch/pytorch/issues/178618,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-06-09T17:13:18Z +issue,143580,"RuntimeError: expect_autograd_hooks_ INTERNAL ASSERT FAILED at ""../torch/csrc/distributed/c10d/reducer.cpp""",https://github.com/pytorch/pytorch/issues/143580,needs triage,89,"triage signal present; open PR references this issue; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-09T16:58:39Z +issue,184841,`torch.export.export` raises `RuntimeError: Unhandled FakeTensor Device Propagation` for CUDA/MPS example inputs instead of a clear user-facing error,https://github.com/pytorch/pytorch/issues/184841,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-06-02T22:55:05Z +issue,185167,"DTensor: `F.nll_loss(weight=..., reduction='mean')` over a sharded batch returns wrong result",https://github.com/pytorch/pytorch/issues/185167,needs triage,89,"triage signal present; open PR references this issue; component/oncall labels: module: dtensor, module: nn, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-05-28T14:09:00Z +issue,163189,[export] Silent fake tensor leakage when doing buffer mutation,https://github.com/pytorch/pytorch/issues/163189,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-05-27T00:13:19Z +issue,184996,"torch.export.export accepts a parameter whose storage is too small, then save/runtime fails with low-level setStorage error",https://github.com/pytorch/pytorch/issues/184996,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-05-25T08:27:05Z +issue,181061,[RFC] M-to-N cast primitive (for RL weight rollout & pipeline parallel),https://github.com/pytorch/pytorch/issues/181061,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed infra",review linked PR state before touching issue,2026-05-23T16:27:41Z +issue,178473,ProcessGroupNCCL scalable communicator init may reuse stale UniqueNCCLID store keys,https://github.com/pytorch/pytorch/issues/178473,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed infra",review linked PR state before touching issue,2026-05-22T16:52:24Z +issue,178798,all_gather_into_tensor in gloo backend does not support stacking,https://github.com/pytorch/pytorch/issues/178798,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed infra",review linked PR state before touching issue,2026-05-22T16:51:38Z +issue,151510,"[eager] [cpu] [edge case] When processing `torch.nan_to_num-.long()`, inductor outputs the `reciprocal` of eager",https://github.com/pytorch/pytorch/issues/151510,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T04:12:11Z +issue,182915,Add NVFP4-aware fused scaled_mm_v2 + reduce_scatter support in symmetric memory,https://github.com/pytorch/pytorch/issues/182915,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed infra",review linked PR state before touching issue,2026-05-17T07:58:27Z +issue,183520,Cannot JIT a deprecated class,https://github.com/pytorch/pytorch/issues/183520,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: jit,review linked PR state before touching issue,2026-05-13T14:20:25Z +issue,183207,FSDP2 with ignore_params does not work with foreach optimizers,https://github.com/pytorch/pytorch/issues/183207,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: dtensor, module: fsdp, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-05-11T19:50:06Z +issue,181238,"torch.cuda.synchronize() syncs on device 0 when current device is not explicitly set, causing incomplete profiling results",https://github.com/pytorch/pytorch/issues/181238,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: profiler,review linked PR state before touching issue,2026-05-08T13:11:50Z +issue,180354,`torch.cond` with tensors created in `[true/false]_fn` fails during `run_decompositions`,https://github.com/pytorch/pytorch/issues/180354,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-05-04T21:32:24Z +issue,182102,DeviceMesh pickling warning spam in DCP process,https://github.com/pytorch/pytorch/issues/182102,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",review linked PR state before touching issue,2026-05-04T08:39:53Z +issue,182031,DTensor.local_map inconsistently handles pytree on input versus placements,https://github.com/pytorch/pytorch/issues/182031,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-04-30T14:46:45Z +issue,153614,Add opentelemetry traces,https://github.com/pytorch/pytorch/issues/153614,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: profiler,review linked PR state before touching issue,2026-04-28T21:19:40Z +issue,178522,AArch64 static runtimes failures,https://github.com/pytorch/pytorch/issues/178522,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: arm, oncall: quantization",review linked PR state before touching issue,2026-04-20T15:40:13Z +issue,143157,can we make not(x) --> sym_not(x) automatically during tracing.,https://github.com/pytorch/pytorch/issues/143157,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",review linked PR state before touching issue,2026-04-01T15:45:37Z +issue,177510,[export] re-raising augmented exceptions resets traceback to except block,https://github.com/pytorch/pytorch/issues/177510,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-03-16T09:01:48Z +issue,176428,torch.export.load() slow due to Python-side schema deserialization,https://github.com/pytorch/pytorch/issues/176428,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-03-09T23:28:26Z +issue,172839,_dynamo_graph_capture_for_export doesn't work on pytree output with tensor closures,https://github.com/pytorch/pytorch/issues/172839,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-03-09T23:27:54Z +issue,171978,`torch.export` fails on models using `pack_padded_sequence` with LSTM due to FakeTensor in dimension,https://github.com/pytorch/pytorch/issues/171978,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-03-09T23:27:43Z +issue,166093,aoti cross compile for windows failed with undefined reference to WinMain,https://github.com/pytorch/pytorch/issues/166093,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-02-27T08:57:01Z +issue,173643,"Typo ""emitDefautlArgsWithOutArgs"" in torch/csrc/jit/serialization/export_module.cpp",https://github.com/pytorch/pytorch/issues/173643,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: jit,review linked PR state before touching issue,2026-01-28T16:20:46Z +issue,173495,`torch.fbgemm_linear_quantize_weight` crashes with segmentation fault (SIGSEGV) with a cuda tensor,https://github.com/pytorch/pytorch/issues/173495,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: error checking, oncall: quantization",review linked PR state before touching issue,2026-01-27T15:44:41Z +issue,170127,Could not gaurd on data-dependent expression for SDPA,https://github.com/pytorch/pytorch/issues/170127,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",review linked PR state before touching issue,2026-01-22T06:27:13Z +issue,171964,`torch.export.export` on unflattened module fails for dynamic shapes,https://github.com/pytorch/pytorch/issues/171964,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-01-21T18:51:08Z +issue,172711,torch.compile produces wrong output channels for mkldnn._convolution_transpose_pointwise (FakeTensor shape inference bug),https://github.com/pytorch/pytorch/issues/172711,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: correctness (silent), oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2026-01-21T17:22:19Z +issue,171897,Torch decomposition has no effect on torch.ops.aten._fused_rms_norm,https://github.com/pytorch/pytorch/issues/171897,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-01-16T02:06:20Z +issue,171977,`torch.export` triggers internal state corruption (`AssertionError: cpu`) via `copy.copy` on FakeTensor,https://github.com/pytorch/pytorch/issues/171977,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-01-13T17:20:20Z +issue,171974,`torch.export` fails on models with KV-cache-like stateful updates using data-dependent slice assignment,https://github.com/pytorch/pytorch/issues/171974,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-01-13T17:18:59Z +issue,171870,[s390x/ZVECTOR] C++ compilation failure in argmax/argmin operations with torch.compile inductor backend,https://github.com/pytorch/pytorch/issues/171870,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; component/oncall labels: module: power, oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2026-01-13T07:14:37Z +issue,147170,[BUG][PyTorch 2.0 Export][quant]:get_source_partitions() may return different matches with same input graph,https://github.com/pytorch/pytorch/issues/147170,needs triage,89,"PyTorch issue lacks triaged label; open PR references this issue; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2, oncall: quantization",review linked PR state before touching issue,2025-11-15T21:57:13Z +issue,130861,torch.fx.Tracer.record_stack_traces is broken in torch 2.4.0,https://github.com/pytorch/pytorch/issues/130861,needs triage,89,PyTorch issue lacks triaged label; open PR references this issue; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: oncall: export,review linked PR state before touching issue,2025-09-23T12:08:48Z +issue,170039,PyTorch crashes with std::__1::system_error: mutex lock failed: Invalid argument,https://github.com/pytorch/pytorch/issues/170039,high-priority/blocker,88,"high-priority/blocker/regression label present; needs design/research label present; component/oncall labels: module: convolution, module: crash, module: macos",route to design/research decision before patch planning,2026-02-08T13:11:27Z +issue,177730,`nanosleep()` call in `socket.cpp` delay method cannot handle signal interrupts (e.g. from sampling-based profilers),https://github.com/pytorch/pytorch/issues/177730,needs triage,83,"PyTorch issue lacks triaged label; needs design/research label present; component/oncall labels: oncall: distributed, oncall: distributed infra",route to design/research decision before patch planning,2026-05-21T17:01:27Z +issue,51856,Assignment target is transposed when using jit.script and avanced indexing,https://github.com/pytorch/pytorch/issues/51856,needs triage,83,PyTorch issue lacks triaged label; label says issue is suitable for a scoped external contribution; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-02-18T19:46:05Z +issue,131941,[export] Schematize nn_module_stack serialization,https://github.com/pytorch/pytorch/issues/131941,needs triage,83,PyTorch issue lacks triaged label; label says issue is suitable for a scoped external contribution; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2026-02-02T09:14:46Z +issue,167872,exported program cannot be loaded with load_pt2: GraphModuleDeserializer' object has no attribute 'symbol_name_to_range',https://github.com/pytorch/pytorch/issues/167872,needs triage,83,"PyTorch issue lacks triaged label; open PR references this issue; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-03T22:00:26Z +issue,161053,[export] Non-strict export of `torch.cond` fails with `UnspecializedNNModuleVariable is already tracked for mutation`,https://github.com/pytorch/pytorch/issues/161053,needs triage,83,"PyTorch issue lacks triaged label; open PR references this issue; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: export, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-26T15:56:28Z +issue,177906,CPU: 2x CPU utilization regression in 2.9+ with no proportional wall-clock improvement,https://github.com/pytorch/pytorch/issues/177906,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: cpu, module: multithreading, module: openmp, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-03T09:21:32Z +issue,169823,Abnormal CPU usage with ComfyUI,https://github.com/pytorch/pytorch/issues/169823,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: performance, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-03T09:18:57Z +issue,188255,"Crash while import torch, in some special virtualization situation.",https://github.com/pytorch/pytorch/issues/188255,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: arm, module: binaries, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-30T06:56:13Z +issue,179214,"cudaErrorLaunchOutOfResources in CTC loss backward on RTX 5090 (sm_120, Blackwell) with CUDA 13.0",https://github.com/pytorch/pytorch/issues/179214,high-priority/blocker,82,high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-23T08:05:36Z +issue,182279,Segmentation fault on RTX 5090 during repeated PyTorch CUDA forward passes,https://github.com/pytorch/pytorch/issues/182279,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-20T19:01:30Z +issue,183812,`torch.scatter_reduce` CPU performance regression (~15–58×) introduced in 2.13.0.,https://github.com/pytorch/pytorch/issues/183812,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: cpu, module: performance, module: regression, module: scatter & gather ops",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-18T22:17:24Z +issue,183815,`F.conv2d` CPU performance regression (~4–16×) for bfloat16 inputs introduced in 2.13.0.,https://github.com/pytorch/pytorch/issues/183815,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: bfloat16, module: convolution, module: cpu, module: intel",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-18T17:06:20Z +issue,176426,Triton kernels with 2 or more tl.load() calls segfault at runtime on sm_120 (NVIDIA RTX PRO 6000 Blackwell). The kernels compile without errors but produce invalid code that crashes on execution.,https://github.com/pytorch/pytorch/issues/176426,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: cuda, module: inductor, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-16T14:26:06Z +issue,175614,"Segfault during device_map=""auto"" weight materialization on RTX 5090 D v2 (GB202-240, sm_120)",https://github.com/pytorch/pytorch/issues/175614,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: cuda, module: windows",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-13T00:06:32Z +issue,183251,PyTorch Segmentation Fault on M2 Mac - Bug Report,https://github.com/pytorch/pytorch/issues/183251,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-11T17:41:40Z +issue,178538,[BUG] [CRASH] Torch cause notebook crash on any exception (import torch),https://github.com/pytorch/pytorch/issues/178538,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: windows",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-08T18:55:28Z +issue,181853,[Bug] DataLoader worker segmentation fault in multi-task training with long sequences and numerical instability,https://github.com/pytorch/pytorch/issues/181853,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: dataloader",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-06T04:44:12Z +issue,181534,CTCLoss backward is inconsistent with its forward derivative with respect to the documented log_probs input.,https://github.com/pytorch/pytorch/issues/181534,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: autograd, module: correctness (silent), module: loss, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-29T22:16:15Z +issue,180458,[Bug] ValueError: I/O operation on closed file during torch.save() on Windows (PyTorch Nightly + CUDA 13.2),https://github.com/pytorch/pytorch/issues/180458,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: multiprocessing, module: regression, module: serialization, module: windows",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-20T17:29:39Z +issue,178038,"torch.dot crashes with SIGFPE (exit code 136) on RTX 5060 Ti (Blackwell, sm_120)",https://github.com/pytorch/pytorch/issues/178038,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: cublas, module: cuda, module: linear algebra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-17T12:04:36Z +issue,177561,CUDA driver error: invalid argument in fft backward when output used in subtraction,https://github.com/pytorch/pytorch/issues/177561,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: autograd, module: cuda, module: fft",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-13T17:22:55Z +issue,178841,Importing torch before QApplication on macOS prevents Qt from discovering cocoa platform plugin (inverse of #11326),https://github.com/pytorch/pytorch/issues/178841,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: binaries, module: crash, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-13T17:10:27Z +issue,178892,[Windows] torch.save triggers 0xC0000005 Access Violation on RTX 4090 Laptop (WDDM Driver Conflict),https://github.com/pytorch/pytorch/issues/178892,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: cuda, module: pickle, module: serialization",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-13T16:39:33Z +issue,177227,Floating point exception (core dumped) with AMD processor,https://github.com/pytorch/pytorch/issues/177227,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: convolution, module: cpu, module: crash, module: mkldnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-16T17:19:33Z +issue,174468,Memory leak by torch.compile,https://github.com/pytorch/pytorch/issues/174468,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: memory usage, module: regression, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-12T02:14:09Z +issue,121101,Segmentation error for torch==2.2.1 on MacOs,https://github.com/pytorch/pytorch/issues/121101,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: crash, module: intel, module: macos, module: openmp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-10T04:09:37Z +issue,171606,cos CPU ops deterministic not ensured,https://github.com/pytorch/pytorch/issues/171606,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: cpu, module: determinism",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-05T14:06:49Z +issue,113180,Higher train loss and worse evaluation metrics when using `torch.compile()`,https://github.com/pytorch/pytorch/issues/113180,high-priority/blocker,82,"high-priority/blocker/regression label present; needs reproduction label present; component/oncall labels: module: pt2 accuracy, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-24T21:28:00Z +pr,189286,Fix DeviceContext mode leaks from checkpoint recompute and set_default_device restore,https://github.com/pytorch/pytorch/pull/189286,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-09T00:20:54Z +pr,188012,Add sparse triangular_solve CPU fallback,https://github.com/pytorch/pytorch/pull/188012,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-08T20:49:23Z +pr,186969,Bump onnx to 1.22.0 (stable) and unify typing-extensions to 4.15.0,https://github.com/pytorch/pytorch/pull/186969,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/docker, ciflow/trunk",final maintainer merge/release decision,2026-07-08T18:44:13Z +pr,187171,[inductor] Fix test_comprehensive_addbmm_cpu_float16 precision mismatch (#187170),https://github.com/pytorch/pytorch/pull/187171,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-08T18:44:12Z +pr,189265,test/inductor: relax FlexAttention invalid-block-size regex,https://github.com/pytorch/pytorch/pull/189265,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-08T18:09:53Z +pr,189142,fx: Fix incorrect return_annotation for tuple types in operator schemas,https://github.com/pytorch/pytorch/pull/189142,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-08T17:00:08Z +pr,186502,Make context decorator frames identify the context manager,https://github.com/pytorch/pytorch/pull/186502,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-08T15:34:38Z +pr,187354,Match Inductor i0/i1 infinities to eager CUDA,https://github.com/pytorch/pytorch/pull/187354,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",final maintainer merge/release decision,2026-07-08T13:52:08Z +pr,188990,[FSDP] Fix to_cpu when flat_param is already on CPU,https://github.com/pytorch/pytorch/pull/188990,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T08:40:41Z +pr,188967,Update logging docs to match current API,https://github.com/pytorch/pytorch/pull/188967,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-08T07:05:30Z +pr,187969,Enable traceable DTensor RNG ops on XPU,https://github.com/pytorch/pytorch/pull/187969,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk",final maintainer merge/release decision,2026-07-08T01:30:36Z +pr,189122,Fix sparse-dense mul dropping data when broadcasting a size-1 sparse dim,https://github.com/pytorch/pytorch/pull/189122,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-08T00:36:19Z +pr,181036,Clamp spdiags diagonal length to zero for out-of-range offsets,https://github.com/pytorch/pytorch/pull/181036,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-07T22:08:00Z +pr,187481,Add overflow check to FBGEMM version of quantize_val_arm,https://github.com/pytorch/pytorch/pull/187481,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-07T20:09:25Z +pr,188698,Make Fsspec filesystem checkpointing public and exposing cache_staged_state_dict to enable async save,https://github.com/pytorch/pytorch/pull/188698,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan,final maintainer merge/release decision,2026-07-07T18:23:02Z +pr,188114,Fix CK SDPA flash attention on gfx1200/gfx1201 (RDNA4),https://github.com/pytorch/pytorch/pull/188114,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/docker, ciflow/h100, ciflow/inductor",final maintainer merge/release decision,2026-07-07T14:35:08Z +pr,186583,Propagate contextvars to Python autograd functions,https://github.com/pytorch/pytorch/pull/186583,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-07T10:42:16Z +pr,178768,Pass tensor data pointer to dlDeviceToTorchDevice from maybeCopyTensor in DLConvertor.cpp,https://github.com/pytorch/pytorch/pull/178768,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-07T08:21:28Z +pr,184745,distributions: fix Kumaraswamy mode formula,https://github.com/pytorch/pytorch/pull/184745,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-05T17:56:11Z +pr,188799,Fix callable value-opaque objects causing graph break on method access,https://github.com/pytorch/pytorch/pull/188799,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-05T09:18:17Z +pr,188378,"Fix linspace/logspace docstrings: start/end are Number, not float",https://github.com/pytorch/pytorch/pull/188378,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-07-04T07:01:07Z +pr,186492,[inductor] Fix stale xoffset for TMA in mix-order reduction loop,https://github.com/pytorch/pytorch/pull/186492,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-02T19:15:00Z +pr,188321,[c10d] Integrate ncclScatter into PT,https://github.com/pytorch/pytorch/pull/188321,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-29T22:05:01Z +pr,185913,[BugFix] Fix torch.kron crash on non-contiguous input (Issue #185650),https://github.com/pytorch/pytorch/pull/185913,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-29T09:47:27Z +pr,188287,Improve FX CodeGen process_inputs typing,https://github.com/pytorch/pytorch/pull/188287,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-29T07:46:43Z +pr,154279,Add Support for transposed convolution with Padding Mode 'same' and 'valid',https://github.com/pytorch/pytorch/pull/154279,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-27T18:40:27Z +pr,188036,Fix THPStream_richcompare reading off the end of non-Stream objects,https://github.com/pytorch/pytorch/pull/188036,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-26T08:44:36Z +pr,182937,Align XPU unit tests,https://github.com/pytorch/pytorch/pull/182937,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/xpu,final maintainer merge/release decision,2026-06-26T06:58:48Z +pr,185812,Fix arange int64 fractional truncation,https://github.com/pytorch/pytorch/pull/185812,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps",final maintainer merge/release decision,2026-06-25T05:49:49Z +pr,186489,Fix division by zero in range when step becomes zero after conversion,https://github.com/pytorch/pytorch/pull/186489,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/mps,final maintainer merge/release decision,2026-06-25T03:17:03Z +pr,187928,test: generate native function type hint checks,https://github.com/pytorch/pytorch/pull/187928,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-24T12:29:19Z +pr,183697,Fix SVE blendv mask handling after bool fill,https://github.com/pytorch/pytorch/pull/183697,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2026-06-22T16:19:01Z +pr,184881,Fix SIGSEGV from functorch key contamination via Tensor.set_data,https://github.com/pytorch/pytorch/pull/184881,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-21T15:57:14Z +pr,169249,[ARM] Disable SVE for GCC 11.5.0 to workaround compiler bug,https://github.com/pytorch/pytorch/pull/169249,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-19T14:20:18Z +pr,186346,Handle invalid CUDA JITerator cache entries,https://github.com/pytorch/pytorch/pull/186346,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-19T05:59:07Z +pr,187396,Fix integer linspace decomposition to match eager,https://github.com/pytorch/pytorch/pull/187396,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-19T05:12:07Z +pr,178781,[MPS] Remove stale @skipIfMPS from test_nn.py,https://github.com/pytorch/pytorch/pull/178781,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/mps, ciflow/trunk",final maintainer merge/release decision,2026-06-19T01:01:59Z +pr,182323,[Mergerules][Temp] Restrict docker changes to just maintainers,https://github.com/pytorch/pytorch/pull/182323,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-19T01:01:50Z +pr,177069,[xpu] Increase tolerance for tests that fail due to non-deterministic operators behavior.,https://github.com/pytorch/pytorch/pull/177069,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-18T20:52:27Z +pr,186566,Fix ldexp gradient for negative integer exponents,https://github.com/pytorch/pytorch/pull/186566,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-18T14:09:56Z +pr,186570,[distributed] Fix Windows gloo allreduce crash by moving ProcessGroupGlooCuda.cpp to base_sources,https://github.com/pytorch/pytorch/pull/186570,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-17T17:28:56Z +pr,186836,Fix Cauchy/Gumbel icdf boundary values at q=0/q=1 (#186824),https://github.com/pytorch/pytorch/pull/186836,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-17T14:56:14Z +pr,187326,Fix torch.special.zeta returning NaN for q=+inf,https://github.com/pytorch/pytorch/pull/187326,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-16T06:06:42Z +pr,179528,[Codemod][PSS3] Upgrade fbcode/mobile-vision to Python Scientific Stack 3 (#179528),https://github.com/pytorch/pytorch/pull/179528,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-15T18:54:22Z +pr,182034,Add no-tree-loop-optimize pragma for apply_grad_input on GCC/AArch64,https://github.com/pytorch/pytorch/pull/182034,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-15T08:47:20Z +pr,178100,docs: mention torch.cuda.memory.mem_get_info in CUDA memory management notes,https://github.com/pytorch/pytorch/pull/178100,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-10T08:17:44Z +pr,182546,Fix sparse compressed set_data device metadata,https://github.com/pytorch/pytorch/pull/182546,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-09T22:15:55Z +pr,185190,Make constant SymInts hashable,https://github.com/pytorch/pytorch/pull/185190,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/dynamo, ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-06-05T08:06:05Z +pr,182344,Inductor: Fix Searchsorted/Bucketize Sliced View Boundaries,https://github.com/pytorch/pytorch/pull/182344,ready for maintainer decision,80,"approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-06-04T15:34:22Z +pr,186133,[ROCm] remove redundant @skipIfRocm from test_cudnn_attention_broken_166211,https://github.com/pytorch/pytorch/pull/186133,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-03T22:56:03Z +pr,179732,Fix torch.compile istft failing when length exceeds reconstructed sig…,https://github.com/pytorch/pytorch/pull/179732,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-06-02T20:34:32Z +pr,174198,Adjust test_linalg_lstsq_input_check to be compatible with other devices,https://github.com/pytorch/pytorch/pull/174198,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers; CI lane labels: ciflow/xpu,final maintainer merge/release decision,2026-06-01T07:16:27Z +pr,180742,[autograd] lazily start worker threads in the autograd engine,https://github.com/pytorch/pytorch/pull/180742,ready for maintainer decision,80,approved with no detected failing/pending checks; PR closes or references open issue numbers,final maintainer merge/release decision,2026-05-29T09:36:24Z +issue,162466,Launch one kernel take long time.,https://github.com/pytorch/pytorch/issues/162466,needs triage,77,triage signal present; needs reproduction label present; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-13T16:37:21Z +issue,124007,gpu memory normal but one of gpus GPU-Util is 0 so cant train,https://github.com/pytorch/pytorch/issues/124007,needs triage,77,"triage signal present; needs reproduction label present; component/oncall labels: oncall: distributed, oncall: distributed parallelisms",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:04:45Z +issue,178085,DDP mode: CUDA error: an illegal memory access was encountered,https://github.com/pytorch/pytorch/issues/178085,needs triage,77,"triage signal present; needs reproduction label present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-28T14:13:15Z +issue,178766,torch.distributed._functional_collectives.reduce_scatter_tensor gets a wrong result,https://github.com/pytorch/pytorch/issues/178766,needs triage,77,"triage signal present; needs reproduction label present; component/oncall labels: oncall: distributed, oncall: distributed infra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-23T16:24:27Z +issue,150011,[Windows][CPU] 51 UT of inductor/test_torchinductor_opinfo.py failed with CPP wrapper,https://github.com/pytorch/pytorch/issues/150011,needs triage,77,"PyTorch issue lacks triaged label; needs reproduction label present; component/oncall labels: module: windows, oncall: cpu inductor, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-22T22:32:45Z +issue,173907,[ARM][QNNPACK] quantized::conv2d.new not supported for TorchScript VisionTransformer models,https://github.com/pytorch/pytorch/issues/173907,needs triage,77,"PyTorch issue lacks triaged label; needs reproduction label present; component/oncall labels: module: arm, module: nnpack, oncall: quantization",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-31T18:08:43Z +issue,172500,`RuntimeError: device >= 0 && device < num_gpus INTERNAL ASSERT FAILED` caused by early CUDA init,https://github.com/pytorch/pytorch/issues/172500,needs triage,77,"PyTorch issue lacks triaged label; needs reproduction label present; component/oncall labels: module: cuda, oncall: profiler",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-22T18:39:17Z +issue,172254,Python 3.14 test error: `AttributeError: 'MobileNetV2' object has no attribute '__annotations__'.`,https://github.com/pytorch/pytorch/issues/172254,needs triage,77,"PyTorch issue lacks triaged label; needs reproduction label present; component/oncall labels: module: python version, oncall: jit",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-21T16:38:13Z +issue,170877,AArch64 unit test failure test_transpose_mxn_16_16_bf16_fp16 on Neoverse-V1,https://github.com/pytorch/pytorch/issues/170877,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: regression",review early; this can change release or regression risk,2025-12-22T15:42:47Z +issue,168288,ICE in gcc < 13.5 for AArch64 on Neoverse-V1,https://github.com/pytorch/pytorch/issues/168288,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: crash, module: third_party",review early; this can change release or regression risk,2025-11-21T10:49:47Z +issue,167086,Unchecked uint32 overflow in CUDA scan (cumsum/cumprod/cummin/etc),https://github.com/pytorch/pytorch/issues/167086,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda",review early; this can change release or regression risk,2025-11-10T18:07:36Z +issue,123800,add `FlopCounterMode` documentation,https://github.com/pytorch/pytorch/issues/123800,has linked PR,76,open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: docs,review early; this can change release or regression risk,2025-10-17T03:14:51Z +issue,139789,"compiling ""input mutation"" followed by ""input mutation + aliasing"" does not properly recompile",https://github.com/pytorch/pytorch/issues/139789,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: guards, oncall: pt2",review early; this can change release or regression risk,2025-10-16T18:25:25Z +issue,27614,scatter_add allows index tensor that doesn't match input size in forward pass but fails on backward pass,https://github.com/pytorch/pytorch/issues/27614,has linked PR,76,open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2025-09-29T23:18:14Z +issue,150779,Decorator `skipIfXpu` disables tests when used on class,https://github.com/pytorch/pytorch/issues/150779,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: regression, module: testing, module: tests",review early; this can change release or regression risk,2025-04-15T13:13:10Z +issue,147077,[inductor] SIGSEGV when using `torch.compile` with `torch.as_strided_copy`,https://github.com/pytorch/pytorch/issues/147077,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking, oncall: pt2",review early; this can change release or regression risk,2025-03-18T13:56:07Z +issue,146790,Segmentation Fault in `torch.ops.aten.multi_margin_loss_backward` with Empty `grad_output`,https://github.com/pytorch/pytorch/issues/146790,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-02-09T20:16:38Z +issue,145220,Regression in the compilation of the torch.all operation in PyTorch version 2.6.0 compared to 2.5.1,https://github.com/pytorch/pytorch/issues/145220,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: empty tensor, module: regression, oncall: pt2",review early; this can change release or regression risk,2025-01-21T19:11:06Z +issue,143649,Floating Point Exception (core dumped) when running floordiv/remainder/fmod under torch.compile,https://github.com/pytorch/pytorch/issues/143649,has linked PR,76,"open PR references this issue; high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: cpu inductor, oncall: pt2",review early; this can change release or regression risk,2024-12-25T06:58:11Z +pr,188980,Fix inconsistent accelerator guard in checkpoint staging test,https://github.com/pytorch/pytorch/pull/188980,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-09T02:09:58Z +pr,189193,[XCCL] Allow XCCL to be set as default in `_new_process_group_helper`,https://github.com/pytorch/pytorch/pull/189193,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:22Z +pr,181559,[XPU] Enable _flash_attention_forward/_flash_attention_backward,https://github.com/pytorch/pytorch/pull/181559,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:52:50Z +pr,189317,[doc] Add accelerator test reuse guide (#181474),https://github.com/pytorch/pytorch/pull/189317,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:41:05Z +pr,188521,Fix OrderingBarrier additional_buffer_deps key mismatch in control_deps lowering,https://github.com/pytorch/pytorch/pull/188521,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:03:51Z +pr,186252,[inductor] Fix truncdiv off-by-one on CUDA with _div_rn,https://github.com/pytorch/pytorch/pull/186252,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T00:21:46Z +pr,188978,"swap isinstance(t, FakeTensor) to is_fake",https://github.com/pytorch/pytorch/pull/188978,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:41Z +pr,186358,Fix output mismtach for F.softshrink with bfloat16 and float scalar,https://github.com/pytorch/pytorch/pull/186358,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T22:48:00Z +pr,189129,inductor: Fix non-deterministic gradients in slice_scatter backward fusion,https://github.com/pytorch/pytorch/pull/189129,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:51:00Z +pr,189294,[CI] Realign scipy after nvmath install to fix H100 smoke NumPy ABI crash,https://github.com/pytorch/pytorch/pull/189294,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/h100,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:23:06Z +pr,188931,Fix matrix-vector linear fallback crashes with 1D weight and bias,https://github.com/pytorch/pytorch/pull/188931,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:16:55Z +pr,188913,[inductor] Add triton.cudagraph_max_recorded_sizes to cap dynamic shape re-recording,https://github.com/pytorch/pytorch/pull/188913,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:14:32Z +pr,188998,[inductor] Fix bf16 matmul accumulation precision in Triton templates (#188492),https://github.com/pytorch/pytorch/pull/188998,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:14:23Z +pr,188996,[inductor] Fix FP8 blockwise scaling shape validation in Triton (#188711),https://github.com/pytorch/pytorch/pull/188996,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:13:10Z +pr,189141,"[test] Add opt-in sign-sensitive special-value parity check (copysign, atan2, reciprocal)",https://github.com/pytorch/pytorch/pull/189141,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:56:01Z +pr,187989,[MPS] Fix linear backward SIGABRT with a 1-D weight,https://github.com/pytorch/pytorch/pull/187989,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/docker, ciflow/dtensor, ciflow/h100","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:52:32Z +pr,181074,[inductor][coordesc] extract get_neighbor_configs and get_all_direction_configs (#181074),https://github.com/pytorch/pytorch/pull/181074,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:49:30Z +pr,188784,Fix `torch.asarray` for buffer-protocol objects with no dtype,https://github.com/pytorch/pytorch/pull/188784,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:48:55Z +pr,189149,"[inductor] Emit pointwise kernel for small mm (K<8, N<8) instead of cuBLAS",https://github.com/pytorch/pytorch/pull/189149,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:37:31Z +pr,188965,Fix non-deterministic node ordering in get_source_partitions(),https://github.com/pytorch/pytorch/pull/188965,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:33:53Z +pr,184279,Avoid MultiOutput for single-output fallback kernels,https://github.com/pytorch/pytorch/pull/184279,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:50:16Z +pr,186349,Improve Dynamo logger method graph-break guidance,https://github.com/pytorch/pytorch/pull/186349,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:47:13Z +pr,186355,Fix AOTAutograd unused output backward pruning,https://github.com/pytorch/pytorch/pull/186355,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:43:09Z +pr,183884,Fix compile worker quiesce draining,https://github.com/pytorch/pytorch/pull/183884,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:35:57Z +pr,186364,Prevent guard validation from mutating ShapeEnv,https://github.com/pytorch/pytorch/pull/186364,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:32:00Z +pr,186372,Fix torch.Tensor construction from traced NumPy arrays,https://github.com/pytorch/pytorch/pull/186372,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:28:15Z +pr,186406,Fix invalid source in AOT repro generation,https://github.com/pytorch/pytorch/pull/186406,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:24:07Z +pr,186415,Fix Dynamo Parameter construction in new modules,https://github.com/pytorch/pytorch/pull/186415,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:21:06Z +pr,186403,Avoid CUDA init during CPU torch.compile,https://github.com/pytorch/pytorch/pull/186403,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:13:22Z +pr,186849,Fix CuTeDSL grouped_mm descriptor init shape,https://github.com/pytorch/pytorch/pull/186849,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:12:36Z +pr,186425,Skip Dynamo inside non-traceable subclass dispatch hooks,https://github.com/pytorch/pytorch/pull/186425,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:08:07Z +pr,184124,Fix Inductor multi-output replacement insertion order,https://github.com/pytorch/pytorch/pull/184124,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:48:16Z +pr,184343,Prune slow Hopper WGMMA autotune configs,https://github.com/pytorch/pytorch/pull/184343,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:45:11Z +pr,186455,Fix AOTAutograd Dynamo memory leak repro,https://github.com/pytorch/pytorch/pull/186455,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:34:47Z +pr,186457,Fix torch.ops float parsing for scalar tensor args,https://github.com/pytorch/pytorch/pull/186457,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:30:28Z +pr,186460,Support dynamic mode in Dynamo explain,https://github.com/pytorch/pytorch/pull/186460,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:21:50Z +pr,185776,Propagate fake RuntimeErrors to Dynamo handlers,https://github.com/pytorch/pytorch/pull/185776,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:10:20Z +pr,186466,Fix scalar Tensor args for custom op float schemas in Dynamo,https://github.com/pytorch/pytorch/pull/186466,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:57:44Z +pr,186467,Preserve non-persistent buffers in compiled modules,https://github.com/pytorch/pytorch/pull/186467,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:54:05Z +pr,186557,Resolve scalar overloads before AOT compilation,https://github.com/pytorch/pytorch/pull/186557,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:50:03Z +pr,189233,[MPS] Fix avg_pool2d/adaptive_avg_pool2d backward on channels_last inputs,https://github.com/pytorch/pytorch/pull/189233,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-07-08T17:32:04Z +pr,185947,Fix Dynamo logit with tensor eps,https://github.com/pytorch/pytorch/pull/185947,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:24:18Z +pr,185952,Support compiled async all_reduce wait,https://github.com/pytorch/pytorch/pull/185952,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:20:37Z +pr,185971,Fix Dynamo random.Random replay on cache hits,https://github.com/pytorch/pytorch/pull/185971,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:07:28Z +pr,185972,Handle malformed MSVC help output,https://github.com/pytorch/pytorch/pull/185972,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:03:22Z +pr,186011,[dynamo] skip empty graph fallback recursively,https://github.com/pytorch/pytorch/pull/186011,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:52:11Z +pr,186016,Reject invalid bitwise dtypes in Inductor lowering,https://github.com/pytorch/pytorch/pull/186016,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:48:18Z +pr,186021,[Inductor][CPU] Guard compiled integer div by zero,https://github.com/pytorch/pytorch/pull/186021,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:39:16Z +pr,186472,Graph break on dynamic unspecialized ModuleList indexing,https://github.com/pytorch/pytorch/pull/186472,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:31:06Z +pr,186487,Run Dynamo compilation with empty functorch stack,https://github.com/pytorch/pytorch/pull/186487,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:24:08Z +pr,186248,Simplify Min/Max using ShapeEnv value ranges,https://github.com/pytorch/pytorch/pull/186248,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:17:54Z +pr,184048,Enable Inductor OpInfo CPU coverage,https://github.com/pytorch/pytorch/pull/184048,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:17:39Z +pr,186493,Optimize AOTAutograd compiled forward wrapper,https://github.com/pytorch/pytorch/pull/186493,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T16:17:07Z +pr,186494,[Dynamo] Constant-fold CUDA Tensor.is_pinned(),https://github.com/pytorch/pytorch/pull/186494,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:12:01Z +pr,188047,fix(dlpack): raise Python RuntimeError instead of aborting for negative DLPack strides,https://github.com/pytorch/pytorch/pull/188047,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-08T16:06:27Z +pr,186498,Add dynamic shape test assertions for input dims,https://github.com/pytorch/pytorch/pull/186498,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T15:58:12Z +pr,186500,Fix Dynamo resume after f-string logging graph breaks,https://github.com/pytorch/pytorch/pull/186500,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:45:31Z +pr,184392,Fix ModularIndexing printer semantics,https://github.com/pytorch/pytorch/pull/184392,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:41:30Z +pr,186504,Allow HOP speculation through torch autograd.Function skipfiles,https://github.com/pytorch/pytorch/pull/186504,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:28:59Z +pr,186507,Invalidate Dynamo caches when module hooks change,https://github.com/pytorch/pytorch/pull/186507,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:20:55Z +pr,186512,[dynamo] Graph break on shape-changing Parameter.data assignment,https://github.com/pytorch/pytorch/pull/186512,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:09:42Z +pr,186520,Preserve SymInt bindings in dynamic-shape minifier repros,https://github.com/pytorch/pytorch/pull/186520,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:01:48Z +pr,186529,Preserve grad state in Dynamo minifier repros,https://github.com/pytorch/pytorch/pull/186529,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T14:45:26Z +pr,186530,[dynamo] Restore autocast dtype after trace-time updates,https://github.com/pytorch/pytorch/pull/186530,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T14:41:41Z +pr,186534,Account for fanout in Inductor reuse realization,https://github.com/pytorch/pytorch/pull/186534,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T14:28:35Z +pr,186538,Fix CUDA driver linking for Inductor CUTLASS builds,https://github.com/pytorch/pytorch/pull/186538,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T14:17:49Z +pr,189224,[xpu][skip] Skip test_combo_kernel_dynamic_scale_rblock on XPU,https://github.com/pytorch/pytorch/pull/189224,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:52:22Z +pr,186539,[Dynamo] Limit FX node growth from loop unrolling,https://github.com/pytorch/pytorch/pull/186539,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:38:21Z +pr,186361,Improve autograd.Function backward graph grad names,https://github.com/pytorch/pytorch/pull/186361,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:36:08Z +pr,185304,Fix while_loop stride metadata preservation,https://github.com/pytorch/pytorch/pull/185304,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:12:21Z +pr,186091,Preserve CUDA LSTM in Inductor,https://github.com/pytorch/pytorch/pull/186091,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T12:47:20Z +pr,185280,Fix Dynamo automatic dynamic int specialization,https://github.com/pytorch/pytorch/pull/185280,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T12:27:31Z +pr,188049,Fix Inductor div floor nonfinite CUDA results,https://github.com/pytorch/pytorch/pull/188049,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T12:27:26Z +pr,186546,Preserve original Dynamo minifier backend failures,https://github.com/pytorch/pytorch/pull/186546,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T12:09:12Z +pr,186549,Fix fake deepcopy for custom module mappings,https://github.com/pytorch/pytorch/pull/186549,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T12:06:30Z +pr,186553,[dynamo] Graph break on packed sequence lower ops,https://github.com/pytorch/pytorch/pull/186553,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T12:03:34Z +pr,186551,Preserve symbolic tensor metadata in after-AOT repros,https://github.com/pytorch/pytorch/pull/186551,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T12:01:01Z +pr,187057,Fix value-opaque graph output replay,https://github.com/pytorch/pytorch/pull/187057,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:53:46Z +pr,186395,Preserve overlap deps without blocking fusion,https://github.com/pytorch/pytorch/pull/186395,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:46:42Z +pr,185871,Fix symbolic stride-order sorting without explicit ShapeEnv,https://github.com/pytorch/pytorch/pull/185871,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:27:18Z +pr,185873,Preserve mutation renames after scheduler recompute,https://github.com/pytorch/pytorch/pull/185873,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:24:49Z +pr,185324,Reject mixed-device fake convolution inputs,https://github.com/pytorch/pytorch/pull/185324,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:15:02Z +pr,186476,Fix Dynamo resume tensor lifetimes across graph breaks,https://github.com/pytorch/pytorch/pull/186476,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:14:32Z +pr,185876,Prevent saving mutated grad inputs in AOT min-cut,https://github.com/pytorch/pytorch/pull/185876,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:05:12Z +pr,186246,Fix stale scalar value in addcdiv graph-break decomposition,https://github.com/pytorch/pytorch/pull/186246,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T11:00:22Z +pr,184508,Fix export with unused DTensor param attrs,https://github.com/pytorch/pytorch/pull/184508,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T10:33:24Z +pr,186564,Skip uninitialized outputs during accuracy minification,https://github.com/pytorch/pytorch/pull/186564,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T10:26:53Z +pr,185939,Fix int8 WoQ GEMM residual epilogue fusion,https://github.com/pytorch/pytorch/pull/185939,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T10:22:03Z +pr,186380,Run python decompositions under functionalization,https://github.com/pytorch/pytorch/pull/186380,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T10:01:47Z +pr,186446,Fix fake overlap handling for Dynamo try/except,https://github.com/pytorch/pytorch/pull/186446,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:56:39Z +pr,186505,Simplify mod-one symbolic shape guards,https://github.com/pytorch/pytorch/pull/186505,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T09:54:20Z +pr,186462,Allow cond branch mutation during export,https://github.com/pytorch/pytorch/pull/186462,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:50:39Z +pr,186923,Graph break on custom op fake tensor failures,https://github.com/pytorch/pytorch/pull/186923,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:45:51Z +pr,186485,Fix export dynamic dims with size-one examples,https://github.com/pytorch/pytorch/pull/186485,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:42:45Z +pr,187493,Fix Inductor nextafter for low precision CUDA tensors,https://github.com/pytorch/pytorch/pull/187493,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:39:13Z +pr,187311,Preserve forward AD tangents for Inductor dual inputs,https://github.com/pytorch/pytorch/pull/187311,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:33:04Z +pr,184323,Share foreach combo kernel bodies,https://github.com/pytorch/pytorch/pull/184323,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:32:51Z +pr,183654,Fix Inductor CPU group norm affine precision,https://github.com/pytorch/pytorch/pull/183654,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:26:14Z +pr,184595,Allow functorch grad fake propagation in Dynamo,https://github.com/pytorch/pytorch/pull/184595,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:20:54Z +pr,187486,Fix unbacked memo reuse during fake replay,https://github.com/pytorch/pytorch/pull/187486,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T09:18:17Z +pr,185469,Fix stable TMA descriptor reconstruction in Dynamo,https://github.com/pytorch/pytorch/pull/185469,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:54:06Z +pr,185804,Fix non-strict export tensor len handling,https://github.com/pytorch/pytorch/pull/185804,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:41:16Z +pr,183511,Fix Inductor addmm beta-zero input handling,https://github.com/pytorch/pytorch/pull/183511,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:41:01Z +pr,186388,Avoid native eager router graph breaks under Dynamo,https://github.com/pytorch/pytorch/pull/186388,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:41:00Z +pr,186656,Autotune XBLOCK for mix-order reductions,https://github.com/pytorch/pytorch/pull/186656,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:40:37Z +pr,184656,Fix CPU SDPA higher-order grad compile,https://github.com/pytorch/pytorch/pull/184656,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T08:38:02Z +pr,186857,Use real flex block counts for autotune inputs,https://github.com/pytorch/pytorch/pull/186857,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:33:36Z +pr,185878,Fix Inductor expm1 on float subnormal inputs,https://github.com/pytorch/pytorch/pull/185878,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:28:20Z +pr,183744,Update ACL dependencies for BF16 inner product,https://github.com/pytorch/pytorch/pull/183744,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/docker, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:23:33Z +pr,183854,Filter C++-wrapper-incompatible extern autotune choices,https://github.com/pytorch/pytorch/pull/183854,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:17:45Z +pr,185501,Fix Issue #185396 inductor complex64 triton signature,https://github.com/pytorch/pytorch/pull/185501,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T07:45:41Z +pr,186150,Fix compile storage semantics for as_strided scatter,https://github.com/pytorch/pytorch/pull/186150,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T07:32:08Z +pr,183866,Preserve native matmul lowp output precision,https://github.com/pytorch/pytorch/pull/183866,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T07:22:01Z +pr,184713,Support Dynamo class setattr for hook handles,https://github.com/pytorch/pytorch/pull/184713,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T07:19:25Z +pr,184793,Fix no-tensor-aliasing recompile reason for stale sources,https://github.com/pytorch/pytorch/pull/184793,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T07:05:29Z +pr,183862,[Inductor] Preserve view_as_complex input layout before fake prop,https://github.com/pytorch/pytorch/pull/183862,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:38:21Z +pr,185063,Handle scalar carries in Inductor while_loop,https://github.com/pytorch/pytorch/pull/185063,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:19:11Z +pr,185800,Fix TORCH_LOGS stderr handling at process exit,https://github.com/pytorch/pytorch/pull/185800,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:14:19Z +pr,185908,[Dynamo] Preserve live dict view semantics around mappingproxy mutation,https://github.com/pytorch/pytorch/pull/185908,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:11:11Z +pr,185085,[dynamo] Support datetime.now attrs as runtime inputs,https://github.com/pytorch/pytorch/pull/185085,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:07:41Z +pr,160585,Add dim=None support to torch.logsumexp,https://github.com/pytorch/pytorch/pull/160585,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T05:58:24Z +pr,183994,Specialize numpy scalar module attrs,https://github.com/pytorch/pytorch/pull/183994,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T05:50:01Z +pr,186913,Fix int64 value_expr arithmetic in Triton codegen,https://github.com/pytorch/pytorch/pull/186913,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T05:47:06Z +pr,188327,[test] Add error_inputs for nn.Conv2d module,https://github.com/pytorch/pytorch/pull/188327,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T05:39:05Z +pr,187614,Use sym_numel in linalg_cond string overload,https://github.com/pytorch/pytorch/pull/187614,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T04:57:51Z +pr,183652,Fix dynamic slice lowering for generic views,https://github.com/pytorch/pytorch/pull/183652,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T04:50:00Z +pr,184934,Validate packed RNN batch_sizes before fallback,https://github.com/pytorch/pytorch/pull/184934,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T04:29:01Z +pr,184950,Fix mkldnn transposed conv pointwise shape inference,https://github.com/pytorch/pytorch/pull/184950,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T04:19:40Z +pr,184803,[dynamo] Avoid key-set guard for dict setitem,https://github.com/pytorch/pytorch/pull/184803,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T03:53:31Z +pr,183681,Fix Triton compile-time autotune fallback,https://github.com/pytorch/pytorch/pull/183681,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T03:40:48Z +pr,186845,[make_fx] Decompose detach by default for higher-order grads,https://github.com/pytorch/pytorch/pull/186845,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T03:40:38Z +pr,184838,Fix Dynamo raw descriptor building,https://github.com/pytorch/pytorch/pull/184838,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T03:35:58Z +pr,184848,Fix nearest upsample scale shortcuts in compile,https://github.com/pytorch/pytorch/pull/184848,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T03:32:45Z +pr,184864,Error on overlapping as_strided mutations in compile,https://github.com/pytorch/pytorch/pull/184864,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T02:34:02Z +pr,185108,Prevent non-strict export from leaking fake buffers,https://github.com/pytorch/pytorch/pull/185108,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T02:29:12Z +pr,183731,[inductor] Validate fallback output dtype metadata,https://github.com/pytorch/pytorch/pull/183731,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T02:23:00Z +pr,179376,fix(export): add SDPA decomposition for contiguous output,https://github.com/pytorch/pytorch/pull/179376,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T02:18:50Z +pr,184964,Handle tensor scalar slice bounds in export setitem,https://github.com/pytorch/pytorch/pull/184964,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T02:13:43Z +pr,184967,Fix export CIA decomps with backend kernels,https://github.com/pytorch/pytorch/pull/184967,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-08T02:08:48Z +pr,185815,Fix Dynamo NumPy scalar identity handling,https://github.com/pytorch/pytorch/pull/185815,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T02:02:30Z +pr,184977,Fix Dynamo wrapper method sources for nested compile,https://github.com/pytorch/pytorch/pull/184977,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:48:44Z +pr,188641,Mark linalg.eigh as cudagraph-unsafe in Inductor,https://github.com/pytorch/pytorch/pull/188641,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:43:28Z +pr,185953,Clear Dynamo weakrefs before graph-break resumes,https://github.com/pytorch/pytorch/pull/185953,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:37:41Z +pr,185288,Support symbolic list lengths in Triton grids,https://github.com/pytorch/pytorch/pull/185288,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:35:05Z +pr,186027,Fix symbolic broadcast inference for unbacked sizes,https://github.com/pytorch/pytorch/pull/186027,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:14:58Z +pr,188636,Handle reciprocal SymPy powers in shape range eval,https://github.com/pytorch/pytorch/pull/188636,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T01:11:14Z +pr,186893,Fix dynamic size-one tensor recompiles,https://github.com/pytorch/pytorch/pull/186893,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-pallas, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:08:57Z +pr,184988,Allow HOPs to pass through unchanged input tensors,https://github.com/pytorch/pytorch/pull/184988,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:06:12Z +pr,184990,[dynamo] Route fake RuntimeError through user exception handlers,https://github.com/pytorch/pytorch/pull/184990,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:03:39Z +pr,186032,Avoid reshape metadata guards on symbolic size-one checks,https://github.com/pytorch/pytorch/pull/186032,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T00:58:50Z +pr,184628,Fix SingletonInt static guard evaluation,https://github.com/pytorch/pytorch/pull/184628,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T00:41:33Z +pr,185033,Support Dynamo Tensor.backward by default,https://github.com/pytorch/pytorch/pull/185033,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T00:29:13Z +pr,185034,Enable Dynamo capture of torch.autograd.grad,https://github.com/pytorch/pytorch/pull/185034,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T00:25:51Z +pr,185042,Avoid value guards for residual lazy constants,https://github.com/pytorch/pytorch/pull/185042,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T00:20:19Z +pr,184590,Fix uncached NestedTensor seqlens under compile,https://github.com/pytorch/pytorch/pull/184590,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T00:13:03Z +pr,184920,[dynamo] Store FunctionIdSet entries by identity weakrefs,https://github.com/pytorch/pytorch/pull/184920,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T23:53:56Z +pr,185775,Avoid storage access for no-storage tensors in MetaConverter,https://github.com/pytorch/pytorch/pull/185775,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T23:50:14Z +pr,186886,Fix keyed user Triton autotune in Inductor,https://github.com/pytorch/pytorch/pull/186886,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T23:43:59Z +pr,185100,Keep fake tensor cache synthesis out of proxy traces,https://github.com/pytorch/pytorch/pull/185100,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T23:35:34Z +pr,185103,Fix unbacked scalar equality reasoning through div/mod,https://github.com/pytorch/pytorch/pull/185103,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T23:31:06Z +pr,185116,Thread effect tokens through cond branches,https://github.com/pytorch/pytorch/pull/185116,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T23:21:55Z +pr,184007,[inductor] Clear stale subprocess cache env vars,https://github.com/pytorch/pytorch/pull/184007,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-07T23:19:03Z +pr,182224,Enable skipped static_runtime tests on AArch64,https://github.com/pytorch/pytorch/pull/182224,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-07T22:49:48Z +pr,189191,Cuda Graph Default Enable (#189191),https://github.com/pytorch/pytorch/pull/189191,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T22:49:03Z +pr,185138,[dynamo] Enforce one compiled frame for fullgraph skipped frames,https://github.com/pytorch/pytorch/pull/185138,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T22:37:36Z +pr,185149,Reject partial-overlap copy aliases,https://github.com/pytorch/pytorch/pull/185149,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T22:28:51Z +pr,185159,Inductor: gate max-autotune 1x1 conv GEMM lowering,https://github.com/pytorch/pytorch/pull/185159,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-07T22:21:35Z +pr,184047,Enable cooperative reductions by default,https://github.com/pytorch/pytorch/pull/184047,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-perf-compare, ciflow/inductor-periodic, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T21:46:44Z +pr,175566,Fix test_extension_backend race condition,https://github.com/pytorch/pytorch/pull/175566,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T20:11:47Z +pr,184276,[inductor] fix torch.randperm for slice_shape node in fx_passes,https://github.com/pytorch/pytorch/pull/184276,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-07-07T19:39:05Z +pr,185084,Avoid contiguity guards on maybe-singleton symbolic dims,https://github.com/pytorch/pytorch/pull/185084,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-07T19:23:43Z +pr,185946,[CUDA] Warn instead of assert in `reportProcessMemoryInfo` to prevent crash on Tegra OOM,https://github.com/pytorch/pytorch/pull/185946,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-07T19:21:47Z +pr,185300,Fix AttrProxy module reuse in Dynamo,https://github.com/pytorch/pytorch/pull/185300,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T19:17:11Z +pr,185320,Improve scan HOP direct call error,https://github.com/pytorch/pytorch/pull/185320,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T19:11:00Z +pr,184059,Fix split reduction kernel metadata,https://github.com/pytorch/pytorch/pull/184059,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T19:02:38Z +pr,185336,[dynamo] Add regression test for prepended forward pre-hooks,https://github.com/pytorch/pytorch/pull/185336,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T18:55:57Z +pr,185349,Fix flex decoding BLOCK_M validation for short GQA,https://github.com/pytorch/pytorch/pull/185349,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T18:47:13Z +pr,185356,Fix DTensor embedding mask state under compile,https://github.com/pytorch/pytorch/pull/185356,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T18:38:46Z +pr,186003,Fix Dynamo condition-dependent frame skip caching,https://github.com/pytorch/pytorch/pull/186003,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T18:37:28Z +pr,185357,Forward container methods on OptimizedModule,https://github.com/pytorch/pytorch/pull/185357,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T18:33:56Z +pr,186622,Fix bf16 autocast precision barriers in Inductor,https://github.com/pytorch/pytorch/pull/186622,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T18:09:44Z +pr,185365,Fix FX subgraph rewrite of adjacent aggregate matches,https://github.com/pytorch/pytorch/pull/185365,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-07T18:06:57Z +pr,188828,Skip conv_transpose3d redispatch test on Windows CI - flaky test,https://github.com/pytorch/pytorch/pull/188828,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-07T18:03:47Z +pr,185371,Fix symbolic antialiased resize export shapes,https://github.com/pytorch/pytorch/pull/185371,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:59:04Z +pr,185373,Fix assume_constant_result tensor constants after graph breaks,https://github.com/pytorch/pytorch/pull/185373,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:55:12Z +pr,184103,Fix FakeTensorUpdater HOP dtype propagation,https://github.com/pytorch/pytorch/pull/184103,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:48:51Z +pr,185419,[dynamo] avoid stale verbose hint for backend failures,https://github.com/pytorch/pytorch/pull/185419,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:26:18Z +pr,185436,Preserve eager exception assertions for eager_noexcept,https://github.com/pytorch/pytorch/pull/185436,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:22:40Z +pr,185438,Scope distribution validation disabling to Dynamo tracing,https://github.com/pytorch/pytorch/pull/185438,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:20:01Z +pr,185451,Wrap internal Dynamo assertion errors,https://github.com/pytorch/pytorch/pull/185451,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:12:06Z +pr,185057,Support torch.cuda.use_mem_pool in compile,https://github.com/pytorch/pytorch/pull/185057,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:08:52Z +pr,185461,[dynamo] Handle module-level logging ignores,https://github.com/pytorch/pytorch/pull/185461,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:06:15Z +pr,185462,Fix NumPy ndarray operators under default device contexts,https://github.com/pytorch/pytorch/pull/185462,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:01:40Z +pr,185466,[dynamo] Show carets in graph break user stacks,https://github.com/pytorch/pytorch/pull/185466,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T16:57:14Z +pr,185467,Keep tensors alive for Dynamo pointer reconstruction,https://github.com/pytorch/pytorch/pull/185467,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T16:53:45Z +pr,185471,Handle Dynamo tracing of torch._C._infer_size,https://github.com/pytorch/pytorch/pull/185471,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T16:50:04Z +pr,185474,Fix nested tensor tangent metadata coercion,https://github.com/pytorch/pytorch/pull/185474,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T16:41:41Z +pr,184086,Fix Inductor scaled_mm scale_result lowering,https://github.com/pytorch/pytorch/pull/184086,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-07T16:30:06Z +pr,184090,Preserve benchmark integer inputs for Inductor debug,https://github.com/pytorch/pytorch/pull/184090,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T16:12:27Z +pr,185504,Support unbacked SymInt closures in cond export,https://github.com/pytorch/pytorch/pull/185504,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T16:09:25Z +pr,185508,[dynamo] Preserve observed exception stack for reraises,https://github.com/pytorch/pytorch/pull/185508,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T16:02:55Z +pr,185528,Fix Dynamo tracing of compile support checks,https://github.com/pytorch/pytorch/pull/185528,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T15:50:42Z +pr,185536,Teach Dynamo PGO about sparse tensor strides,https://github.com/pytorch/pytorch/pull/185536,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T15:46:33Z +pr,185545,Handle symbolic bounds in slice decomposition,https://github.com/pytorch/pytorch/pull/185545,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T15:43:08Z +pr,185568,Fix export of arange with data-dependent float step,https://github.com/pytorch/pytorch/pull/185568,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T15:30:11Z +pr,185588,[inductor] Codegen backed SymInt input runtime asserts,https://github.com/pytorch/pytorch/pull/185588,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T15:19:08Z +pr,185591,Canonicalize avg_pool2d singleton lists during export,https://github.com/pytorch/pytorch/pull/185591,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T15:14:17Z +pr,185604,Improve infer_size unbacked broadcast error,https://github.com/pytorch/pytorch/pull/185604,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T15:11:46Z +pr,185624,Fix subclass static input remapping for SymInt args,https://github.com/pytorch/pytorch/pull/185624,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T15:02:30Z +pr,185632,Make AOTAutograd metadata repr subclass-safe,https://github.com/pytorch/pytorch/pull/185632,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T14:59:34Z +pr,184109,Fix Inductor replacement patterns across view-to-reshape,https://github.com/pytorch/pytorch/pull/184109,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T14:52:56Z +pr,185722,[dynamo] Wrap module.compile leaf module call frames,https://github.com/pytorch/pytorch/pull/185722,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T14:10:05Z +pr,184119,Avoid fp8 bf16 prologue fusion on pre-sm90,https://github.com/pytorch/pytorch/pull/184119,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-07T14:03:15Z +pr,185724,Fix scalar searchsorted decomposition for unbacked values,https://github.com/pytorch/pytorch/pull/185724,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:58:50Z +pr,185726,Improve Dynamo stack for nonstrict_trace input errors,https://github.com/pytorch/pytorch/pull/185726,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:55:55Z +pr,185731,Clear stale prefix exception table entries in Dynamo,https://github.com/pytorch/pytorch/pull/185731,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:50:06Z +pr,185732,Fix nested compile during subclass fake tracing,https://github.com/pytorch/pytorch/pull/185732,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:44:45Z +pr,185734,Preserve torch._check messages in runtime asserts,https://github.com/pytorch/pytorch/pull/185734,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T13:40:16Z +pr,185740,Support inspect.signature on nested functions in Dynamo,https://github.com/pytorch/pytorch/pull/185740,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:32:17Z +pr,185742,Support None leaves in nonstrict_trace,https://github.com/pytorch/pytorch/pull/185742,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:28:04Z +pr,185730,Fix dynamic shapes for variadic kwargs,https://github.com/pytorch/pytorch/pull/185730,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-07T13:27:38Z +pr,188783,[xpu][fix]Fix bmm_outer_product override to support XPU tensors,https://github.com/pytorch/pytorch/pull/188783,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:27:33Z +pr,185902,Enable torch.compile for RNN modules by default,https://github.com/pytorch/pytorch/pull/185902,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:19:06Z +pr,185752,Make OptimizedModule state_dict use public keys,https://github.com/pytorch/pytorch/pull/185752,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:14:15Z +pr,185762,[Dynamo] Support _AutoDispatchBelowAutograd context manager,https://github.com/pytorch/pytorch/pull/185762,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:07:26Z +pr,185763,Improve disable API graph break messages,https://github.com/pytorch/pytorch/pull/185763,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T13:03:37Z +pr,185766,Dynamo: support torch.jit.isinstance without graph breaks,https://github.com/pytorch/pytorch/pull/185766,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:00:43Z +pr,185768,Improve Dynamo error for dynamic module hook registration,https://github.com/pytorch/pytorch/pull/185768,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T12:56:11Z +pr,185769,Preserve generator return values in Dynamo,https://github.com/pytorch/pytorch/pull/185769,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T12:46:39Z +pr,185771,Support Dynamo tracing through textwrap.indent string formatting,https://github.com/pytorch/pytorch/pull/185771,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T12:44:02Z +pr,186469,Wrap Dynamo-generated bytecode runtime errors,https://github.com/pytorch/pytorch/pull/186469,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T12:40:35Z +pr,185773,Use max dynamic bounds for Inductor autotuning,https://github.com/pytorch/pytorch/pull/185773,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T12:33:23Z +pr,185456,Fix Dynamo tracing for Tensor.random_,https://github.com/pytorch/pytorch/pull/185456,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T12:26:47Z +pr,185781,Avoid unbound symbols in insignificant output strides,https://github.com/pytorch/pytorch/pull/185781,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T12:20:40Z +pr,185784,Fix flex attention captured vector gradient scatters,https://github.com/pytorch/pytorch/pull/185784,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T12:07:52Z +pr,185795,Preserve overlapping input layouts when cloning Dynamo inputs,https://github.com/pytorch/pytorch/pull/185795,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T11:28:13Z +pr,185799,Avoid dataclass constructors in Dynamo and make_fx graphs,https://github.com/pytorch/pytorch/pull/185799,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T11:24:34Z +pr,185809,Fix as_strided_copy storage bounds in Inductor,https://github.com/pytorch/pytorch/pull/185809,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T11:09:45Z +pr,185916,Fix FlexAttention nested index backward,https://github.com/pytorch/pytorch/pull/185916,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T11:01:20Z +pr,185823,[Dynamo] Handle class method_descriptor calls generically,https://github.com/pytorch/pytorch/pull/185823,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T10:58:48Z +pr,185920,[Dynamo] Detect tensor mutations when reconstructing generators,https://github.com/pytorch/pytorch/pull/185920,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T10:53:51Z +pr,185829,Fix autograd.grad graph breaks under aot_eager,https://github.com/pytorch/pytorch/pull/185829,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T10:24:02Z +pr,184820,[dynamo] respect custom metaclass __call__ during class construction,https://github.com/pytorch/pytorch/pull/184820,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T09:07:38Z +pr,186043,Lower symbolic bool not to torch.sym_not,https://github.com/pytorch/pytorch/pull/186043,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T09:04:49Z +pr,186051,Fix disabled no-op module pre-hooks in compile,https://github.com/pytorch/pytorch/pull/186051,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T09:01:09Z +pr,186052,Fix Inductor C++ pybinding source literal escaping,https://github.com/pytorch/pytorch/pull/186052,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:58:18Z +pr,186554,Preserve hooks for dependent AOTAutograd outputs,https://github.com/pytorch/pytorch/pull/186554,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:54:46Z +pr,187589,Make FX symbolic tracing state thread-local,https://github.com/pytorch/pytorch/pull/187589,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T08:44:58Z +pr,184185,Use memory stats for persistent reductions,https://github.com/pytorch/pytorch/pull/184185,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:42:28Z +pr,186053,Fix AOTAutograd cache save for call-free lazy backwards,https://github.com/pytorch/pytorch/pull/186053,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:37:22Z +pr,184189,Make generated Triton code use runtime device properties,https://github.com/pytorch/pytorch/pull/184189,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:34:30Z +pr,186058,Preserve export graph metadata during normalization,https://github.com/pytorch/pytorch/pull/186058,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:30:24Z +pr,186063,[dynamo] Route OptimizedModule delegated methods through compiled call,https://github.com/pytorch/pytorch/pull/186063,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:27:25Z +pr,186069,Fix Dynamo round on NumPy scalar results,https://github.com/pytorch/pytorch/pull/186069,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:24:40Z +pr,189009,[xpu][fix] Enable test_fallback_kernel_with_symexpr_output on XPU with correct head_dim,https://github.com/pytorch/pytorch/pull/189009,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T08:21:52Z +pr,186614,Delay RNN graph break until module call,https://github.com/pytorch/pytorch/pull/186614,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:24:16Z +pr,186072,Make constant warnings reorderable in Dynamo by default,https://github.com/pytorch/pytorch/pull/186072,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T07:20:53Z +pr,186898,Add always-on AOTI input rank checks,https://github.com/pytorch/pytorch/pull/186898,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:18:27Z +pr,185866,Fix CPU Inductor numerics for BatchNorm erf normalize,https://github.com/pytorch/pytorch/pull/185866,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:15:18Z +pr,186078,Fix OptimizedModule _parameters access,https://github.com/pytorch/pytorch/pull/186078,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:07:26Z +pr,184460,Remove obsolete Python wrapper symbol extraction,https://github.com/pytorch/pytorch/pull/184460,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:05:51Z +pr,188556,[inductor] Preserve eager semantics for infinite Bessel inputs,https://github.com/pytorch/pytorch/pull/188556,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:04:47Z +pr,187353,Preserve signed zero in Triton neg lowering,https://github.com/pytorch/pytorch/pull/187353,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:03:49Z +pr,185333,Graph break existing list tensor assignments in Dynamo,https://github.com/pytorch/pytorch/pull/185333,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:01:15Z +pr,186094,Fix AOTI cond predicate unbacked symbol dependency,https://github.com/pytorch/pytorch/pull/186094,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T06:55:13Z +pr,186900,Improve pytree missing registration error,https://github.com/pytorch/pytorch/pull/186900,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T06:43:21Z +pr,186942,[inductor] Round fmod/remainder scalars for bf16/fp16 CPU,https://github.com/pytorch/pytorch/pull/186942,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T06:40:25Z +pr,186119,Improve Dynamo observed exception messages,https://github.com/pytorch/pytorch/pull/186119,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T06:36:59Z +pr,186166,Respect fake/functional modes for backend autograd kernels,https://github.com/pytorch/pytorch/pull/186166,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-07T06:29:52Z +pr,186044,Stop Dynamo from importing optional libraries at import time,https://github.com/pytorch/pytorch/pull/186044,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T06:24:21Z +pr,186168,Fix Dynamo tracing of torch.random.fork_rng,https://github.com/pytorch/pytorch/pull/186168,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T06:19:28Z +pr,186180,Fix Dynamo cProfile under active Python profilers,https://github.com/pytorch/pytorch/pull/186180,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T06:16:58Z +pr,186920,Route overrideable SDPA AOTI calls through dispatcher,https://github.com/pytorch/pytorch/pull/186920,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T06:06:41Z +pr,184367,Fix AotAutograd aliased-output replay for traceable wrapper subclasses,https://github.com/pytorch/pytorch/pull/184367,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T06:00:06Z +pr,186192,Fix CompiledFxGraph callable state across threads,https://github.com/pytorch/pytorch/pull/186192,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:58:43Z +pr,184256,[inductor] Cap reduction XBLOCK to max block,https://github.com/pytorch/pytorch/pull/184256,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:50:52Z +pr,184705,[dynamo] Preserve exception chaining in InternalTorchDynamoError rewrapping,https://github.com/pytorch/pytorch/pull/184705,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:47:17Z +pr,186501,Collapse foreach input mutation copies in AOTAutograd,https://github.com/pytorch/pytorch/pull/186501,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:40:00Z +pr,186210,Fix cond branch-local shape assertions,https://github.com/pytorch/pytorch/pull/186210,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:36:21Z +pr,186214,Profile lazy backward compilation with cProfile,https://github.com/pytorch/pytorch/pull/186214,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:32:50Z +pr,186217,Move FX graph cache hash details out of debug logs,https://github.com/pytorch/pytorch/pull/186217,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:28:59Z +pr,186222,Include epilogue guards in Dynamo code_parts,https://github.com/pytorch/pytorch/pull/186222,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:23:25Z +pr,186249,[dynamo] Support time.time as a runtime scalar input,https://github.com/pytorch/pytorch/pull/186249,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T05:05:55Z +pr,187890,Fix Dynamo metadata sync after inplace tensor mutations,https://github.com/pytorch/pytorch/pull/187890,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T04:57:36Z +pr,186263,Increase small fake tensor constant limit,https://github.com/pytorch/pytorch/pull/186263,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T04:55:06Z +pr,186280,Disable profiler Python tracing during compile,https://github.com/pytorch/pytorch/pull/186280,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T04:50:59Z +pr,186287,Reject non-functional export decomposition entries,https://github.com/pytorch/pytorch/pull/186287,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T04:46:25Z +pr,184294,Avoid unused storage size guards for indexing,https://github.com/pytorch/pytorch/pull/184294,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T04:40:13Z +pr,186340,Ensure explicit compile attempts skipped code,https://github.com/pytorch/pytorch/pull/186340,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T04:35:47Z +pr,188546,[Inductor] Fix log_sigmoid_forward flushing float32 subnormals to +0.0 on CUDA,https://github.com/pytorch/pytorch/pull/188546,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T03:07:28Z +pr,188466,Fix inductor scatter crash with empty index,https://github.com/pytorch/pytorch/pull/188466,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T00:07:42Z +pr,184133,[aot_autograd] Reduce runtime_wrapper Python overhead: 4 micro-optimizations,https://github.com/pytorch/pytorch/pull/184133,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-06T22:02:10Z +pr,175081,Add shim config includes for linting,https://github.com/pytorch/pytorch/pull/175081,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/binaries, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T21:59:20Z +pr,132135,"[torch.special] Adding betainc, betaincc, betaincinv, betainccinv, betaln and beta with backward operation",https://github.com/pytorch/pytorch/pull/132135,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T21:23:21Z +pr,188286,Fix out= variant recompilations for CompositeImplicitAutograd linalg ops,https://github.com/pytorch/pytorch/pull/188286,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:53:33Z +pr,188902,[ATen] Add reduction tag to missing base reduction ops,https://github.com/pytorch/pytorch/pull/188902,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:52:22Z +pr,188406,[dynamo] preserve Stream/Event subclass type across graph-break rec…,https://github.com/pytorch/pytorch/pull/188406,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:13:36Z +pr,188853,Register DTensorSpec as an opaque reference type for torch.export,https://github.com/pytorch/pytorch/pull/188853,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:11:31Z +pr,188859,Raise a clear error when linalg.eigh/svd workspace exceeds 32-bit LAPACK limit,https://github.com/pytorch/pytorch/pull/188859,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-06T18:51:14Z +pr,186294,[dynamo] Proxy OptimizedModule __class__ to wrapped module,https://github.com/pytorch/pytorch/pull/186294,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T18:38:06Z +pr,186306,Support tensor torch._assert in export,https://github.com/pytorch/pytorch/pull/186306,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T18:35:10Z +pr,186874,Fix Dynamo resume lifetime for _EnableTorchFunction,https://github.com/pytorch/pytorch/pull/186874,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T18:26:10Z +pr,186319,Avoid full simplify in vector-range index analysis,https://github.com/pytorch/pytorch/pull/186319,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T18:15:25Z +pr,186330,Fix compiled autograd.Function manual vmap rules,https://github.com/pytorch/pytorch/pull/186330,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T18:03:39Z +pr,184305,Fix zero-bias SDPA flash selection,https://github.com/pytorch/pytorch/pull/184305,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T17:13:30Z +pr,186336,Support multiword CXX commands in Inductor cpp_builder,https://github.com/pytorch/pytorch/pull/186336,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T17:06:55Z +pr,187596,Fall back on AOTAutograd version-check capture errors,https://github.com/pytorch/pytorch/pull/187596,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-06T17:03:15Z +pr,185791,Fix cond branch shape guards leaking to export constraints,https://github.com/pytorch/pytorch/pull/185791,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T16:59:35Z +pr,188268,Fix positional arg overflow in TritonBenchmarkRequest when template inputs are deduplicated,https://github.com/pytorch/pytorch/pull/188268,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-06T10:01:56Z +pr,188328,[meta] Add output_padding and bias validation to transposed convolution meta path,https://github.com/pytorch/pytorch/pull/188328,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-06T05:32:45Z +pr,179619,[PowerPC] Guard deviceTypeFromActivity with USE_KINETO check,https://github.com/pytorch/pytorch/pull/179619,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-05T22:54:03Z +pr,187782,[dynamo] Skip guard creation for unused function inputs,https://github.com/pytorch/pytorch/pull/187782,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-05T14:29:15Z +pr,188401,dtensor: include local tensor device in _stable_hash_for_caching,https://github.com/pytorch/pytorch/pull/188401,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-05T14:03:48Z +pr,187664,[MPS] Fix `baddbmm` operation to handle zero-sized dimensions correctly,https://github.com/pytorch/pytorch/pull/187664,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-07-05T12:45:29Z +pr,186479,[dynamo] Respect mark_dynamic on Parameter inputs,https://github.com/pytorch/pytorch/pull/186479,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-05T09:40:32Z +pr,168999,[Pt. 1] Make some aspects of Dynamo/Inductor global state thread safe,https://github.com/pytorch/pytorch/pull/168999,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-05T05:49:00Z +pr,178664,Implement igamma/igammac backward w.r.t. first argument,https://github.com/pytorch/pytorch/pull/178664,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-04T22:49:12Z +pr,188394,Speed up CPU quantile/nanquantile with partial selection,https://github.com/pytorch/pytorch/pull/188394,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-04T19:03:01Z +pr,182216,Add Half/BFloat16 copysign overloads to fix NaN sign bit loss on CUDA,https://github.com/pytorch/pytorch/pull/182216,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-04T18:50:13Z +pr,187653,Add NanDetectMode for forward-pass NaN/Inf detection,https://github.com/pytorch/pytorch/pull/187653,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-04T15:13:42Z +pr,175082,[pytree] make `torch.utils.pytree` as public API,https://github.com/pytorch/pytorch/pull/175082,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-04T12:38:27Z +pr,188291,"Fix dims docstring for torch.flip and torch.rot90 (axes, element type)",https://github.com/pytorch/pytorch/pull/188291,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-04T07:01:06Z +pr,175947,[dynamo] Remove the dict over guarding on dict setitem,https://github.com/pytorch/pytorch/pull/175947,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-03T00:10:44Z +pr,186127,[AI Codemod][PerfAICT-General] fbcode/caffe2/torch/csrc/jit/serialization/pickler.cpp (#186127),https://github.com/pytorch/pytorch/pull/186127,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T19:51:19Z +pr,181582,[AI Codemod][RlAutotuneFbsourceWearablesMemory] torch::jit::mobile::Function::append_instruction (function.cpp) (#181582),https://github.com/pytorch/pytorch/pull/181582,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T19:15:32Z +pr,188426,Fix incorrect CPU bf16 conv3d when innermost output dim is 1 and strided,https://github.com/pytorch/pytorch/pull/188426,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-02T17:33:35Z +pr,185911,[dynamo] Keep relational guards out of dict-tag skips,https://github.com/pytorch/pytorch/pull/185911,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/tsan","identify whether block is CI, merge conflict, or review gate",2026-07-02T09:24:15Z +pr,186532,Fix index_add decomposition bounds checks,https://github.com/pytorch/pytorch/pull/186532,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T08:37:48Z +pr,187032,[dynamo] Support member access on protobuf enums,https://github.com/pytorch/pytorch/pull/187032,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T21:09:05Z +pr,182092,Allow torch.meshgrid() with empty input to return empty tuple,https://github.com/pytorch/pytorch/pull/182092,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-01T21:03:34Z +pr,188337,[inductor] Handle non-comparable arguments in index_propagation Min/Max,https://github.com/pytorch/pytorch/pull/188337,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T12:50:55Z +pr,185661,[fx] Fix pytree codegen for tuple/dict positional inputs,https://github.com/pytorch/pytorch/pull/185661,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T11:59:59Z +pr,181571,Generalize test_pooling.py to be device-agnostic and enable it on XPU,https://github.com/pytorch/pytorch/pull/181571,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-01T09:44:24Z +pr,185738,Make HOP alias and mutation support opt-in,https://github.com/pytorch/pytorch/pull/185738,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T08:19:03Z +pr,185988,Fix FlexAttention scalar captured grad backward,https://github.com/pytorch/pytorch/pull/185988,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-01T02:59:52Z +pr,187786,Support int step tensors in foreach optimizer implementations,https://github.com/pytorch/pytorch/pull/187786,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-01T01:54:05Z +pr,187207,Use abs() in lp_pool to avoid NaN for non-integer norm_type,https://github.com/pytorch/pytorch/pull/187207,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-01T01:54:04Z +pr,188368,[Inductor] Fix ValueRangeError in Inductor for zero-iteration loops,https://github.com/pytorch/pytorch/pull/188368,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T01:22:00Z +pr,185891,[dynamo] Guard AOTAutograd input mutation overlap changes,https://github.com/pytorch/pytorch/pull/185891,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-30T18:31:24Z +pr,188511,[AI Codemod][PerfAICT-General] Replace cast with castRaw in TensorType::isSubtypeOfExt to avoid shared_ptr refcount overhead (#188511),https://github.com/pytorch/pytorch/pull/188511,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-30T18:30:30Z +pr,188357,[ROCm] Add MIOpen convolution ops to FORBIDDEN_CUDAGRAPH_OPS for cudagraph safety,https://github.com/pytorch/pytorch/pull/188357,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-30T17:57:21Z +pr,176044,[opaque object] Disallow creation of reference-type opaque objects under custom ops in torch.compile,https://github.com/pytorch/pytorch/pull/176044,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-30T17:19:21Z +pr,188549,[AI Codemod][PerfAICT-General] Vectorize index_select bounds check via fused unsigned compare (#188549),https://github.com/pytorch/pytorch/pull/188549,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-30T16:30:32Z +pr,176084,ci: Cleanup dead centos code,https://github.com/pytorch/pytorch/pull/176084,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/binaries_wheel,"identify whether block is CI, merge conflict, or review gate",2026-06-30T05:55:32Z +pr,187714,test: Command to reproduce error is incorrect,https://github.com/pytorch/pytorch/pull/187714,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-30T04:53:50Z +pr,187927,docs: document FlopCounterMode semantics,https://github.com/pytorch/pytorch/pull/187927,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-29T22:34:38Z +pr,188320,[c10d] Integrate ncclGather into PT,https://github.com/pytorch/pytorch/pull/188320,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-29T22:04:19Z +pr,188410,Use typing_extensions.TypeAliasType for better reexport of __module__,https://github.com/pytorch/pytorch/pull/188410,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-29T21:21:22Z +pr,188314,Fix tutorial link for GPU Direct Storage,https://github.com/pytorch/pytorch/pull/188314,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-29T18:14:58Z +pr,169241,[XPU] Migrate 10 dynamo test cases for XPU,https://github.com/pytorch/pytorch/pull/169241,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-06-29T11:55:14Z +pr,169798,"[XPU] Migrate 2 test cases autograd, creation_ops for XPU",https://github.com/pytorch/pytorch/pull/169798,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-06-29T11:55:00Z +pr,188380,docs: fix broken gpu_direct_storage tutorial URL in serialization notes,https://github.com/pytorch/pytorch/pull/188380,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-29T11:54:58Z +pr,187871,Fix scaled_dot_product_attention: explicitly check for both attn_mask…,https://github.com/pytorch/pytorch/pull/187871,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-28T23:36:42Z +pr,181561,[AI Codemod][DevmateCppModernizeFindToContains] fbcode/caffe2/torch/csrc/jit/backends/xnnpack/xnnpack_graph_builder.cpp (#181561),https://github.com/pytorch/pytorch/pull/181561,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-28T20:49:58Z +pr,181516,Fix heap-buffer-overflow in torch.linalg.eig on subnormal real inputs,https://github.com/pytorch/pytorch/pull/181516,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-27T17:50:56Z +pr,181475,Add `tensor` and `from_numpy` to `torch.__all__`,https://github.com/pytorch/pytorch/pull/181475,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-27T15:53:33Z +pr,184986,Handle unbatched input for non-zeros padding modes in Conv1d/2d/3d,https://github.com/pytorch/pytorch/pull/184986,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-27T08:37:46Z +pr,185177,Fix DTensor nll_loss weighted mean bug (closes #185167),https://github.com/pytorch/pytorch/pull/185177,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-26T20:50:55Z +pr,180561,[optim] Support ndim >= 2 in Muon and add param_groups_for_muon helper,https://github.com/pytorch/pytorch/pull/180561,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-26T09:09:03Z +pr,181546,Fix AssertionError: failed to set ranges (#181546),https://github.com/pytorch/pytorch/pull/181546,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-26T07:48:29Z +pr,185077,Avoid folding mutated buffer storages,https://github.com/pytorch/pytorch/pull/185077,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-26T01:27:26Z +pr,179592,Fix MPS SDPA correctness when score matrix exceeds 2^32 elements,https://github.com/pytorch/pytorch/pull/179592,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-25T22:02:58Z +pr,183509,[distributed] Add fused scaled_mm_v2 + reduce_scatter for NVFP4 support,https://github.com/pytorch/pytorch/pull/183509,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T17:01:21Z +pr,184193,Preserve FX graph cache guard provenance,https://github.com/pytorch/pytorch/pull/184193,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-25T11:35:14Z +pr,186779,Error on unsupported batch norm third derivatives,https://github.com/pytorch/pytorch/pull/186779,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-25T07:44:26Z +pr,187368,[inductor] add input validation to normal and bernoulli decompositions,https://github.com/pytorch/pytorch/pull/187368,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-pallas, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T06:03:51Z +pr,184704,Capturable param in optimizers restored using custom vts and cleanup hooks,https://github.com/pytorch/pytorch/pull/184704,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T04:59:44Z +pr,181481,[Codemod][MigrateInternmcUrls] fbcode/caffe2 [A] (#181481),https://github.com/pytorch/pytorch/pull/181481,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-25T01:24:33Z +pr,175526,while_loop vmap batching rule,https://github.com/pytorch/pytorch/pull/175526,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-25T01:18:10Z +pr,175246,Fix NLLLoss backward failure for non-contiguous 4D inputs,https://github.com/pytorch/pytorch/pull/175246,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-24T21:58:09Z +pr,186085,Fix transient dispatch-mode skips in Dynamo,https://github.com/pytorch/pytorch/pull/186085,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-24T19:58:40Z +pr,186178,Fix matmul folding when an inner folded dim has size 1,https://github.com/pytorch/pytorch/pull/186178,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-24T19:21:25Z +pr,182342,[Inductor] Deal with choices that fail to compile in async pipelined autotuning path (#182342),https://github.com/pytorch/pytorch/pull/182342,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-24T16:00:27Z +pr,183668,[ROCm] fix flaky test_honor_sm_carveout,https://github.com/pytorch/pytorch/pull/183668,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-24T06:14:11Z +pr,178584,Remove unused expected failure decorator from FakeTensorOperatorInvariants test and clean up condition in Module class,https://github.com/pytorch/pytorch/pull/178584,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-23T19:24:57Z +pr,187726,Expose public API for torch.cuda.clear_cublas_workspaces,https://github.com/pytorch/pytorch/pull/187726,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T15:22:42Z +pr,181252,[Codemod][MigrateInternmcUrls] fbcode/caffe2 (#181252),https://github.com/pytorch/pytorch/pull/181252,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-22T23:54:49Z +pr,166876,[ci][cpu] Update AArch64 manylinux compiler to GCC14,https://github.com/pytorch/pytorch/pull/166876,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/docker, ciflow/linux-aarch64, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-22T16:03:37Z +pr,180557,[docs] Improve documentation of normalization layers,https://github.com/pytorch/pytorch/pull/180557,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-22T14:33:15Z +pr,186049,Fix bool argmax meta validation for Inductor,https://github.com/pytorch/pytorch/pull/186049,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-21T16:27:42Z +pr,187250,Support NumPy 2 __array__ copy kwarg and __array_wrap__ context on Tensor,https://github.com/pytorch/pytorch/pull/187250,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-21T15:57:16Z +pr,176070,[rnn] Prevent LSTMCell segfault on input/hidden shape mismatch,https://github.com/pytorch/pytorch/pull/176070,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-21T03:48:22Z +pr,183525,[MPS] truncate fractional start/end to integer type in linspace when …,https://github.com/pytorch/pytorch/pull/183525,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-20T14:23:28Z +pr,180783,"docs: fix index_add_ `index` description (points into self, not source)",https://github.com/pytorch/pytorch/pull/180783,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-19T13:44:46Z +pr,180782,docs: remove unsupported out= from torch.linalg._powsum docstring,https://github.com/pytorch/pytorch/pull/180782,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-19T13:44:45Z +pr,180319,[ROCm] Handle profiler trace format differences in combo kernel tests,https://github.com/pytorch/pytorch/pull/180319,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-rocm-mi200, ciflow/inductor-rocm-mi300, ciflow/inductor-rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-06-19T13:44:45Z +pr,180315,Add _Float16 support for Tensor::item() C++ API,https://github.com/pytorch/pytorch/pull/180315,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-19T08:36:53Z +pr,187661,Return empty result for FFT with empty batch on MKL backend,https://github.com/pytorch/pytorch/pull/187661,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-18T19:36:04Z +pr,184696,Normalize export slice default bounds from getitem,https://github.com/pytorch/pytorch/pull/184696,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-18T13:50:10Z +pr,185316,[dynamo] Resolve autograd.Function static attrs in getattr,https://github.com/pytorch/pytorch/pull/185316,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-18T01:24:49Z +pr,187405,[dtensor] Propagate `_scaled_mm` tensor meta without eager scale checks,https://github.com/pytorch/pytorch/pull/187405,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-17T23:32:38Z +pr,180570,Fix torch.cond crash with FunctionalTensor constants (#180354),https://github.com/pytorch/pytorch/pull/180570,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-17T21:23:49Z +pr,180810,[inductor][static launcher] validate pointer arg dtypes before launch,https://github.com/pytorch/pytorch/pull/180810,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-17T21:22:49Z +pr,180737,[inductor] Support pin_memory for torch.rand and torch.tensor,https://github.com/pytorch/pytorch/pull/180737,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-17T15:47:07Z +pr,185578,Fix addmm max-autotune SliceView bias guard,https://github.com/pytorch/pytorch/pull/185578,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-17T14:54:20Z +pr,179235,[Inductor] Add _debug_assert_fused option to foreach_map,https://github.com/pytorch/pytorch/pull/179235,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-17T08:52:35Z +pr,184968,Fix autocast state propagation for normalization layers in functorch batched autograd,https://github.com/pytorch/pytorch/pull/184968,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-16T21:25:19Z +pr,182431,[Inductor] Elide redundant .to(tl.float32) in Triton codegen when value already has target dtype,https://github.com/pytorch/pytorch/pull/182431,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-16T19:05:49Z +pr,178209,Fix CUDA batch_norm eval returning populated save_mean/save_invstd,https://github.com/pytorch/pytorch/pull/178209,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-16T18:53:29Z +pr,178298,torch.linalg.solve_triangular: support for eligible non-dense args,https://github.com/pytorch/pytorch/pull/178298,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-16T12:23:01Z +pr,169937,Fix vectorized float32 to uint8 conversion on AArch64,https://github.com/pytorch/pytorch/pull/169937,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-15T12:50:01Z +pr,170538,[XPU] Migrate test_testing for xpu,https://github.com/pytorch/pytorch/pull/170538,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-06-15T02:42:17Z +pr,179627,Fix all masked Transformer eager/compile inconsistency,https://github.com/pytorch/pytorch/pull/179627,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-14T16:59:38Z +pr,179584,fix inference_mode getitem compile error,https://github.com/pytorch/pytorch/pull/179584,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-14T16:59:32Z +pr,179404,[DTensor] Replace __module__ hacks with @exposed_in decorators,https://github.com/pytorch/pytorch/pull/179404,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-14T16:59:31Z +pr,184143,Bound Inductor op count for high-order diff,https://github.com/pytorch/pytorch/pull/184143,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-13T14:18:46Z +pr,179658,Add OpaqueGenerator to allow make_fx trace Generator,https://github.com/pytorch/pytorch/pull/179658,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-12T23:00:10Z +pr,187169,Support weights_only and safetensors format in torch.export.save and …,https://github.com/pytorch/pytorch/pull/187169,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-12T17:55:13Z +pr,185122,Validate parameter/buffer storage bounds at export time,https://github.com/pytorch/pytorch/pull/185122,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-12T17:39:33Z +pr,186573,[Inductor] Stabilize checkpointed reduction autotuning,https://github.com/pytorch/pytorch/pull/186573,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-12T17:00:15Z +pr,186619,[pytorch] Fix Inductor missing wait on AsyncCollectiveTensor nested inside DTensor,https://github.com/pytorch/pytorch/pull/186619,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-11T18:11:55Z +pr,179526,inductor: handle matmuls gracefully in foreach_map,https://github.com/pytorch/pytorch/pull/179526,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:40:41Z +pr,176295,[inductor] Fix cudagraph output->input feedback crashes,https://github.com/pytorch/pytorch/pull/176295,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T15:45:18Z +pr,171617,[FSDP] Preserve container object identity in _recursive_to function,https://github.com/pytorch/pytorch/pull/171617,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T13:06:41Z +pr,186061,Fixing torch.pow crash with integral dtypes on CUDA by gating reciprocal/sqrt/rsqrt shortcuts,https://github.com/pytorch/pytorch/pull/186061,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-11T11:41:37Z +pr,180162,Add CPU tensor validation to FBGEMM operators to prevent segfault,https://github.com/pytorch/pytorch/pull/180162,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-11T06:06:57Z +pr,180971,"[inductor] Using is_gpu() instead of hardcoded ""cuda"" check",https://github.com/pytorch/pytorch/pull/180971,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T01:24:30Z +pr,176670,[CUDAGraph] Enable `graph_capture_record_stream_reuse` with conditional nodes,https://github.com/pytorch/pytorch/pull/176670,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-10T23:50:28Z +pr,178805,[openreg] Add comprehensive test documentation for OpenReg backend test suite,https://github.com/pytorch/pytorch/pull/178805,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-10T07:55:21Z +pr,184334,[CUDA] Enable bf16 grouped_mm CUTLASS fast path on SM12x and Windows,https://github.com/pytorch/pytorch/pull/184334,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-10T06:44:23Z +pr,184459,Decompose supported aten._trilinear in Inductor,https://github.com/pytorch/pytorch/pull/184459,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-10T00:11:38Z +pr,184155,Fix Inductor bool vector mask casts,https://github.com/pytorch/pytorch/pull/184155,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-09T23:14:06Z +pr,184114,Add cpp wrapper fallback alignment assertions,https://github.com/pytorch/pytorch/pull/184114,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-09T23:13:58Z +pr,179882,[Devmate Tasks][Butterfly] Fix flaky fbcode//caffe2/test/distributed/tensor:dtensor_ops - test_single_dim_strategy_i0_cpu_float (#179882),https://github.com/pytorch/pytorch/pull/179882,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-09T05:53:40Z +pr,176234,[autograd] Explicitly error when layer_norm computes 3rd order derivatives,https://github.com/pytorch/pytorch/pull/176234,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-08T18:20:19Z +pr,184409,Fix local_map deferred inlining to functionalize mutable operations,https://github.com/pytorch/pytorch/pull/184409,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-05T12:26:59Z +pr,185415,[ROCm] Fix LayerNorm backward kernel for AMD Strix Halo GPUs,https://github.com/pytorch/pytorch/pull/185415,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-05T12:18:38Z +pr,177546,[inductor] Fix pad_mm leaking padded strides to user-visible outputs,https://github.com/pytorch/pytorch/pull/177546,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-04T18:58:21Z +pr,174241,[MPS] Fix conv backward to preserve input memory format,https://github.com/pytorch/pytorch/pull/174241,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-04T01:30:31Z +pr,185973,[Accelerator] Enable Default Stream API,https://github.com/pytorch/pytorch/pull/185973,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-03T22:01:12Z +pr,185851,[autograd] Fall back to tensor backward for symbolic pow exponent,https://github.com/pytorch/pytorch/pull/185851,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-03T18:37:37Z +pr,185834,[inductor] Preserve signed zero in Triton neg codegen,https://github.com/pytorch/pytorch/pull/185834,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-03T03:44:46Z +pr,185846,[decomp] Match eager's device-dependent threshold comparison dtype,https://github.com/pytorch/pytorch/pull/185846,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-03T00:04:20Z +pr,185842,[decomp] Fix multilabel_margin_loss for padded targets,https://github.com/pytorch/pytorch/pull/185842,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-02T21:59:02Z +pr,185838,[inductor] Propagate subgraph tensor constants to the root graph,https://github.com/pytorch/pytorch/pull/185838,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-02T17:39:02Z +pr,181915,Support returning None from torch.cond() true_fn and false_fn,https://github.com/pytorch/pytorch/pull/181915,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-02T03:46:13Z +pr,185727,"Document bernoulli(input, p) overload in torch docs",https://github.com/pytorch/pytorch/pull/185727,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-01T23:46:24Z +pr,179205,[functorch] Fix vmap/functionalize under inference_mode,https://github.com/pytorch/pytorch/pull/179205,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-01T22:47:09Z +pr,185518,[inductor] Add probability checks for Bernoulli lowering,https://github.com/pytorch/pytorch/pull/185518,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-01T07:24:06Z +pr,185358,[PrivateUse1] Add dedicated LSTM dispatch branch for PU1 backends.,https://github.com/pytorch/pytorch/pull/185358,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-05-29T21:41:39Z +pr,185579,Improve error message in linalg.eigh for large matrices,https://github.com/pytorch/pytorch/pull/185579,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-05-29T18:28:37Z +pr,185231,[C++] Ensure wrapped tensors can be printed,https://github.com/pytorch/pytorch/pull/185231,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-05-29T14:55:50Z +pr,184164,Fix while_loop compile with empty carry and mutation-only state,https://github.com/pytorch/pytorch/pull/184164,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-29T08:25:29Z +pr,182068,Make custom op aliasing check warn (not error) in CI,https://github.com/pytorch/pytorch/pull/182068,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-29T02:03:32Z +pr,177170,[FSDP] Fix alias annotation for fsdp::all_gather_copy_in,https://github.com/pytorch/pytorch/pull/177170,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-29T01:57:56Z +pr,182414,fix(decomp): cast padding to int in replication/reflection pad decomposition,https://github.com/pytorch/pytorch/pull/182414,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-28T17:01:06Z +pr,178106,Support Optional[list[torch.Tensor]] in infer_schema,https://github.com/pytorch/pytorch/pull/178106,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-05-27T12:58:49Z +pr,185227,Error out when getting symmetric memory device pointers for NCCL < 2.28,https://github.com/pytorch/pytorch/pull/185227,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/h100-symm-mem,"identify whether block is CI, merge conflict, or review gate",2026-05-27T08:38:20Z +pr,184420,"Fix pow(-0.0, +/-0.5) sign on sqrt/rsqrt shortcut",https://github.com/pytorch/pytorch/pull/184420,PR blocked,73,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-26T23:49:52Z +pr,184822,Add Sphinx redirects for renamed autosummary Stream/stream pages,https://github.com/pytorch/pytorch/pull/184822,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-05-26T16:10:55Z +pr,184631,[MPS] Skip strided NDArray path in _mps_linear_nograph when inputs are contiguous,https://github.com/pytorch/pytorch/pull/184631,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-05-26T13:45:10Z +pr,181823,Docs: update torch.multinomial num_samples requirement note for bette…,https://github.com/pytorch/pytorch/pull/181823,PR blocked,73,failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-05-25T11:25:22Z +pr,183857,Support float8 dtypes in index_copy on CPU and CUDA,https://github.com/pytorch/pytorch/pull/183857,ready for maintainer decision,72,approved with no detected failing/pending checks; PR closes or references open issue numbers; PR has not updated for at least 45 days,final maintainer merge/release decision,2026-05-18T16:41:02Z +pr,171101,workaround for transpose_mxn test failure on AArch64,https://github.com/pytorch/pytorch/pull/171101,ready for maintainer decision,72,"approved with no detected failing/pending checks; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2026-05-11T20:08:35Z +pr,141912,Add stdlib includes to c10d debug handlers,https://github.com/pytorch/pytorch/pull/141912,ready for maintainer decision,72,approved with no detected failing/pending checks; PR closes or references open issue numbers; PR has not updated for at least 45 days,final maintainer merge/release decision,2025-03-10T19:21:46Z +pr,121445,[NJT] Actually inline NT torch function during dynamo,https://github.com/pytorch/pytorch/pull/121445,ready for maintainer decision,72,"approved with no detected failing/pending checks; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2024-08-06T22:35:39Z +issue,170672,"torch.export.export Fails with RuntimeError and ""Please file an issue on github"" for In-Place Parameter Updates in forward",https://github.com/pytorch/pytorch/issues/170672,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2026-01-04T04:13:24Z +issue,170539,"test_export_opinfo: 9 tests fail with ""Unexpected success"" due to outdated xfail markers",https://github.com/pytorch/pytorch/issues/170539,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-12-17T00:26:40Z +issue,163713,[export] 8*s72 is not tracked with proxy for torch.fx.experimental.proxy_tensor._ModuleStackTracer object,https://github.com/pytorch/pytorch/issues/163713,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-12-15T10:45:37Z +issue,163323,[export] Bug Report: module.to(device) on exported programs causes assertion failures (device-agnostic export broken since 2.8),https://github.com/pytorch/pytorch/issues/163323,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-12-12T01:09:12Z +issue,169991,aot_export_joint_with_descriptors don't work with compile inside due to dynamo config,https://github.com/pytorch/pytorch/issues/169991,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-12-09T23:28:48Z +issue,169210,Segmentation Fault when Converting QInt32Storage to bfloat16,https://github.com/pytorch/pytorch/issues/169210,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, oncall: quantization",review linked PR state before touching issue,2025-12-05T00:42:42Z +issue,164313,Torch export with dynamic shapes does not work with Hugging Face models that has TransformersKwargs in forward signature,https://github.com/pytorch/pytorch/issues/164313,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-11-11T22:17:58Z +issue,167007,"torch.export failed with tensor subclass, worked in 2.9.0",https://github.com/pytorch/pytorch/issues/167007,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-11-11T07:44:20Z +issue,166121,[profile] profile_memory doesn't work if _ExperimentalConfig(profile_all_threads = True) is set,https://github.com/pytorch/pytorch/issues/166121,needs triage,71,PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,review linked PR state before touching issue,2025-11-10T22:01:15Z +issue,146719,Operations that fail under `torch.export.export`(`torch.autograd.grad`) -> `torch.compile`,https://github.com/pytorch/pytorch/issues/146719,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-11-07T16:22:01Z +issue,149497,symbolic_trace failed on deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B,https://github.com/pytorch/pytorch/issues/149497,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: fx",review linked PR state before touching issue,2025-10-31T03:37:04Z +issue,165641,strict export doesn't work with global constants,https://github.com/pytorch/pytorch/issues/165641,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-21T21:37:23Z +issue,165444,AOTInductor not updating buffers inplace,https://github.com/pytorch/pytorch/issues/165444,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-19T23:42:43Z +issue,154559,[cond] support for unbacked symbols in compiled region,https://github.com/pytorch/pytorch/issues/154559,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-18T05:59:01Z +issue,165032,"torch.export fails when you slice an input at a dynamic dimension and Dim(“t”, min=, max=)",https://github.com/pytorch/pytorch/issues/165032,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-13T23:34:32Z +issue,165073,torch.export lift symint input to a graph during dynamo,https://github.com/pytorch/pytorch/issues/165073,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-10T17:00:38Z +issue,153247,Cannot use torch.arange in torch.ops.higher_order.scan,https://github.com/pytorch/pytorch/issues/153247,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-02T05:53:16Z +issue,159468,Dynamo ONNX Export of resize operation doesn't work as expected,https://github.com/pytorch/pytorch/issues/159468,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-01T07:16:38Z +issue,159918,[export] Tensor subclass decomposition breaks state dict contracts,https://github.com/pytorch/pytorch/issues/159918,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-24T15:20:27Z +issue,163303,[export] non-strict export creates extra tensor constants when forward runs with a container of params,https://github.com/pytorch/pytorch/issues/163303,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-23T16:52:11Z +issue,163312,[symbolic trace] fuction with decorator,https://github.com/pytorch/pytorch/issues/163312,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: fx",review linked PR state before touching issue,2025-09-20T00:51:03Z +issue,162734,Run decomposition fails for prims.broadcast_in_dim,https://github.com/pytorch/pytorch/issues/162734,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-16T00:08:19Z +issue,162386,Cannot trace out joint from non-strict export for model with activation checkpointing (AC),https://github.com/pytorch/pytorch/issues/162386,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-16T00:07:20Z +issue,140607,Bad error message for malformed args during export,https://github.com/pytorch/pytorch/issues/140607,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:16:08Z +issue,161807,torch.fx.experimental.optimization.optimize_for_inference changes OrderedDict outputs to immutable_dict in torch.export CallSpec,https://github.com/pytorch/pytorch/issues/161807,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-08T18:50:49Z +issue,161671,Saving a modified ExportedProgram,https://github.com/pytorch/pytorch/issues/161671,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-08-29T18:44:31Z +issue,159346,[AOTI] Severe Performance Regression with FP16 Autocast in AOTInductor for Small Batch Sizes,https://github.com/pytorch/pytorch/issues/159346,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-08-21T23:25:42Z +issue,156786,[inductor] [cpu] `torch.nn.Embedding-torch.index_copy` outputs inconsistent results on cpu inductor,https://github.com/pytorch/pytorch/issues/156786,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2025-08-12T02:56:15Z +issue,158540,can't torch.export.export tinyllama model,https://github.com/pytorch/pytorch/issues/158540,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-07-26T00:27:15Z +issue,156511,functorch_maml_omniglot is a bad CPU performance smoketest model,https://github.com/pytorch/pytorch/issues/156511,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: infra, oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2025-07-07T22:25:25Z +issue,147380,[export] fail to export joint graph of a model with tied weights using experimental `_export_forward_backward` API,https://github.com/pytorch/pytorch/issues/147380,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-06-06T22:59:38Z +issue,154647,"Recreation of unbacked symint leads to ""possible memo disaster"" when running decompositions",https://github.com/pytorch/pytorch/issues/154647,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-06-05T00:27:42Z +issue,154137,[Dynamo][Inductor] `detectron2_fcos_r_50_fpn` in export config failure on dashboard,https://github.com/pytorch/pytorch/issues/154137,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: guards, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-05-27T18:21:07Z +issue,153387,Pre-dispatch export doesn't work with non-param/buffer tensor subclasses,https://github.com/pytorch/pytorch/issues/153387,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-05-20T15:45:46Z +issue,153056,Export doesn't move embedding to correct device,https://github.com/pytorch/pytorch/issues/153056,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-05-07T19:09:01Z +issue,150371,Dynamic shapes doesn't work with kwargs,https://github.com/pytorch/pytorch/issues/150371,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-04-02T20:09:53Z +issue,145687,[Inductor-CPU] `add` is not being fused with templated int8 WoQ GEMM while running LLaMA2,https://github.com/pytorch/pytorch/issues/145687,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2025-02-04T22:33:10Z +issue,145785,export serde turn hop's tuple arg into list,https://github.com/pytorch/pytorch/issues/145785,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-01-31T23:44:47Z +issue,145574,[compile / strict export] torch._dynamo.exc.Unsupported: CollectiveFunctionRewriteVariable can't support async_op=True for ,https://github.com/pytorch/pytorch/issues/145574,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-01-24T17:53:07Z +issue,144247,[inductor][cpu] torch.bitwise_and/or/xor incorrectly accepts float32 tensors,https://github.com/pytorch/pytorch/issues/144247,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2025-01-06T18:04:38Z +issue,140683,[inductor][cpu]maml fp32/amp_fp16 performance regression in 2024-11-11 nightly release,https://github.com/pytorch/pytorch/issues/140683,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2024-12-17T01:19:15Z +issue,141640,Relaxing Export IR node.meta requirements,https://github.com/pytorch/pytorch/issues/141640,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2024-12-04T18:17:29Z +issue,141115,[AOTInductor] AOT compiled models do not consistently check input shapes,https://github.com/pytorch/pytorch/issues/141115,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2024-11-20T19:06:14Z +issue,140615,Non-actionable error message when output type is not pytree-able for aoti artifact,https://github.com/pytorch/pytorch/issues/140615,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2024-11-18T01:45:45Z +issue,140842,torch.cond with unbacked symint failed to compile in AOTI,https://github.com/pytorch/pytorch/issues/140842,needs triage,71,"PyTorch issue lacks triaged label; open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",review linked PR state before touching issue,2024-11-16T00:02:36Z +pr,187631,[dynamo] Fix segfault in list comprehension graph break,https://github.com/pytorch/pytorch/pull/187631,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-09T01:04:06Z +pr,188175,[dynamo] fix autograd cache and split_module tests under canonicalization,https://github.com/pytorch/pytorch/pull/188175,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-08T23:22:38Z +pr,189292,Fix dead ExecuTorch XNNPACK doc link in mobile_optimizer,https://github.com/pytorch/pytorch/pull/189292,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-08T23:22:32Z +pr,186809,[cudaMallocAsync] Count graph-mem pool in memory_reserved(),https://github.com/pytorch/pytorch/pull/186809,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-08T22:47:34Z +pr,189256,[BE][MPS] Factor host-buffer wrapping into buffer_with_offset_from_tensor,https://github.com/pytorch/pytorch/pull/189256,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/mps,final maintainer merge/release decision,2026-07-08T21:46:11Z +pr,188901,Implement Double Backward for `cdist` and `pdist`,https://github.com/pytorch/pytorch/pull/188901,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-08T21:37:19Z +pr,189187,Retune against triton 3.8,https://github.com/pytorch/pytorch/pull/189187,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T21:22:11Z +pr,189021,[dynamo] Support empty tuple-subclass construction (MyTuple()),https://github.com/pytorch/pytorch/pull/189021,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T21:21:53Z +pr,187887,[Inductor] Add decomposition for prims.uniform,https://github.com/pytorch/pytorch/pull/187887,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",final maintainer merge/release decision,2026-07-08T21:12:15Z +pr,188825,"[Dynamo] Implement subgenerator support in "".throw"" / "".close"" methods",https://github.com/pytorch/pytorch/pull/188825,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T21:06:07Z +pr,188834,[Dynamo] Track generator attribute mutations for proper closure handling,https://github.com/pytorch/pytorch/pull/188834,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T21:06:06Z +pr,188639,[Dynamo] Corretly implement SEND opcode,https://github.com/pytorch/pytorch/pull/188639,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",final maintainer merge/release decision,2026-07-08T21:06:04Z +pr,188638,[Dynamo] Fix CLEANUP_THROW,https://github.com/pytorch/pytorch/pull/188638,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T21:06:03Z +pr,185309,Add sparse COO dispatch for linalg.vector_norm,https://github.com/pytorch/pytorch/pull/185309,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-08T20:52:04Z +pr,182906,Port 11 tests under test/distributed folder to XPU,https://github.com/pytorch/pytorch/pull/182906,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/torchtitan,final maintainer merge/release decision,2026-07-08T20:48:29Z +pr,189201,[TEST][Inductor] Fix some `test_buffer_mutation*` tests,https://github.com/pytorch/pytorch/pull/189201,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-08T20:40:29Z +pr,189172,Fix protobuf resource-compiler build with clang-cl (RC1106 /bigobj),https://github.com/pytorch/pytorch/pull/189172,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-08T20:20:05Z +pr,186854,Simplify endianness check using std::endian,https://github.com/pytorch/pytorch/pull/186854,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/trunk, ciflow/vllm",final maintainer merge/release decision,2026-07-08T19:23:37Z +pr,184261,[Test] Make test_serialization.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/184261,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-08T19:12:10Z +issue,189287,"[vllm] vllm-x-pytorch-test multimodal pooling tests fail with EngineCore ""CUDA driver initialization failed"" since Triton 3.8 bump (#188251)",https://github.com/pytorch/pytorch/issues/189287,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: ci, module: cuda, module: regression, module: tests",review early; this can change release or regression risk,2026-07-08T18:15:41Z +issue,189283,"[CPU] operator_benchmark: isolated ~2x regressions (index_add, baddbmm) since ~Nov 2025",https://github.com/pytorch/pytorch/issues/189283,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: performance, module: regression",review early; this can change release or regression risk,2026-07-08T17:34:26Z +pr,187509,test: add backend-agnostic c10d collective tests,https://github.com/pytorch/pytorch/pull/187509,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-08T17:00:29Z +pr,188472,[inductor] Bucket all_reduce in the manual overlap scheduler,https://github.com/pytorch/pytorch/pull/188472,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T16:51:13Z +pr,187236,Make Overlap Runtime Estimation Device Aware,https://github.com/pytorch/pytorch/pull/187236,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-08T15:25:16Z +pr,188330,Dispatch `index_put_` to `masked_fill_.Scalar` when possible,https://github.com/pytorch/pytorch/pull/188330,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-08T12:14:25Z +pr,188480,[Dynamo] Add UNSPECIFIED_BACKEND lint rule requiring explicit backend=,https://github.com/pytorch/pytorch/pull/188480,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2026-07-08T05:57:36Z +issue,185405,"[Inductor] AssertionError: len(index) == len(stride) in ir.py when using torch.compile(mode=""max-autotune"") on CPU BMM/GEMM",https://github.com/pytorch/pytorch/issues/185405,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: complex, module: cpu, module: crash, module: inductor",review early; this can change release or regression risk,2026-07-08T05:55:34Z +pr,189181,[CI] Advisor: split garbage verdict into infra_issue + garbage (minimal),https://github.com/pytorch/pytorch/pull/189181,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-08T03:15:15Z +pr,189017,Respect the blocking parameter in torch.Event,https://github.com/pytorch/pytorch/pull/189017,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/mps, ciflow/trunk",final maintainer merge/release decision,2026-07-08T01:25:26Z +pr,189004,Add priority&priority_range to torch.Stream,https://github.com/pytorch/pytorch/pull/189004,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-08T01:09:05Z +pr,189033,Auto-triage open source PRs when a maintainer engages,https://github.com/pytorch/pytorch/pull/189033,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-07T22:10:04Z +pr,188396,[Test] Improve test reuse in test/test_tensor_iterator.py for out-of-tree backends,https://github.com/pytorch/pytorch/pull/188396,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-07T18:13:06Z +pr,185207,[Test] Make dynamo/test_aot_autograd_cache.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/185207,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-07T16:34:21Z +pr,176404,Add prioritized text linker optimization for lld.,https://github.com/pytorch/pytorch/pull/176404,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/linux-aarch64, ciflow/trunk",final maintainer merge/release decision,2026-07-07T14:18:47Z +pr,184593,[Test] Make test_autograd.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/184593,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-07T13:56:39Z +pr,185586,Update README of OpenReg module to consistent with its structure and principle,https://github.com/pytorch/pytorch/pull/185586,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-07T09:43:23Z +issue,188332,Inductor: allgather fusion creates rank-specific cache that crashes when loaded by other ranks,https://github.com/pytorch/pytorch/issues/188332,high-priority/blocker,70,high-priority/blocker/regression label present; component/oncall labels: oncall: pt2,review early; this can change release or regression risk,2026-07-07T06:55:08Z +issue,185839,"torch.linalg.qr (float64) repeated in a loop crashes the process (SIGSEGV, exit 139) on ROCm 7.x",https://github.com/pytorch/pytorch/issues/185839,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cudacachingallocator, module: linear algebra, module: regression",review early; this can change release or regression risk,2026-07-07T03:32:56Z +pr,188666,Enable xpu for TestRefs and TestDecomp.,https://github.com/pytorch/pytorch/pull/188666,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/trunk, ciflow/xpu",final maintainer merge/release decision,2026-07-07T02:30:51Z +pr,182784,"[xpu][test] Port distributed _tools,algorithms,launcher tests cases on Intel GPUs",https://github.com/pytorch/pytorch/pull/182784,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/h100-distributed,final maintainer merge/release decision,2026-07-07T01:50:18Z +pr,189016,[Test] Improve test reuse in test_torch.py for out-of-tree backends,https://github.com/pytorch/pytorch/pull/189016,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-07T01:49:04Z +issue,189018,`_scaled_mm_v2` meta implementation checks for device-specific constraints,https://github.com/pytorch/pytorch/issues/189018,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: dispatch, module: meta tensors, module: regression",review early; this can change release or regression risk,2026-07-06T22:46:50Z +pr,186639,[CUDA graphs] Add resolve_and_remap convenience for kernel annotations,https://github.com/pytorch/pytorch/pull/186639,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-06T20:11:11Z +pr,188832,[pytorch][cuda] make_graphed_callables: expose capture_error_mode (#188832),https://github.com/pytorch/pytorch/pull/188832,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-06T20:11:00Z +pr,182062,[CUDA] Early-exit in `release_cached_blocks` if `captures_underway` isn't empty,https://github.com/pytorch/pytorch/pull/182062,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-06T19:08:20Z +issue,188733,[vllm] [2.12.1 regression][Spec Decode] CUDA WARP_ILLEGAL_ADDRESS in FlashAttnFwdSm90 during CUDA graph capture (EAGLE),https://github.com/pytorch/pytorch/issues/188733,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: cuda graphs, module: regression",review early; this can change release or regression risk,2026-07-06T18:05:06Z +issue,188835,DistillGPT2 fails accuracy on A100 in inductor smoke test,https://github.com/pytorch/pytorch/issues/188835,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: inductor, module: numerical-stability, module: regression, oncall: pt2",review early; this can change release or regression risk,2026-07-06T17:32:29Z +issue,188958,[CUDA] illegal memory read on batch_norm_backward_reduce_channels_last_kernel,https://github.com/pytorch/pytorch/issues/188958,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: edge cases, module: norms and normalization",review early; this can change release or regression risk,2026-07-06T17:31:39Z +issue,188959,[CUDA] illegal memory read on col2im_batched_kernel,https://github.com/pytorch/pytorch/issues/188959,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: convolution, module: crash, module: cuda, module: edge cases",review early; this can change release or regression risk,2026-07-06T17:31:22Z +issue,188841,triton 3.8 pin update regressing compile + float8 training numerics,https://github.com/pytorch/pytorch/issues/188841,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: floatx (formerly float8), module: inductor, module: numerical-stability, module: regression",review early; this can change release or regression risk,2026-07-06T17:23:33Z +issue,188557,operator_microbenchmark ROCm gfx950 job never completes -- runs 8h+ and is cancelled at the workflow timeout,https://github.com/pytorch/pytorch/issues/188557,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: benchmark, module: ci, module: regression, module: rocm",review early; this can change release or regression risk,2026-07-06T17:19:02Z +pr,188656,[Test] Improve test reuse in test/test_unary_ufuncs.py for out-of-tree backends,https://github.com/pytorch/pytorch/pull/188656,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-06T15:58:54Z +pr,187926,[Test] Refactor test/test_shape_ops.py,https://github.com/pytorch/pytorch/pull/187926,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-06T13:43:44Z +pr,186351,[Test] Refactor test/test_scatter_gather_ops.py,https://github.com/pytorch/pytorch/pull/186351,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-06T13:43:43Z +pr,186352,[Test] Refactor test/test_ops.py,https://github.com/pytorch/pytorch/pull/186352,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-06T13:43:37Z +pr,183278,[Inductor][Refactor] 4/N Move reduction heuristics from triton_heuristics.py into triton_codegen/,https://github.com/pytorch/pytorch/pull/183278,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",final maintainer merge/release decision,2026-07-06T10:55:42Z +issue,188674,DISABLED test_workspace_runtime_integration (__main__.TestNVUniversalGemmEpilogueFusion),https://github.com/pytorch/pytorch/issues/188674,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: inductor, module: tests, oncall: pt2",review early; this can change release or regression risk,2026-07-06T10:19:40Z +issue,188955,rms_norm autocast fp32 policy (2.12) materializes standalone cast kernels under torch.compile when dtypes already match. Peak-memory and throughput regression for bf16 models,https://github.com/pytorch/pytorch/issues/188955,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: amp (automated mixed precision), module: inductor, module: memory usage, module: performance",review early; this can change release or regression risk,2026-07-06T10:13:27Z +issue,188912,vit_base_patch16_siglip_256 amp training accuracy regression on H100 (2.13.0),https://github.com/pytorch/pytorch/issues/188912,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: amp (automated mixed precision), module: correctness (silent), module: cudnn, module: inductor",review early; this can change release or regression risk,2026-07-06T10:12:49Z +pr,188044,[Test] Refactor test/test_view_ops.py,https://github.com/pytorch/pytorch/pull/188044,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-06T09:24:25Z +pr,185798,[Test] Refactor test/test_indexing.py,https://github.com/pytorch/pytorch/pull/185798,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-07-06T08:27:13Z +pr,185802,[Test] Refactor test/test_out_dtype_op.py,https://github.com/pytorch/pytorch/pull/185802,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-06T08:26:56Z +pr,183272,[DTensor] Switch test_math_ops and test_matrix_ops to DTensorOpTestBase,https://github.com/pytorch/pytorch/pull/183272,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/dtensor,final maintainer merge/release decision,2026-07-05T17:51:54Z +pr,184947,[dynamo] fix pyright typing on torch.compiler.disable,https://github.com/pytorch/pytorch/pull/184947,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-05T00:19:46Z +pr,182014,[inductor] Fix spmd_check deadlock in recursive post_grad_passes,https://github.com/pytorch/pytorch/pull/182014,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-07-04T18:50:04Z +pr,178051,[pytree][dynamo] add generic_pytree (torch.utils.pytree) to test_tree_map.py parametrization,https://github.com/pytorch/pytorch/pull/178051,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-07-04T12:10:38Z +issue,188864,[MPS] torch.fft.hfft aborts on 5D complex64 input with dim=0,https://github.com/pytorch/pytorch/issues/188864,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: complex, module: crash, module: error checking, module: fft",review early; this can change release or regression risk,2026-07-03T03:42:04Z +pr,188229,Fix 64-bit indexing in avg_pool3d backward atomic kernel,https://github.com/pytorch/pytorch/pull/188229,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-02T19:08:59Z +pr,184795,Autogenerated operator!= in C++20,https://github.com/pytorch/pytorch/pull/184795,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/trunk, ciflow/vllm",final maintainer merge/release decision,2026-07-02T17:23:39Z +pr,188785,Upgrade submodule oneDNN to v3.12.2,https://github.com/pytorch/pytorch/pull/188785,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-07-02T16:15:03Z +issue,188691,DISABLED test_stream_invalid_device_index (__main__.TestCuda),https://github.com/pytorch/pytorch/issues/188691,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cuda, module: regression, module: tests",review early; this can change release or regression risk,2026-07-01T16:15:45Z +issue,188678,DISABLED test_redispatch_nn_functional_conv_transpose3d_cuda_float16 (__main__.TestTorchFunctionRedispatchOpsCUDA),https://github.com/pytorch/pytorch/issues/188678,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: __torch_function__, module: convolution, module: cuda, module: half",review early; this can change release or regression risk,2026-07-01T15:24:50Z +issue,188336,AArch64 CPU - Inductor benchmark Regressions,https://github.com/pytorch/pytorch/issues/188336,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: arm, module: ci, module: performance, module: regression",review early; this can change release or regression risk,2026-07-01T08:42:20Z +pr,181431,test split group,https://github.com/pytorch/pytorch/pull/181431,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-30T21:07:50Z +pr,187952,Fix out= variant recompilations for torch.max/min/topk,https://github.com/pytorch/pytorch/pull/187952,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-06-30T14:01:44Z +pr,184545,Fix UntypedStorage.new null allocator crash,https://github.com/pytorch/pytorch/pull/184545,has linked issue,70,PR closes or references open issue numbers; high-priority/blocker/regression label present,review when ready,2026-06-29T23:48:11Z +issue,187444,`torch.compile` vs eager DDP divergence with ResNet50 + AdamW lr=1.0 on two-node NCCL,https://github.com/pytorch/pytorch/issues/187444,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: ddp, module: inductor, oncall: distributed",review early; this can change release or regression risk,2026-06-29T22:37:29Z +issue,188425,torch.compile produces drastically incorrect outputs for model with symbolic dimensions and random operations,https://github.com/pytorch/pytorch/issues/188425,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: determinism, module: inductor, module: random",review early; this can change release or regression risk,2026-06-29T18:27:42Z +issue,188127,torch.compile causes persistent RSS growth during repeated inference in PyTorch 2.9 with ROCM,https://github.com/pytorch/pytorch/issues/188127,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: inductor, module: memory usage, module: performance, module: regression",review early; this can change release or regression risk,2026-06-29T18:20:38Z +pr,182033,Add a utility method for interrupting long running subprocess during run_test.py,https://github.com/pytorch/pytorch/pull/182033,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-29T18:16:28Z +issue,188218,[inductor] `LoweringException: AssertionError: max not supported for zero-dimension tensors!`,https://github.com/pytorch/pytorch/issues/188218,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: inductor, module: pooling, oncall: pt2",review early; this can change release or regression risk,2026-06-29T17:26:28Z +pr,181614,[TEST] Fix `_debug_handlers` import error in `test_stacks_handler_all_success`,https://github.com/pytorch/pytorch/pull/181614,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/torchtitan,final maintainer merge/release decision,2026-06-26T20:10:01Z +pr,186468,Fix incorrect LR curves in Chained/Sequential scheduler doc images,https://github.com/pytorch/pytorch/pull/186468,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-26T19:21:51Z +issue,187951,inductor_timm tests has been broken since 6/12,https://github.com/pytorch/pytorch/issues/187951,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: ci, module: inductor, module: regression, module: tests",review early; this can change release or regression risk,2026-06-26T19:18:48Z +issue,187343,expandable_segments:True crashes on ROCm,https://github.com/pytorch/pytorch/issues/187343,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cudacachingallocator, module: rocm",review early; this can change release or regression risk,2026-06-25T23:29:55Z +pr,185123,[c10d] Add missing runCollectiveChecks for allgather_into_tensor_coalesced in ProcessGroupWrapper,https://github.com/pytorch/pytorch/pull/185123,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-25T13:27:19Z +pr,187625,Widen optimizer hyperparameter type hints to accept int and Tensor,https://github.com/pytorch/pytorch/pull/187625,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-24T19:58:26Z +issue,187449,`torch.compile` vs eager DDP logits-sum divergence with ResNet50 + AdamW lr=0.01 on two-node NCCL,https://github.com/pytorch/pytorch/issues/187449,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: ddp, module: inductor, module: nccl",review early; this can change release or regression risk,2026-06-24T15:11:18Z +issue,65766,Temporarily disabling grad doesn't work if AMP is enabled,https://github.com/pytorch/pytorch/issues/65766,high-priority/blocker,70,high-priority/blocker/regression label present; component/oncall labels: module: amp (automated mixed precision),review early; this can change release or regression risk,2026-06-24T14:17:20Z +pr,187720,Fix stale_issues year cutoff clamping the day to 28 in every month,https://github.com/pytorch/pytorch/pull/187720,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-24T14:02:16Z +issue,187319,torch.compile FX graph does not distinguish custom ops with the same name but different mutates_args,https://github.com/pytorch/pytorch/issues/187319,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: custom-operators, module: fx, module: inductor",review early; this can change release or regression risk,2026-06-24T01:32:04Z +pr,187779,Fix MPS fused RMSNorm: do the weight multiply in fp32 to match CPU/CUDA,https://github.com/pytorch/pytorch/pull/187779,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/mps,final maintainer merge/release decision,2026-06-24T01:12:50Z +issue,187445,"`torch.compile` vs eager DDP loss/grad/logits divergence with ResNet34 + SGD lr=0.1, batch=2, TF32 allowed, and 1MB buckets",https://github.com/pytorch/pytorch/issues/187445,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: ddp, module: determinism, module: inductor",review early; this can change release or regression risk,2026-06-23T18:28:31Z +issue,187446,"`torch.compile` vs eager DDP grad/parameter divergence with ResNet50 + Adam lr=0.05, batch=1, sqrt LR scaling and step scheduler",https://github.com/pytorch/pytorch/issues/187446,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: ddp, module: inductor, module: pt2 accuracy",review early; this can change release or regression risk,2026-06-23T18:28:23Z +issue,187447,"`torch.compile` vs eager DDP parameter-fingerprint divergence with ResNet50 + Adam lr=0.1, grad accumulation, grad clipping, and 1MB buckets",https://github.com/pytorch/pytorch/issues/187447,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: ddp, module: inductor, module: pt2 optimizer",review early; this can change release or regression risk,2026-06-23T18:19:52Z +issue,187448,"`torch.compile` vs eager DDP grad/logits divergence with ResNet50 + AdamW lr=0.01, warmup cosine schedule, grad accumulation, and DDP bucket views",https://github.com/pytorch/pytorch/issues/187448,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: ddp, module: inductor, oncall: distributed",review early; this can change release or regression risk,2026-06-23T18:19:45Z +pr,187731,Fix extract_scripts step-index zero-padding off-by-one,https://github.com/pytorch/pytorch/pull/187731,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-23T16:59:39Z +issue,176784,[CUDA] AdaptiveAvgPool2d: START_IND macro causes int32 overflow → OOB read,https://github.com/pytorch/pytorch/issues/176784,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: 64-bit, module: crash, module: cuda, module: edge cases",review early; this can change release or regression risk,2026-06-23T08:05:36Z +pr,174275,Adjust to use __riscv_v to detect for RISC-V Vector Extension,https://github.com/pytorch/pytorch/pull/174275,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-20T07:48:12Z +pr,174826,add mem_get_info usage notes to CUDA memory management docs,https://github.com/pytorch/pytorch/pull/174826,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-20T01:26:30Z +pr,174667,Update tolerances for tan* functions w/ complex dtypes on XPU,https://github.com/pytorch/pytorch/pull/174667,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-19T03:48:39Z +pr,186012,Add ownership headers to GitHub workflows and a linter to enforce them,https://github.com/pytorch/pytorch/pull/186012,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/docker, ciflow/inductor-pallas",final maintainer merge/release decision,2026-06-19T01:01:43Z +pr,187130,[BE] Do not try to install CUDA-12.8 wheels in torchbench images,https://github.com/pytorch/pytorch/pull/187130,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/docker,final maintainer merge/release decision,2026-06-19T01:01:39Z +issue,184880,[ROCm] RX 9060 XT 16GB hard-capped at ~8GB usable VRAM in PyTorch — segfault on allocation beyond cap,https://github.com/pytorch/pytorch/issues/184880,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: rocm",review early; this can change release or regression risk,2026-06-18T15:40:41Z +pr,170051,Add pivoted QR decomposition to ATen and torch.linalg,https://github.com/pytorch/pytorch/pull/170051,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-06-18T11:06:22Z +issue,187320,Stable ABI isn't stable when PyTorch changes the default CUDA version,https://github.com/pytorch/pytorch/issues/187320,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: abi, module: binaries, module: cpp-extensions, module: cuda",review early; this can change release or regression risk,2026-06-15T23:57:42Z +issue,186952,GPU is not accessible on some MacOS-15 runners,https://github.com/pytorch/pytorch/issues/186952,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: arm, module: ci, module: macos",review early; this can change release or regression risk,2026-06-15T18:23:28Z +issue,186516,torch.cuda.device_count() returns 0 on ROCm when amdsmi is installed (duplicate libamd_smi ODR clash); HIP reports 1,https://github.com/pytorch/pytorch/issues/186516,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: binaries, module: rocm",review early; this can change release or regression risk,2026-06-15T14:23:41Z +pr,176321,test: replaces deprecated numpy.linalg.linalg module,https://github.com/pytorch/pytorch/pull/176321,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-14T20:03:43Z +pr,187068,[torch][profiler] Fix refcount leaks in _TensorMetadata layout/dtype getters,https://github.com/pytorch/pytorch/pull/187068,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-14T16:17:54Z +pr,186635,c10/util/ArrayRef.h: specialize std::ranges::enable_borrowed_range,https://github.com/pytorch/pytorch/pull/186635,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-12T21:53:32Z +pr,171780,Add support shared mempool for accelerator Graph,https://github.com/pytorch/pytorch/pull/171780,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-11T16:50:34Z +pr,178258,[submodule] Upgrade to protobuf v33.6,https://github.com/pytorch/pytorch/pull/178258,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/linux-aarch64, ciflow/trunk",final maintainer merge/release decision,2026-06-11T07:36:22Z +issue,184431,[vllm] [2.12 regression][multimodal] Qwen2-Audio text-then-audio_embeds: prompt_embeds vs raw-text outputs diverge under --enforce-eager,https://github.com/pytorch/pytorch/issues/184431,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: bfloat16, module: correctness (silent), module: numerical-stability, module: regression",review early; this can change release or regression risk,2026-06-11T01:16:13Z +pr,183389,Port 7 tests under test/distributed folder to XPU,https://github.com/pytorch/pytorch/pull/183389,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2026-06-10T06:08:23Z +issue,178880,`torch.compile` produces different classification output for model using E8M0 bit-manipulation quantization (`view(int32) → bitshift → clamp → uint8`),https://github.com/pytorch/pytorch/issues/178880,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-06-09T23:11:14Z +pr,185189,[Test] Make binary_ufuncs file device agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/185189,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-09T21:20:23Z +issue,184549,[vllm] [2.12 regression][Inductor] assert_size_stride (8192 vs 512) in BERT-with-rope AOT-compiled path for Nomic v2 MoE pooling model,https://github.com/pytorch/pytorch/issues/184549,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: custom-operators, module: inductor, module: pt2-dispatcher, module: regression",review early; this can change release or regression risk,2026-06-09T01:23:00Z +issue,186174,ROCm: Flash/mem-efficient SDPA hangs during batched autoregressive generation with left-padded sequences,https://github.com/pytorch/pytorch/issues/186174,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: deadlock, module: regression, module: rocm, module: sdpa",review early; this can change release or regression risk,2026-06-08T17:08:02Z +issue,186212,C++ API: at::batch_norm crashes with an internal assertion on invalid inputs,https://github.com/pytorch/pytorch/issues/186212,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cpp, module: cpu, module: crash, module: norms and normalization",review early; this can change release or regression risk,2026-06-08T14:29:40Z +pr,186519,c10/util/intrusive_ptr.h: add [[nodiscard]] to query methods,https://github.com/pytorch/pytorch/pull/186519,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-06-08T13:55:27Z +pr,186575,Update FindMKLDNN CMake configuration,https://github.com/pytorch/pytorch/pull/186575,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-08T03:07:22Z +issue,73294,Internal assert failed at rref_context.cpp,https://github.com/pytorch/pytorch/issues/73294,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: rpc, oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-06T16:29:49Z +pr,186038,add distortion benchmark,https://github.com/pytorch/pytorch/pull/186038,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-05T21:12:04Z +pr,185821,[BE][Ez]: Add c10::SmallVector reserve calls and simplify log func calls,https://github.com/pytorch/pytorch/pull/185821,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/mps,final maintainer merge/release decision,2026-06-05T19:51:27Z +issue,76865,Object-base collectives create tensors at unexpected devices,https://github.com/pytorch/pytorch/issues/76865,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",review early; this can change release or regression risk,2026-06-05T17:12:58Z +issue,82293,pytorch's checkpoint_wrapper does not save memory while fairscale's checkpoint_wrapper saves huge memory,https://github.com/pytorch/pytorch/issues/82293,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: activation checkpointing, module: fsdp, oncall: distributed, oncall: distributed parallelisms",review early; this can change release or regression risk,2026-06-05T17:11:27Z +pr,186268,[TEST][CUDA] Add some additional cleanup to `TestCudaallocator`and re-enable `test_memory_plots`,https://github.com/pytorch/pytorch/pull/186268,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-06-05T00:21:48Z +issue,169741,ReplicationPad2dwith extremely large padding value causes segmentation fault,https://github.com/pytorch/pytorch/issues/169741,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: nn",review early; this can change release or regression risk,2026-06-03T22:45:46Z +pr,185934,[NVSHMEM] Fix link_whole and -fPIC for nvshmem_device_link_target,https://github.com/pytorch/pytorch/pull/185934,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/h100-symm-mem, ciflow/trunk",final maintainer merge/release decision,2026-06-03T18:44:54Z +issue,185494,"[dynamo] LazyBatchNorm2d fails under torch.compile(dynamic=True) — ""SymIntArrayRef expected to contain only concrete integers""",https://github.com/pytorch/pytorch/issues/185494,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dynamic shapes, module: dynamo, module: nn",review early; this can change release or regression risk,2026-06-03T03:36:56Z +issue,185346,[Inductor] Performance regression of phlippe_densenet model due to non-fused Cat kernels from new fusion mode,https://github.com/pytorch/pytorch/issues/185346,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: inductor, module: performance, module: regression, oncall: pt2",review early; this can change release or regression risk,2026-06-02T18:23:29Z +issue,156819,FakeTensorUpdater doesn't support HOPs,https://github.com/pytorch/pytorch/issues/156819,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: faketensor, module: inductor, module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-06-02T17:04:18Z +pr,183056,[lint] Fix PYI036 violations and re-enable the rule,https://github.com/pytorch/pytorch/pull/183056,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/inductor, ciflow/torchtitan",final maintainer merge/release decision,2026-06-01T15:07:15Z +issue,113586,mps bug: failed assertion `[MPSNDArrayDescriptor sliceDimension:withSubrange:] error: subRange.start (8192) is not less than length of dimension[1] (512)',https://github.com/pytorch/pytorch/issues/113586,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2026-06-01T06:06:24Z +pr,182924,[Fix] DLPack toDLPackNonOwning: handle MPS storage_offset,https://github.com/pytorch/pytorch/pull/182924,ready for maintainer decision,70,approved with no detected failing/pending checks,final maintainer merge/release decision,2026-05-31T15:02:25Z +issue,179030,XPU: Segmentation Fault when using newer drivers,https://github.com/pytorch/pytorch/issues/179030,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: regression, module: xpu",review early; this can change release or regression risk,2026-05-30T22:16:27Z +pr,180828,Reuse ACCELERATOR_TYPE in test UT files,https://github.com/pytorch/pytorch/pull/180828,ready for maintainer decision,70,"approved with no detected failing/pending checks; CI lane labels: ciflow/dtensor, ciflow/h100-symm-mem, ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2026-05-29T01:10:22Z +pr,184163,[Test]Refactor test/nn/test_dropout.py to be device-agnostic,https://github.com/pytorch/pytorch/pull/184163,ready for maintainer decision,70,approved with no detected failing/pending checks; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-05-28T16:46:02Z +issue,173313,Bad perf for torch.compile for merge_attention with transposed inputs,https://github.com/pytorch/pytorch/issues/173313,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-05-26T17:55:01Z +issue,181869,[vllm] [torch nightly] AOTAutogradCache.save: _pickle.PicklingError on triton launcher (DFlashDraftModel speculative-decode path),https://github.com/pytorch/pytorch/issues/181869,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: inductor, module: pickle, module: pt2-dispatcher",review early; this can change release or regression risk,2026-05-26T17:54:17Z +issue,184027,[inductor] `torch.compile` crashes on convolutions when weight is a plain tensor — stride prediction mismatch,https://github.com/pytorch/pytorch/issues/184027,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: convolution, module: crash, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-05-26T16:49:54Z +issue,120601,Torchrun arguments shadowing script arguments,https://github.com/pytorch/pytorch/issues/120601,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: intel, module: regression",review early; this can change release or regression risk,2026-05-26T02:36:50Z +issue,183401,[inductor][cpu] resnet50_quantized_qat CPP wrapper torch._inductor.exc.InductorError: CppCompileError: C++ compile error in 2026-05-11 nightly release,https://github.com/pytorch/pytorch/issues/183401,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: inductor, module: regression, oncall: pt2",review early; this can change release or regression risk,2026-05-21T18:26:50Z +issue,161295,UNSTABLE inductor-periodic / cuda12.8-py3.10-gcc9-sm80 / test (inductor_torchbench_smoketest_perf),https://github.com/pytorch/pytorch/issues/161295,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: ci, module: inductor, oncall: pt2",review early; this can change release or regression risk,2026-05-21T18:16:03Z +issue,177834,`torch.nn.LSTM.forward` produces incorrect outputs (or crashes) on ROCm,https://github.com/pytorch/pytorch/issues/177834,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: crash, module: rnn, module: rocm",review early; this can change release or regression risk,2026-05-20T22:35:45Z +issue,178726,Greatly increased CPU memory use on aarch64 H100 node from torch 2.9.0 to torch 2.10.0,https://github.com/pytorch/pytorch/issues/178726,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: arm, module: cuda, module: memory usage, module: regression",review early; this can change release or regression risk,2026-05-20T21:20:51Z +issue,87960,crash when call `torch.set_num_interop_threads` twice,https://github.com/pytorch/pytorch/issues/87960,high-priority/blocker,70,high-priority/blocker/regression label present; component/oncall labels: module: crash,review early; this can change release or regression risk,2026-05-20T18:32:03Z +issue,183890,torch.compile hits RecursionError on nn.Module with custom __getattr__,https://github.com/pytorch/pytorch/issues/183890,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: dynamo, oncall: pt2",review early; this can change release or regression risk,2026-05-18T21:35:07Z +issue,182949,[CD] Drop CPython 3.13t from binary build matrix (manylinux upstream removed it 2026-05-07),https://github.com/pytorch/pytorch/issues/182949,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: binaries, module: ci, module: dependency bug, module: docker",review early; this can change release or regression risk,2026-05-14T14:23:49Z +issue,13246,DataLoader num_workers > 0 causes CPU memory from parent process to be replicated in all worker processes,https://github.com/pytorch/pytorch/issues/13246,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: dataloader, module: dependency bug, module: memory usage, module: molly-guard",review early; this can change release or regression risk,2026-05-10T18:38:26Z +issue,91570,Security policy impractical / lacks contact information?,https://github.com/pytorch/pytorch/issues/91570,high-priority/blocker,70,high-priority/blocker/regression label present; component/oncall labels: module: docs,review early; this can change release or regression risk,2026-05-06T20:41:45Z +issue,182263,"libtorch_cpu.so BLAS symbols are interposable, causing crashes when another BLAS is present in the process",https://github.com/pytorch/pytorch/issues/182263,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: binaries, module: crash, module: openblas",review early; this can change release or regression risk,2026-05-04T17:30:57Z +issue,179891,"torch.xpu.get_device_properties() segfaults on Intel ARC PRO B70 (Battlemage/BMG, device 0xe223)",https://github.com/pytorch/pytorch/issues/179891,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: xpu",review early; this can change release or regression risk,2026-05-02T17:28:20Z +issue,177714,XPU: Fatal Level Zero `RuntimeErrors` instead of recoverable `OutOfMemoryError`,https://github.com/pytorch/pytorch/issues/177714,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: memory usage, module: xpu",review early; this can change release or regression risk,2026-04-29T21:55:24Z +issue,181362,DISABLED deit_tiny_patch16_224.fb_in1k fail_accuracy on ROCm dynamic_inductor_timm_training,https://github.com/pytorch/pytorch/issues/181362,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: inductor, module: pt2 accuracy, module: regression, module: rocm",review early; this can change release or regression risk,2026-04-24T11:30:22Z +issue,139989,Support AC with graph break,https://github.com/pytorch/pytorch/issues/139989,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: activation checkpointing, module: dynamo, module: higher order operators, module: pt2-dispatcher",review early; this can change release or regression risk,2026-04-23T22:12:01Z +issue,143487,Aborted (core dumped) in `replication_pad1d`,https://github.com/pytorch/pytorch/issues/143487,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2026-04-22T05:25:08Z +issue,143484,Segmentation fault (core dumped) in `gru_cell`,https://github.com/pytorch/pytorch/issues/143484,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: empty tensor, module: error checking",review early; this can change release or regression risk,2026-04-22T05:23:11Z +issue,181024,[CUDA] `test_dlpack_tensor_on_different_device` fails with `torch.device` instead of `torch.cuda.device`,https://github.com/pytorch/pytorch/issues/181024,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cuda, module: dlpack, module: regression, module: tests",review early; this can change release or regression risk,2026-04-21T21:05:52Z +issue,179804,torch.compile + aot_eager: backward ignores inner autocast(enabled=False) vs eager,https://github.com/pytorch/pytorch/issues/179804,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: aotdispatch, module: autograd, module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-04-21T15:44:21Z +issue,141896,"CheckpointError with checkpoint(..., use_reentrant=False) & autocast()",https://github.com/pytorch/pytorch/issues/141896,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: activation checkpointing, module: amp (automated mixed precision)",review early; this can change release or regression risk,2026-04-19T10:12:12Z +issue,178939,Pytorch needs Kineto even when using USE_KINETO=OFF,https://github.com/pytorch/pytorch/issues/178939,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: build, module: regression",review early; this can change release or regression risk,2026-04-13T17:10:46Z +issue,168058,[DTensor] inconsistency of distributed RNG (Shard(0) != Shard(1) or 8 GPU != 4 GPU != 1 GPU),https://github.com/pytorch/pytorch/issues/168058,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: dtensor, oncall: distributed",review early; this can change release or regression risk,2026-04-06T17:45:17Z +issue,41486,"LSTMs leak memory in CPU PyTorch 1.5.1, 1.6, and 1.7 on Linux",https://github.com/pytorch/pytorch/issues/41486,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: memory usage, module: rnn",review early; this can change release or regression risk,2026-03-27T05:49:45Z +issue,145702,PyTorch VS2022 official build Windows binary illegal instruction on AVX2(max ISA level) CPU,https://github.com/pytorch/pytorch/issues/145702,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: build, module: cpu, module: crash, module: windows",review early; this can change release or regression risk,2026-03-26T08:22:43Z +issue,165448,`import torch` fails on 3.13 Linux nightly jobs with AST parsing errors,https://github.com/pytorch/pytorch/issues/165448,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: binaries, module: regression",review early; this can change release or regression risk,2026-03-25T11:16:27Z +issue,152302,NCCL out of memory error after updating to PyTorch 2.7,https://github.com/pytorch/pytorch/issues/152302,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: nccl, module: regression, oncall: distributed",review early; this can change release or regression risk,2026-03-21T02:43:04Z +issue,152954,DTensor placement propagation for `as_strided` missing,https://github.com/pytorch/pytorch/issues/152954,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: pt2-dispatcher, oncall: distributed, oncall: pt2",review early; this can change release or regression risk,2026-03-19T22:56:29Z +issue,58743,Python Array API Compatibility Tracker,https://github.com/pytorch/pytorch/issues/58743,high-priority/blocker,70,high-priority/blocker/regression label present; component/oncall labels: module: python array api,review early; this can change release or regression risk,2026-03-16T20:56:13Z +issue,142048,Device check missing in torch.linalg.solve_triangular leading to hard crash,https://github.com/pytorch/pytorch/issues/142048,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: linear algebra, module: mps",review early; this can change release or regression risk,2026-03-14T10:13:35Z +issue,172537,[Regression] 20x slower vision model forward pass in PyTorch 2.9.1 vs 2.8.0,https://github.com/pytorch/pytorch/issues/172537,high-priority/blocker,70,high-priority/blocker/regression label present; component/oncall labels: module: performance,review early; this can change release or regression risk,2026-03-13T19:23:37Z +issue,170787,[ARM] Unit test regression AArch64 - int4mm kernels require bf16 hw support,https://github.com/pytorch/pytorch/issues/170787,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: arm, module: regression",review early; this can change release or regression risk,2026-03-12T14:42:25Z +issue,30774,Natively Declarable Fast-path Functions,https://github.com/pytorch/pytorch/issues/30774,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: dispatch, module: internals",review early; this can change release or regression risk,2026-03-11T19:31:57Z +issue,124484,HOP dispatch isn't faithful,https://github.com/pytorch/pytorch/issues/124484,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review early; this can change release or regression risk,2026-03-11T18:36:02Z +issue,175928,[release 2.11][vllm] Language Models Tests (Extra Standard) 1 test_models[float-jason9693/Qwen2.5-1.5B-apeach],https://github.com/pytorch/pytorch/issues/175928,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: regression, module: tests, module: vllm, oncall: releng",review early; this can change release or regression risk,2026-03-11T14:47:43Z +issue,167373,Torch Tensor -> Nanobind ndarray breaks cuda graphs on torch 2.9,https://github.com/pytorch/pytorch/issues/167373,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cuda graphs, module: dlpack, module: regression",review early; this can change release or regression risk,2026-03-10T22:12:43Z +issue,163501,Improve Registration Error message with const SymIntArrayRef&in PyTorch 2.5+,https://github.com/pytorch/pytorch/issues/163501,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cpp-extensions, module: dispatch, module: dynamic shapes, module: regression",review early; this can change release or regression risk,2026-03-10T16:09:43Z +issue,157128,"Windows Source Build Fails with OSError: [WinError 126] on aoti_custom_ops.dll for RTX 5080 (sm_120), Pre-built PyTorch Works",https://github.com/pytorch/pytorch/issues/157128,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: aotinductor, module: build, module: regression, module: windows",review early; this can change release or regression risk,2026-03-09T23:25:25Z +issue,174704,Recompile time with torch.compile 2.10,https://github.com/pytorch/pytorch/issues/174704,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: compile-time, module: dynamic shapes, module: fsdp, module: regression",review early; this can change release or regression risk,2026-03-07T01:47:55Z +issue,176285,"AArch64 test_cpu_repro - Expected to not find ""transpose_mxn"" but found it",https://github.com/pytorch/pytorch/issues/176285,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: arm, module: ci, module: inductor, module: regression",review early; this can change release or regression risk,2026-03-04T11:59:53Z +issue,142836,[MPS] Incorrect output from convolution ops with large dimensions,https://github.com/pytorch/pytorch/issues/142836,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: correctness (silent), module: mps, module: regression",review early; this can change release or regression risk,2026-03-01T20:13:15Z +issue,142834,Segmentation fault in `replication_pad2d_backward`,https://github.com/pytorch/pytorch/issues/142834,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: edge cases, module: nn",review early; this can change release or regression risk,2026-02-24T15:42:59Z +issue,175436,Intermittent SIGILL in `torch.sin` on CPU (`torch==2.10.0+cpu`) on Xeon E5-2670; crash in `mkl_vml_kernel_dSin_*`,https://github.com/pytorch/pytorch/issues/175436,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: cpu, module: crash, module: intel, module: mkl",review early; this can change release or regression risk,2026-02-23T18:47:29Z +issue,142459,Segmentation fault (core dumped) in `max_pool3d_with_indices_backward`,https://github.com/pytorch/pytorch/issues/142459,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: edge cases, module: error checking, module: nn",review early; this can change release or regression risk,2026-02-18T14:32:11Z +issue,142457,Aborted (core dumped) in `slow_conv_transpose3d`,https://github.com/pytorch/pytorch/issues/142457,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: convolution, module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2026-02-18T14:31:11Z +issue,168997,[Umbrella issue] Deployment of index updates to download.pytorch.org - Switch dependencies that are on pypi to utilize pypi's CDN,https://github.com/pytorch/pytorch/issues/168997,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: binaries, module: ci",review early; this can change release or regression risk,2026-02-03T19:38:02Z +issue,94765,MPS internal error in `torch.gather` when last dimension is a singleton dimension,https://github.com/pytorch/pytorch/issues/94765,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2026-01-30T06:39:30Z +issue,173574,`torch.arange` triggers `SIGFPE` (Floating Point Exception) instead of RuntimeError when writing float results to `int64` out tensor,https://github.com/pytorch/pytorch/issues/173574,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2026-01-28T01:25:30Z +issue,150098,Aborted (core dumped) raw pointer API is exposed directly to Python,https://github.com/pytorch/pytorch/issues/150098,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: crash, module: cuda, module: error checking",review early; this can change release or regression risk,2026-01-27T13:07:06Z +issue,172907,[CI] Changes to install scripts in lintrunner.toml are not tested in CI,https://github.com/pytorch/pytorch/issues/172907,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: ci, module: typing",review early; this can change release or regression risk,2026-01-26T18:14:46Z +issue,172798,`backward` does not respect `with torch.device(device)`,https://github.com/pytorch/pytorch/issues/172798,high-priority/blocker,70,"high-priority/blocker/regression label present; component/oncall labels: module: autograd, module: regression",review early; this can change release or regression risk,2026-01-26T18:12:31Z +issue,51039,Status of pip wheels with _GLIBCXX_USE_CXX11_ABI=1,https://github.com/pytorch/pytorch/issues/51039,high-priority/blocker,70,"high-priority/blocker/regression label present; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: abi, module: binaries, module: cpp",route to design/research decision before patch planning,2025-05-01T18:32:44Z +issue,93880,"`PYTORCH_DEBUG_MODE`, better invalid index embedding lookup error message on cuda",https://github.com/pytorch/pytorch/issues/93880,high-priority/blocker,70,high-priority/blocker/regression label present; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2024-09-14T15:43:15Z +issue,23756,[feature request] Core API for invertible/inplace and flow-like ops + memory-saving (hookless?) reversible sequential container for RevNets to allow for much larger batch-sizes in academic setting,https://github.com/pytorch/pytorch/issues/23756,high-priority/blocker,70,"high-priority/blocker/regression label present; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: nn",route to design/research decision before patch planning,2023-08-21T20:47:19Z +issue,68768,[feature request] making pytorch less noisy,https://github.com/pytorch/pytorch/issues/68768,high-priority/blocker,70,"high-priority/blocker/regression label present; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: ux",route to design/research decision before patch planning,2022-06-02T05:17:44Z +issue,44428,Support for CUDA matrix multiplication on long (and other integer) tensors,https://github.com/pytorch/pytorch/issues/44428,high-priority/blocker,70,"high-priority/blocker/regression label present; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: linear algebra",route to design/research decision before patch planning,2021-11-12T03:38:36Z +pr,186418,Fix AOT export GraphModule serialization,https://github.com/pytorch/pytorch/pull/186418,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T19:16:31Z +pr,186431,Fix ModuleTracker with Dynamo runtime modules,https://github.com/pytorch/pytorch/pull/186431,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T18:56:47Z +pr,186443,Validate random_ and uniform_ meta arguments,https://github.com/pytorch/pytorch/pull/186443,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T18:53:27Z +pr,184347,[inductor] Add third reduction tile for strided block pointers,https://github.com/pytorch/pytorch/pull/184347,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T18:41:19Z +pr,186454,Warn on unsupported custom op schema types,https://github.com/pytorch/pytorch/pull/186454,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T18:37:58Z +pr,186459,Fix fake view strides for AOT group norm tracing,https://github.com/pytorch/pytorch/pull/186459,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T18:25:42Z +pr,186463,Split foreach bool inplace error samples,https://github.com/pytorch/pytorch/pull/186463,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T18:05:31Z +pr,186464,Fix singleton quotient divisibility replacement,https://github.com/pytorch/pytorch/pull/186464,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T18:02:24Z +pr,186470,Raise user error for joint parameter mutation export,https://github.com/pytorch/pytorch/pull/186470,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T17:47:10Z +pr,186471,Handle symbolic tensors in overlap status,https://github.com/pytorch/pytorch/pull/186471,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T17:42:15Z +pr,185926,Preserve tuple arguments in export serde,https://github.com/pytorch/pytorch/pull/185926,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T17:39:36Z +pr,185929,Fix Dynamo benchmark model state isolation,https://github.com/pytorch/pytorch/pull/185929,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T17:33:56Z +pr,186007,Preserve eager post accumulate hook ordering in AOTAutograd,https://github.com/pytorch/pytorch/pull/186007,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T16:56:34Z +pr,186833,Remove stale export OpInfo xfails,https://github.com/pytorch/pytorch/pull/186833,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T16:34:31Z +pr,186478,Fix AOTAutograd resized out tensor mutation replay,https://github.com/pytorch/pytorch/pull/186478,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T16:27:56Z +pr,186488,Improve schema errors for symbolic int list mismatches,https://github.com/pytorch/pytorch/pull/186488,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T16:20:31Z +pr,186495,Initialize compile context for non-strict export tracing,https://github.com/pytorch/pytorch/pull/186495,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T16:09:24Z +pr,186503,Preserve input aliases for multi-output view returns,https://github.com/pytorch/pytorch/pull/186503,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T15:31:42Z +pr,186506,Register missing CIA decomposition py_impls,https://github.com/pytorch/pytorch/pull/186506,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T15:23:46Z +pr,186517,Detect aliased subclass inputs after desugaring,https://github.com/pytorch/pytorch/pull/186517,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T15:05:12Z +pr,185793,Add ONNX lowering for prims.remainder,https://github.com/pytorch/pytorch/pull/185793,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T15:02:11Z +pr,184432,Serialize cudagraph tree runtime,https://github.com/pytorch/pytorch/pull/184432,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T14:57:12Z +pr,184436,Fix Triton async compile cache keys,https://github.com/pytorch/pytorch/pull/184436,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T14:52:50Z +pr,186531,Remove stale Dynamo skip for max_pool1d pooling test,https://github.com/pytorch/pytorch/pull/186531,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T14:39:01Z +pr,186533,Fix SymInt vector dispatcher signature detection,https://github.com/pytorch/pytorch/pull/186533,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T14:33:01Z +pr,187601,[functorch] Fix dynamic layers under inference_mode,https://github.com/pytorch/pytorch/pull/187601,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T14:25:35Z +pr,187592,Support registered HOPs in FlopCounterMode,https://github.com/pytorch/pytorch/pull/187592,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T13:16:23Z +pr,186561,Preserve autograd metadata for functionalized inplace views,https://github.com/pytorch/pytorch/pull/186561,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T11:56:59Z +pr,186409,Fix subclass view replay with dynamic metadata,https://github.com/pytorch/pytorch/pull/186409,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T11:44:39Z +pr,185861,Fix ViewMeta serialization for special views,https://github.com/pytorch/pytorch/pull/185861,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T11:37:57Z +pr,185897,Support concrete enum types in torch.library.define schemas,https://github.com/pytorch/pytorch/pull/185897,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T10:30:52Z +pr,183628,Fix fractional max pool RNG in Inductor,https://github.com/pytorch/pytorch/pull/183628,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T09:29:00Z +pr,183701,Fix CPU vectorized reduction tail stores,https://github.com/pytorch/pytorch/pull/183701,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T08:51:21Z +pr,185995,Fix stale pending-backward state for dead CUDA graph paths,https://github.com/pytorch/pytorch/pull/185995,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T08:46:38Z +pr,184667,Fix cond branch tensor constants in run_decompositions,https://github.com/pytorch/pytorch/pull/184667,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T08:20:27Z +pr,184695,Fix all-masked TransformerEncoder compile consistency,https://github.com/pytorch/pytorch/pull/184695,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T07:35:30Z +pr,184709,Fix Dynamo tensor subclass constructor side effects,https://github.com/pytorch/pytorch/pull/184709,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T07:25:01Z +pr,184917,Make FX traceback preserve state thread-local,https://github.com/pytorch/pytorch/pull/184917,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T07:03:00Z +pr,184928,Fix linalg_pinv tensor tolerance broadcasting,https://github.com/pytorch/pytorch/pull/184928,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T06:57:51Z +pr,185059,Fix non-strict export tensor subclass attr tracing,https://github.com/pytorch/pytorch/pull/185059,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T06:35:29Z +pr,184015,[inductor] Handle scalar workaround args in pattern checks,https://github.com/pytorch/pytorch/pull/184015,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T06:31:50Z +pr,185088,Avoid graphsafe RNG inputs in descriptor compile,https://github.com/pytorch/pytorch/pull/185088,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T06:04:40Z +pr,184747,Remove no-op SDPA masks before backend dispatch,https://github.com/pytorch/pytorch/pull/184747,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T04:16:39Z +pr,183879,[inductor][cpu] Enable int8 WOQ GEMM template on Arm,https://github.com/pytorch/pytorch/pull/183879,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T04:13:41Z +pr,183908,Fix deferred runtime assert symbol availability,https://github.com/pytorch/pytorch/pull/183908,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T04:05:11Z +pr,183958,Enable SDPA constraint cache by default,https://github.com/pytorch/pytorch/pull/183958,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T03:48:03Z +pr,184948,Fix dynamo export for tensor closures in output pytrees,https://github.com/pytorch/pytorch/pull/184948,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T02:25:10Z +pr,184966,Preserve unflattened module forward signatures,https://github.com/pytorch/pytorch/pull/184966,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T02:11:28Z +pr,184974,Fix register_autograd wrappers under torch.func.grad,https://github.com/pytorch/pytorch/pull/184974,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T01:57:49Z +pr,184975,Support FakeTensor module materialization via to_empty,https://github.com/pytorch/pytorch/pull/184975,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T01:53:49Z +pr,184976,Fix FakeTensor Module._apply conversion with weakrefs,https://github.com/pytorch/pytorch/pull/184976,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T01:51:21Z +pr,184978,Improve export error for untraceable lifted constants,https://github.com/pytorch/pytorch/pull/184978,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T01:31:12Z +pr,184981,Fix torch._numpy bool numeric subtract,https://github.com/pytorch/pytorch/pull/184981,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T01:27:26Z +pr,184983,Make SDPA backend checks safe for unbacked head dims,https://github.com/pytorch/pytorch/pull/184983,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-08T01:23:08Z +pr,184987,Fix index_select refs bounds semantics,https://github.com/pytorch/pytorch/pull/184987,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T01:18:24Z +pr,184995,Fix AOT getitem replay under inference_mode,https://github.com/pytorch/pytorch/pull/184995,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T01:01:23Z +pr,185008,Add DebugMode tensor hash export,https://github.com/pytorch/pytorch/pull/185008,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor,review diff and decide acceptance path,2026-07-08T00:53:37Z +pr,185009,Add fuzzy DebugMode hash mismatch comparison,https://github.com/pytorch/pytorch/pull/185009,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor,review diff and decide acceptance path,2026-07-08T00:50:28Z +pr,184079,Enable graph deduplication for Inductor compile,https://github.com/pytorch/pytorch/pull/184079,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T00:35:13Z +pr,185025,[export][serde] Initialize symbol ranges before constants,https://github.com/pytorch/pytorch/pull/185025,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T00:32:29Z +pr,186953,Fix CPU amin/amax backward rematerialization mismatch,https://github.com/pytorch/pytorch/pull/186953,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-08T00:16:47Z +pr,185050,Handle SymInt sizes in lazy module initialization,https://github.com/pytorch/pytorch/pull/185050,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-08T00:03:22Z +pr,185051,Fix FX GraphModule serialization of string annotations,https://github.com/pytorch/pytorch/pull/185051,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T23:59:51Z +pr,185052,Fix Inductor reuse for disjoint strided slices,https://github.com/pytorch/pytorch/pull/185052,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T23:56:48Z +pr,186879,Fix export dynamic shapes for variadic kwargs,https://github.com/pytorch/pytorch/pull/186879,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T23:46:33Z +pr,185105,Fix export move_to_device_pass device rewrites,https://github.com/pytorch/pytorch/pull/185105,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T23:28:41Z +pr,185106,Fix non-strict export state aliases,https://github.com/pytorch/pytorch/pull/185106,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T23:24:33Z +pr,185121,Raise TypeError for invalid torch.export.load inputs,https://github.com/pytorch/pytorch/pull/185121,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T23:18:53Z +pr,185126,Optimize GB_REGISTRY linter scanning,https://github.com/pytorch/pytorch/pull/185126,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T22:44:44Z +pr,185132,Fix custom op autograd defaults with tensorlist outputs,https://github.com/pytorch/pytorch/pull/185132,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T22:41:33Z +pr,185140,Preserve reentrant checkpoint in non-strict export,https://github.com/pytorch/pytorch/pull/185140,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T22:34:40Z +pr,185147,Fix Z3 translation validation for torch.sym_not,https://github.com/pytorch/pytorch/pull/185147,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T22:32:09Z +pr,185155,Avoid warning spam for backend fallback graph breaks,https://github.com/pytorch/pytorch/pull/185155,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T22:25:38Z +pr,185166,Fix export slice bounds across dynamic dim,https://github.com/pytorch/pytorch/pull/185166,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T22:16:53Z +pr,185170,Fix ExportedProgram pickle after graph_module edits,https://github.com/pytorch/pytorch/pull/185170,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T22:13:22Z +pr,185191,Support wrapper subclass storages in HOP alias checks,https://github.com/pytorch/pytorch/pull/185191,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T21:53:38Z +pr,185293,Require bool conditions for torch.where,https://github.com/pytorch/pytorch/pull/185293,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/mps,review diff and decide acceptance path,2026-07-07T21:18:40Z +pr,186859,Recover fake/proxy views from functional tensor bases,https://github.com/pytorch/pytorch/pull/186859,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T21:17:17Z +pr,184787,[export] Deserialize PT2 schema in C++ during load,https://github.com/pytorch/pytorch/pull/184787,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T21:11:39Z +pr,185073,[export] Treat free new-tracer state as constants,https://github.com/pytorch/pytorch/pull/185073,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T20:37:31Z +pr,187012,Support optional Tensor list schemas,https://github.com/pytorch/pytorch/pull/187012,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T18:59:07Z +pr,185369,Fix dynamic output sizes for adaptive avg pool lowering,https://github.com/pytorch/pytorch/pull/185369,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T18:01:43Z +pr,185404,[inductor] Fold small-row expanded bmm to mm in post-grad,https://github.com/pytorch/pytorch/pull/185404,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T17:44:24Z +pr,186894,Fix non-strict export of vmap tensor indexing,https://github.com/pytorch/pytorch/pull/186894,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T17:37:06Z +pr,184065,Fix static launcher 16-bit float scalars,https://github.com/pytorch/pytorch/pull/184065,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T17:30:43Z +pr,185450,Fix reinplace alias checks for index_put,https://github.com/pytorch/pytorch/pull/185450,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T17:16:47Z +pr,184083,Preserve user Triton allocator in Inductor,https://github.com/pytorch/pytorch/pull/184083,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T16:45:45Z +pr,185483,Add fake support for pad_packed_sequence,https://github.com/pytorch/pytorch/pull/185483,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T16:36:25Z +pr,185564,Handle pre-dispatch export of subclass tensor constants,https://github.com/pytorch/pytorch/pull/185564,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T15:33:38Z +pr,187956,Propagate FX subgraph replacement meta values,https://github.com/pytorch/pytorch/pull/187956,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T15:27:38Z +pr,185569,Handle symbolic bool short-circuit with unbacked terms,https://github.com/pytorch/pytorch/pull/185569,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T15:22:19Z +pr,185718,Fix fake functorch-wrapped tensor handling in jacfwd vmap,https://github.com/pytorch/pytorch/pull/185718,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T14:44:05Z +pr,185755,Use gmem efficient attention kernel for long fp32 KV,https://github.com/pytorch/pytorch/pull/185755,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T14:12:52Z +pr,185772,Fix CPU small-output bilinear upsample precision in Inductor,https://github.com/pytorch/pytorch/pull/185772,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T12:38:12Z +pr,185774,Fix non-strict export for Tensor.numpy,https://github.com/pytorch/pytorch/pull/185774,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T12:30:33Z +pr,185408,Fix torch logging to avoid closed pytest stderr,https://github.com/pytorch/pytorch/pull/185408,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T12:19:00Z +pr,185526,Fix Dynamo tracing of LazyGraphModule,https://github.com/pytorch/pytorch/pull/185526,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T12:16:19Z +pr,185786,Handle non-natural SymPy powers in value range interp,https://github.com/pytorch/pytorch/pull/185786,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T12:03:15Z +pr,185803,Fix joint export for tied parameter aliases,https://github.com/pytorch/pytorch/pull/185803,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T11:21:52Z +pr,185914,Log aot_export fw_metadata instead of printing it,https://github.com/pytorch/pytorch/pull/185914,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T11:13:26Z +pr,185922,Fix fake one_hot invalid class count checks,https://github.com/pytorch/pytorch/pull/185922,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T10:34:05Z +pr,184178,Fix masked Triton conv load pointers,https://github.com/pytorch/pytorch/pull/184178,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T10:19:14Z +pr,186074,Fix multilabel_margin_loss decomposition sentinel handling,https://github.com/pytorch/pytorch/pull/186074,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T07:11:13Z +pr,184207,Release max autotune benchmark artifacts,https://github.com/pytorch/pytorch/pull/184207,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-07T06:52:34Z +pr,186111,Fix stale mutated input history in AOTAutograd,https://github.com/pytorch/pytorch/pull/186111,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T06:50:00Z +pr,186118,Preserve guard eval defaulting without callback churn,https://github.com/pytorch/pytorch/pull/186118,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T06:46:54Z +pr,186188,Fix FakeTensor wrapping for meta tensor constructor,https://github.com/pytorch/pytorch/pull/186188,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T06:14:15Z +pr,186194,Canonicalize trunc of symbolic integer division by one,https://github.com/pytorch/pytorch/pull/186194,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T05:55:45Z +pr,186199,Fix functionalization return mapping for mixed mutable out ops,https://github.com/pytorch/pytorch/pull/186199,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T05:47:07Z +pr,186201,Deduplicate symbolic cond inputs after export passes,https://github.com/pytorch/pytorch/pull/186201,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T05:44:48Z +pr,187604,Fix fake slice pending unbacked replacements,https://github.com/pytorch/pytorch/pull/187604,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T05:25:43Z +pr,186225,Preserve NJT nested ints in AOT subclass outputs,https://github.com/pytorch/pytorch/pull/186225,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T05:20:37Z +pr,186261,Fix pre-dispatch tracing for saved tensor detach,https://github.com/pytorch/pytorch/pull/186261,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-07T05:03:06Z +pr,186291,Fix export distribution validation assertions,https://github.com/pytorch/pytorch/pull/186291,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-07T04:31:47Z +pr,186315,Preserve extra autocast keys on FakeTensor,https://github.com/pytorch/pytorch/pull/186315,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-06T18:18:35Z +pr,186321,Fix PythonMod axiom reuse after size refinement,https://github.com/pytorch/pytorch/pull/186321,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-06T18:12:26Z +pr,186322,Add explicit schema registration for TorchBind methods,https://github.com/pytorch/pytorch/pull/186322,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-06T18:10:11Z +pr,186328,Preserve strides for functionalized in-place mutations,https://github.com/pytorch/pytorch/pull/186328,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers,review diff and decide acceptance path,2026-07-06T18:06:16Z +pr,186339,Improve torch.compile sibling output grad error,https://github.com/pytorch/pytorch/pull/186339,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-06T16:55:55Z +pr,186342,Preserve nested fake subclass inputs in AOTAutograd,https://github.com/pytorch/pytorch/pull/186342,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-07-06T16:48:47Z +pr,185079,Fix invalid Dynamo setattr exception timing,https://github.com/pytorch/pytorch/pull/185079,ready for maintainer decision,67,"needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review diff and decide acceptance path,2026-07-03T03:44:16Z +pr,186527,Support export of packed LSTM with data-dependent lengths,https://github.com/pytorch/pytorch/pull/186527,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review diff and decide acceptance path,2026-06-23T19:28:50Z +pr,185000,debug_mode: avoid hashing async collectives before wait,https://github.com/pytorch/pytorch/pull/185000,ready for maintainer decision,67,needs-review label present; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor,review diff and decide acceptance path,2026-06-11T03:18:46Z +issue,189322,DISABLED test_aoti_debug_printer_codegen_cuda (__main__.AOTInductorTestDualWrapper),https://github.com/pytorch/pytorch/issues/189322,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2",categorize and decide next maintainer owner/action,2026-07-09T01:17:51Z +issue,189323,DISABLED test_fake_export_contiguous_cpu_float32 (__main__.TestExportOpInfoCPU),https://github.com/pytorch/pytorch/issues/189323,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-07-09T01:17:16Z +issue,119537,DISABLED test_source_multithreaded_open_in_scope_work_in_main_thread_False (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119537,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-09T01:17:04Z +issue,119576,DISABLED test_source_multithreaded_multiple_preexisting_work_in_main_thread_True (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119576,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-09T01:17:01Z +issue,189308,`torch.distributed.checkpoint`: `.metadata` loaded with raw `pickle.load` on the public `load()` path — same class you already warn about on `torch.load`,https://github.com/pytorch/pytorch/issues/189308,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",categorize and decide next maintainer owner/action,2026-07-08T21:48:14Z +issue,189309,[Stable ABI] Add a has_side_effects op tag. DCE silently deletes side-effect-only op calls,https://github.com/pytorch/pytorch/issues/189309,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-07-08T21:45:56Z +issue,189298,torch.jit.trace fails with MaxUnpool in 2.13,https://github.com/pytorch/pytorch/issues/189298,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-07-08T19:55:36Z +issue,189293,AttributeError: Can't get local object 'WeakValueDictionary.__init__..remove' during AOTAutograd graph caching on Python 3.13,https://github.com/pytorch/pytorch/issues/189293,needs triage,65,"triage signal present; component/oncall labels: module: aotdispatch, module: pickle, module: pt2-dispatcher, module: python version",categorize and decide next maintainer owner/action,2026-07-08T19:26:04Z +issue,119526,DISABLED test_source_multithreaded_multiple_preexisting_work_in_main_thread_False (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119526,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-08T19:02:44Z +issue,189280,[CPU][quant] operator_benchmark: qcat ~3-4x slower (likely) and fake_quantize ~700x slower (possible) since ~Nov 2025,https://github.com/pytorch/pytorch/issues/189280,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-07-08T17:31:39Z +issue,188833,[pipelining] Interleaved1F1B warmup head-of-line bubble from front-loaded RECV,https://github.com/pytorch/pytorch/issues/188833,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2026-07-08T17:31:15Z +issue,188836,DISABLED test_dropout_parity_dynamic_shapes (__main__.TestDropoutAlignRandomEager),https://github.com/pytorch/pytorch/issues/188836,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: pt2,categorize and decide next maintainer owner/action,2026-07-08T17:17:44Z +issue,188896,DISABLED test_launch_elastic_worker_raise_exception (__main__.ElasticLaunchTest),https://github.com/pytorch/pytorch/issues/188896,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: distributed, oncall: r2p",categorize and decide next maintainer owner/action,2026-07-08T17:16:02Z +issue,189245,[cpp_extension] Stale .lock file from FileBaton causes permanent deadlock when compilation process is killed,https://github.com/pytorch/pytorch/issues/189245,needs triage,65,"triage signal present; component/oncall labels: module: cpp-extensions, module: deadlock",categorize and decide next maintainer owner/action,2026-07-08T16:35:40Z +issue,119536,DISABLED test_source_multithreaded_complex_work_in_main_thread_True (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119536,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-08T13:12:38Z +issue,119490,DISABLED test_source_multithreaded_complex_work_in_main_thread_False (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119490,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-08T07:30:07Z +issue,189228,[symm_mem] one/two-shot all-reduce deadlocks when issued from multiple streams: ops lack channel plumbing,https://github.com/pytorch/pytorch/issues/189228,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-07-08T03:18:28Z +issue,119514,DISABLED test_source_multithreaded_close_in_scope_work_in_main_thread_True (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119514,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-08T01:09:15Z +issue,119364,DISABLED test_source_multithreaded_close_in_scope_work_in_main_thread_False (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119364,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-07T13:25:19Z +issue,75171,torch.jit.load fails when path contains non-ascii characters,https://github.com/pytorch/pytorch/issues/75171,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: windows, oncall: jit",categorize and decide next maintainer owner/action,2026-07-07T09:23:51Z +issue,119357,DISABLED test_source_multithreaded_basic_work_in_main_thread_True (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119357,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-07T07:52:38Z +issue,119346,DISABLED test_source_multithreaded_basic_work_in_main_thread_False (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/119346,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-07-07T01:15:49Z +issue,189098,[RFC] Expose Kineto custom logger registration and protocol-based export to PyTorch users,https://github.com/pytorch/pytorch/issues/189098,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-07-06T21:38:18Z +issue,188553,test_cholesky_solve_batched_many_batches_cuda_float64 fails on ROCm gfx942 (Tensor-likes are not close),https://github.com/pytorch/pytorch/issues/188553,needs triage,65,"triage signal present; component/oncall labels: module: linear algebra, module: rocm",categorize and decide next maintainer owner/action,2026-07-06T17:17:16Z +issue,144683,torch.export treats two of the same parameters as the same node,https://github.com/pytorch/pytorch/issues/144683,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-07-06T04:36:19Z +issue,188567,"torch.distributed._symmetric_memory: NCCL-backend rendezvous() fails (""Failed to window register / invalid argument"") when it is the first NCCL op on the process group",https://github.com/pytorch/pytorch/issues/188567,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-30T18:41:40Z +issue,188490,test_profiler_stack_trace_augmentation fails on Windows due to stale profiler golden,https://github.com/pytorch/pytorch/issues/188490,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-06-29T22:35:00Z +issue,188481,export_chrome_trace produces invalid timestamps,https://github.com/pytorch/pytorch/issues/188481,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-06-29T21:05:13Z +issue,188227,[inductor] generated kernel calls `empty_strided_cpu` with a negative stride → `Expected result >= 0`,https://github.com/pytorch/pytorch/issues/188227,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2026-06-28T13:48:56Z +issue,172568,Control flow operators issue tracker,https://github.com/pytorch/pytorch/issues/172568,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-06-27T08:39:48Z +issue,188304,Update NVSHMEM to Resolve RoCE Failures,https://github.com/pytorch/pytorch/issues/188304,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-26T22:26:32Z +issue,188266,SIGABRT inside `gloo::transport::tcp::Address::fromPeerName(int)`,https://github.com/pytorch/pytorch/issues/188266,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-26T10:23:22Z +issue,185268,DISABLED test_correctness_with_reusing_ir_ne_cpu_float32 (__main__.TestLazyOpInfoCPU),https://github.com/pytorch/pytorch/issues/185268,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: jit",categorize and decide next maintainer owner/action,2026-06-26T01:29:50Z +issue,185267,DISABLED test_correctness_ne_cpu_float32 (__main__.TestLazyOpInfoCPU),https://github.com/pytorch/pytorch/issues/185267,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: jit",categorize and decide next maintainer owner/action,2026-06-26T01:29:49Z +issue,188133,Pallas CPU mishandles synthetic-base mutated aliased inputs,https://github.com/pytorch/pytorch/issues/188133,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2026-06-25T05:43:56Z +issue,181086,[Symmetric Memory] symm buffer init and rendezvous cannot be assigned to a specific CUDA stream,https://github.com/pytorch/pytorch/issues/181086,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-19T11:10:46Z +issue,52741,torch.jit.script is swallowing a rfloordiv warning,https://github.com/pytorch/pytorch/issues/52741,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-06-19T10:24:32Z +issue,182373,"[ROCm] torch.profiler drops GPU activity in spawn child if parent calls torch.cuda.init() (regression in 2.12, 2.10/2.11 pass)",https://github.com/pytorch/pytorch/issues/182373,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: rocm, oncall: profiler",categorize and decide next maintainer owner/action,2026-06-18T16:12:02Z +issue,187659,Segmentation fault in `TestQuantizedOps.test_empty_batch` on ARM,https://github.com/pytorch/pytorch/issues/187659,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-06-18T14:07:03Z +issue,177254,AArch64 Unit Test Failure - test_aot_inductor_arrayref - RuntimeError: unknown architecure,https://github.com/pytorch/pytorch/issues/177254,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: arm, oncall: quantization",categorize and decide next maintainer owner/action,2026-06-18T10:11:47Z +issue,186363,test_quantization.py int8 tests fail on AArch64,https://github.com/pytorch/pytorch/issues/186363,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-06-18T10:10:52Z +issue,187432,TypeError: can only join an iterable within `default_planner.py`,https://github.com/pytorch/pytorch/issues/187432,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",categorize and decide next maintainer owner/action,2026-06-16T11:51:08Z +issue,186591,Importing Pytorch results in undefined symbol ncclCommResume,https://github.com/pytorch/pytorch/issues/186591,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-16T05:35:09Z +issue,184147,[AMD] MoRI support for MoE EP Combine/Dispatch torch.distributed.TokenSwitch Interface & AMD Pollara Support,https://github.com/pytorch/pytorch/issues/184147,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-15T23:24:39Z +issue,187094,torch._int_mm dispatches Ampere-era cutlass_80 kernels on sm_103 (GB300): 6.4-13.4x SLOWER than the fp32 tf32 matmul of the same shape,https://github.com/pytorch/pytorch/issues/187094,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-06-14T13:45:25Z +issue,173815,[RFC] Abort a DeviceMesh,https://github.com/pytorch/pytorch/issues/173815,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: devicemesh, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-13T16:39:14Z +issue,176439,[wip][RFC] Symmetric Memory pool explicit lending semantics,https://github.com/pytorch/pytorch/issues/176439,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-13T16:39:11Z +issue,167966,Symmtric_memory: torch.ops.symm_mem.all_to_all_vdev_2d restricts the number of experts per rank (ne) to less than 16,https://github.com/pytorch/pytorch/issues/167966,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-13T16:38:39Z +issue,172873,Need for all_reduce to initialize NCCL communicator,https://github.com/pytorch/pytorch/issues/172873,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, module: symm_mem, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-13T16:38:22Z +issue,172617,Symmetric utils missing,https://github.com/pytorch/pytorch/issues/172617,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-13T16:38:18Z +issue,167975,"Multi-node distributed training hangs indefinitely after NCCL initialization completes, specifically blocking at `verify_param_shape_across_processes` in `torch.distributed.utils`.",https://github.com/pytorch/pytorch/issues/167975,needs triage,65,"triage signal present; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-13T16:34:44Z +issue,172348,Build failure in NCCLSymmetricMemory.cu: undefined identifier `offset_`,https://github.com/pytorch/pytorch/issues/172348,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: build, module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-12T17:12:35Z +issue,170707,Update Rendezvous etcd backend to use etcd v3 API,https://github.com/pytorch/pytorch/issues/170707,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, module: elastic, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-12T17:11:17Z +issue,172218,`_expand_group` can get DeviceMesh ranks more directly,https://github.com/pytorch/pytorch/issues/172218,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: devicemesh, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-12T17:11:16Z +issue,170045,ncclUnhandledCudaError: Call to CUDA function failed. Last error: Cuda failure 999 'unknown error',https://github.com/pytorch/pytorch/issues/170045,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-12T17:10:48Z +issue,169448,FSDP2 model saving broken unless ignoring frozen params,https://github.com/pytorch/pytorch/issues/169448,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed checkpointing",categorize and decide next maintainer owner/action,2026-06-12T17:09:50Z +issue,161147,RuntimeError in DTensor evenly shard view op,https://github.com/pytorch/pytorch/issues/161147,needs triage,65,"triage signal present; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-12T17:04:15Z +issue,163193,Incorrect results when adding partial DTensor with a scalar,https://github.com/pytorch/pytorch/issues/163193,needs triage,65,"triage signal present; component/oncall labels: module: correctness (silent), module: dtensor, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-12T17:04:06Z +issue,186026,Annotate Cuda Graph kernel nodes from profiler with the originating fully qualified layer names,https://github.com/pytorch/pytorch/issues/186026,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-06-11T21:36:50Z +issue,167304,RPC cannot run in jetson orin because of the specific uuid of orin,https://github.com/pytorch/pytorch/issues/167304,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: cuda, module: rpc, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-11T17:24:35Z +issue,166577,Pipelining - cannot calculate dWeight if a custom torch.autograd.Function is used with a schedule that performs separate dWeight and dInput backward passes,https://github.com/pytorch/pytorch/issues/166577,needs triage,65,"triage signal present; component/oncall labels: module: autograd, module: pipelining, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-11T17:24:09Z +issue,155924,"Activation Checkpointing breaks ""torch.distributed.checkpoint.state_dict._get_fqns""",https://github.com/pytorch/pytorch/issues/155924,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: activation checkpointing, oncall: distributed, oncall: distributed checkpointing",categorize and decide next maintainer owner/action,2026-06-10T17:14:28Z +issue,186823,DISABLED test_cupti_monitor_trace_has_expected_events (__main__.TestProfilerCUDA),https://github.com/pytorch/pytorch/issues/186823,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-06-10T01:29:42Z +issue,182988,ROCm MI300 (gfx942): native segfault in mixed-precision BatchNorm2d backward,https://github.com/pytorch/pytorch/issues/182988,needs triage,65,"triage signal present; component/oncall labels: module: fsdp, module: rocm, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-09T17:29:37Z +issue,153779,NCCL timeout while training on multi-gpu,https://github.com/pytorch/pytorch/issues/153779,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-09T17:04:13Z +issue,130251,[funcol] numerical error when funcol.reduce-scatter a non-contiguous tensor,https://github.com/pytorch/pytorch/issues/130251,needs triage,65,"triage signal present; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-09T16:58:51Z +issue,140183,FSDP Hybrid Shard worse loss than than Full Shard,https://github.com/pytorch/pytorch/issues/140183,needs triage,65,"triage signal present; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-09T16:58:45Z +issue,186773,DISABLED test_cupti_monitor_record_shapes (__main__.TestProfilerCUDA),https://github.com/pytorch/pytorch/issues/186773,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-06-09T13:33:41Z +issue,186670,_load_mobile_module_from_bytes SIGSEGVs on malformed flatbuffer model (unchecked Function.class_type index),https://github.com/pytorch/pytorch/issues/186670,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2026-06-08T23:59:37Z +issue,186637,DISABLED test_cupti_monitor_multithread_runtime_thread_assignment (__main__.TestProfilerCUDA),https://github.com/pytorch/pytorch/issues/186637,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-06-08T19:28:23Z +issue,143574,Torch elastic restart fails with torch-2.6.0 nightly build: NCCL unhandled system error,https://github.com/pytorch/pytorch/issues/143574,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-08T17:17:44Z +issue,137016,[RFC] Accelerate intra-node communication via NCCL buffer registration,https://github.com/pytorch/pytorch/issues/137016,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-08T17:16:28Z +issue,143593,TCPStore crash when initializing from multiple threads,https://github.com/pytorch/pytorch/issues/143593,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-08T17:16:05Z +issue,137007,[RFC] ProcessGroupNCCL uses non-blocking API by default,https://github.com/pytorch/pytorch/issues/137007,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-08T17:14:29Z +issue,130113,ncclUnhandledCudaError: Call to CUDA function failed.,https://github.com/pytorch/pytorch/issues/130113,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: ddp, module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-08T17:14:11Z +issue,129341,NCCL Blocking Send/Recv are Non-blocking in practice,https://github.com/pytorch/pytorch/issues/129341,needs triage,65,"triage signal present; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-08T17:14:02Z +issue,186587,DISABLED test_cupti_monitor_matches_stock_op_and_kernel_names (__main__.TestProfilerCUDA),https://github.com/pytorch/pytorch/issues/186587,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-06-08T14:01:08Z +issue,115388,`torch.distributed.destroy_process_group()` hangs after CUDA graph capture of NCCL operations,https://github.com/pytorch/pytorch/issues/115388,needs triage,65,"triage signal present; component/oncall labels: module: c10d, module: cuda graphs, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-07T16:38:36Z +issue,114396,is PGNCCL abortCommsFromMap working correctly?,https://github.com/pytorch/pytorch/issues/114396,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-07T16:37:33Z +issue,106633,Refcount problem for torch.distributed.Store objects defined in Python,https://github.com/pytorch/pytorch/issues/106633,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-07T16:37:12Z +issue,106711,'CUDA out of memory' when using a GPU services for reinforcement learning in Torch rpc tutorial,https://github.com/pytorch/pytorch/issues/106711,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: rpc, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-07T16:36:53Z +issue,109941,Need latest NCCL support to reduce GPU HBM consumption,https://github.com/pytorch/pytorch/issues/109941,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-07T16:36:40Z +issue,85671,nn.Embedding weights are not synced across processes with DistributedDataParallel when other parameters are present,https://github.com/pytorch/pytorch/issues/85671,needs triage,65,"triage signal present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-07T16:33:46Z +issue,102596,Jetson NX with torch 1.12.0 :cannot import name 'ProcessGroup' from 'torch.distributed'.,https://github.com/pytorch/pytorch/issues/102596,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:36:35Z +issue,103221,[help] did torch.distributed.launch can be applied on k8s cluster with pytorch-operator,https://github.com/pytorch/pytorch/issues/103221,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:36:24Z +issue,101962,SummaryWriter background thread holds the GIL for too long,https://github.com/pytorch/pytorch/issues/101962,needs triage,65,"triage signal present; component/oncall labels: module: tensorboard, oncall: distributed",categorize and decide next maintainer owner/action,2026-06-06T16:35:54Z +issue,100310,Using ddp training with different machine,https://github.com/pytorch/pytorch/issues/100310,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:35:38Z +issue,91173,not able to import pipelines as torch.distributed is missing,https://github.com/pytorch/pytorch/issues/91173,needs triage,65,triage signal present; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2026-06-06T16:35:12Z +issue,86962,PyTorch RPC crashed when using IB,https://github.com/pytorch/pytorch/issues/86962,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: rpc, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:35:02Z +issue,86849,`torch.distributed.all_reduce` allocates excess GPU memory when using NCCL backend,https://github.com/pytorch/pytorch/issues/86849,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:34:44Z +issue,74590,[CTA] Let's Stamp Out Flaky Tests!,https://github.com/pytorch/pytorch/issues/74590,needs triage,65,"triage signal present; component/oncall labels: module: autograd, module: cuda, module: data parallel, module: dataloader",categorize and decide next maintainer owner/action,2026-06-06T16:34:00Z +issue,83824,RuntimeError: Interrupted system call when doing distributed training,https://github.com/pytorch/pytorch/issues/83824,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:33:27Z +issue,85234,c10d all_gather aborts with Signal 8 (SIGFPE) when tensor.numel() == 0,https://github.com/pytorch/pytorch/issues/85234,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:33:19Z +issue,82306,when distribute training load pretrain model error,https://github.com/pytorch/pytorch/issues/82306,needs triage,65,"triage signal present; component/oncall labels: module: serialization, oncall: distributed",categorize and decide next maintainer owner/action,2026-06-06T16:33:02Z +issue,76039,[RFC] NPU device for PyTorch,https://github.com/pytorch/pytorch/issues/76039,needs triage,65,triage signal present; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2026-06-06T16:32:36Z +issue,71049,Distributed broadcast fails with simple GPU tensor on Windows + GLOO,https://github.com/pytorch/pytorch/issues/71049,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:32:08Z +issue,73542,Investigate why tensor shapes are not populated when printing WorkNCCL for broadcast,https://github.com/pytorch/pytorch/issues/73542,needs triage,65,"triage signal present; component/oncall labels: module: c10d, module: ddp, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:29:40Z +issue,74809,Incorrect results for `torch.distributed.gather` for tensor created from permuted NumPy array,https://github.com/pytorch/pytorch/issues/74809,needs triage,65,"triage signal present; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-06T16:29:06Z +issue,76282,`torch.cuda.amp.GradScaler` may skip parameter synchronization required by post localSGD optimizer,https://github.com/pytorch/pytorch/issues/76282,needs triage,65,"triage signal present; component/oncall labels: module: amp (automated mixed precision), oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-06T16:28:23Z +issue,186400,DISABLED test_cupti_monitor_collection_repeated_lifecycle (__main__.TestProfilerCUDA),https://github.com/pytorch/pytorch/issues/186400,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-06-05T19:15:42Z +issue,82308,torch.einsum gets wrong results randomly when training with multi-gpu,https://github.com/pytorch/pytorch/issues/82308,needs triage,65,"triage signal present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:11:22Z +issue,82464,SyncBatchNorm does not work on CPU,https://github.com/pytorch/pytorch/issues/82464,needs triage,65,"triage signal present; component/oncall labels: module: nn, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:11:17Z +issue,87995,multi-node distributed training rank0 hang at dataloader after a few epochs,https://github.com/pytorch/pytorch/issues/87995,needs triage,65,"triage signal present; component/oncall labels: module: dataloader, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:11:01Z +issue,91535,"When dist.broadcast float32 to int64, it will silently generate wrong results",https://github.com/pytorch/pytorch/issues/91535,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-05T17:10:02Z +issue,102186,torch/distributed/_spmd/api.py should aot_module_export instead of make_fx directly,https://github.com/pytorch/pytorch/issues/102186,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:09:11Z +issue,106690,DDP grads not synced when static_graph=True and module output is a dict subclass?,https://github.com/pytorch/pytorch/issues/106690,needs triage,65,"triage signal present; component/oncall labels: module: ddp, module: pytree, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:08:16Z +issue,111138,Module states cannot be fully synchronized due to the DDP broadcast_buffers breaking change,https://github.com/pytorch/pytorch/issues/111138,needs triage,65,"triage signal present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:07:32Z +issue,112152,DDP backpropagated gradients not the same across all gpus when forward inference not using all declared modules,https://github.com/pytorch/pytorch/issues/112152,needs triage,65,"triage signal present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:07:07Z +issue,123983,Segfault in TCPStore and FileStore compare_set(),https://github.com/pytorch/pytorch/issues/123983,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-05T17:04:50Z +issue,124666,"When using DDP, model outputs that are are custom dataclasses would make things silently wrong",https://github.com/pytorch/pytorch/issues/124666,needs triage,65,"triage signal present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:04:00Z +issue,130345,Provide shutdown() method for ProcessGroupGloo distributed backend,https://github.com/pytorch/pytorch/issues/130345,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-05T17:03:45Z +issue,130515,[Bug Report] inconsistent global src and local rank in dist.nn.functional.broadcast,https://github.com/pytorch/pytorch/issues/130515,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-05T17:03:40Z +issue,130958,Document ProcessGroupOptions,https://github.com/pytorch/pytorch/issues/130958,needs triage,65,"triage signal present; component/oncall labels: module: docs, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-05T17:03:35Z +issue,143536,FSDP learning hangs when the program tries to save the model,https://github.com/pytorch/pytorch/issues/143536,needs triage,65,"triage signal present; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:01:25Z +issue,144359,Incorrect Results with Tensor Parallelism,https://github.com/pytorch/pytorch/issues/144359,needs triage,65,"triage signal present; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:01:16Z +issue,146331,DeepSeek: hierarchical a2a,https://github.com/pytorch/pytorch/issues/146331,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:00:33Z +issue,146907,"INTERNAL ASSERT FAILED at ""/pytorch/torch/csrc/distributed/c10d/reducer.cpp"":2134, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/146907,needs triage,65,"triage signal present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T17:00:09Z +issue,165727,DDP hangs at first AllGather during parameter broadcast with tied weights on Blackwell GPUs (sm_120),https://github.com/pytorch/pytorch/issues/165727,needs triage,65,"triage signal present; component/oncall labels: module: ddp, module: nccl, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-06-05T16:54:20Z +issue,166301,Improve Docs: Reproducibility and Randomness in dist training with multi-process data loading and explicit RNG seed,https://github.com/pytorch/pytorch/issues/166301,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T16:54:07Z +issue,169970,Does torch._grouped.mm work with cudagraphs over multiple nodes?,https://github.com/pytorch/pytorch/issues/169970,needs triage,65,"triage signal present; component/oncall labels: module: cuda, module: cuda graphs, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-06-05T16:52:53Z +issue,186344,DISABLED test_cupti_monitor_collection_raw_dump_smoke (__main__.TestProfilerCUDA),https://github.com/pytorch/pytorch/issues/186344,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: flaky-tests, oncall: profiler",categorize and decide next maintainer owner/action,2026-06-05T13:39:39Z +issue,153705,Unable to export a model using scan with inplace modification,https://github.com/pytorch/pytorch/issues/153705,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-06-04T20:08:38Z +issue,151654,[sparse][exported-graph] Wrong Exported Graph for the Sparse Tensor,https://github.com/pytorch/pytorch/issues/151654,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-06-04T20:08:31Z +issue,185443,[JIT] Uncontrolled memory allocation in Pickle Unpickler opcode handlers (setInput / readBytes),https://github.com/pytorch/pytorch/issues/185443,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-06-04T17:36:27Z +issue,184647,Replacing a submodule's `__call__` with `fx.Interpreter` on its exported `gm` causes the parent's output to contain `FakeTensor`,https://github.com/pytorch/pytorch/issues/184647,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-06-04T00:50:16Z +issue,121471,[profiler]Not passed shapes for lists of tensors to kineto_result,https://github.com/pytorch/pytorch/issues/121471,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-06-01T16:59:16Z +issue,181621,torch.compile (Inductor) generates invalid C++ kernel: tmp8 used outside its declared scope when torch.gather index is a self-referential expression,https://github.com/pytorch/pytorch/issues/181621,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: cpu, module: inductor, module: scatter & gather ops, oncall: cpu inductor",categorize and decide next maintainer owner/action,2026-06-01T03:58:14Z +issue,181638,torch.compile(backend='inductor') hangs indefinitely on LayerNorm → flatten → cumprod,https://github.com/pytorch/pytorch/issues/181638,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2026-06-01T03:56:17Z +issue,151525,[inductor] [cpu] [silent incorrectness] `nn.LazyConvTranspose2d-torch.randn-F.linear-torch.argmax` output incorrect results on CPU inductor,https://github.com/pytorch/pytorch/issues/151525,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2026-05-30T01:11:09Z +issue,151726,Exported Module cannot call train() or eval(),https://github.com/pytorch/pytorch/issues/151726,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-05-29T16:37:47Z +issue,185442,[JIT] Use-after-free in Source::_M_dispose() during TorchScript error-path cleanup,https://github.com/pytorch/pytorch/issues/185442,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-05-27T22:43:56Z +issue,185440,[JIT] Out-of-bounds read in MemoryReadAdapter::read() via crafted TorchScript model,https://github.com/pytorch/pytorch/issues/185440,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-05-27T22:42:07Z +issue,184010,[Feature Request] Support PrivateUse1 device in ProcessGroupGloo,https://github.com/pytorch/pytorch/issues/184010,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-27T20:29:52Z +issue,183913,[RFC] Unify Python dependency declarations under pyproject.toml + uv.lock,https://github.com/pytorch/pytorch/issues/183913,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: ci, oncall: releng",categorize and decide next maintainer owner/action,2026-05-27T14:33:16Z +issue,172561,[export] Constraint expression suggested by torch export is not supported,https://github.com/pytorch/pytorch/issues/172561,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: onnx, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-05-25T20:30:42Z +issue,184982,"deprecation of derived quantized dtypes (quint8, qint8, qint32)",https://github.com/pytorch/pytorch/issues/184982,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: deprecation, oncall: quantization",categorize and decide next maintainer owner/action,2026-05-25T17:33:27Z +issue,152532,The 2.7.0 release tarball is missing `.ci/docker/ci_commit_pins/nccl-cu12.txt` required for building,https://github.com/pytorch/pytorch/issues/152532,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2026-05-23T23:24:23Z +issue,181429,"Make torchrun options ( --nnodes, --nproc-per-node, etc.) same a mpirun/mpiexec",https://github.com/pytorch/pytorch/issues/181429,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-23T16:28:35Z +issue,180550,[Discussion] Parallelizing serial for-loops with free-threaded Python — and the `torch.distributed` challenge,https://github.com/pytorch/pytorch/issues/180550,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-23T16:27:54Z +issue,178138,Conflict between PyTorch CUDACachingAllocator and NCCL symmetric memory (ncclCommWindowRegister) causing incorrect all-reduce results,https://github.com/pytorch/pytorch/issues/178138,needs triage,65,"triage signal present; component/oncall labels: module: nccl, module: symm_mem, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-23T16:24:32Z +issue,179848,Race condition updating pgStatus_ in ProcessGroupGloo::runLoop,https://github.com/pytorch/pytorch/issues/179848,needs triage,65,"triage signal present; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-23T16:24:21Z +issue,179238,Shutdown race in Gloo backend,https://github.com/pytorch/pytorch/issues/179238,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-22T16:51:58Z +issue,178756,Case mismatch in `_new_process_group_helper` prevents default backend type from being set for custom out-of-tree backends,https://github.com/pytorch/pytorch/issues/178756,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-22T16:51:57Z +issue,178191,[distributed] ProcessGroupMPI: cudaAwareMpiCheck() does not support Cray MPICH GPU-aware MPI,https://github.com/pytorch/pytorch/issues/178191,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-22T16:51:13Z +pr,180501,Fix clip_grad_norm_ failure with mixed DeviceMesh DTensor norms,https://github.com/pytorch/pytorch/pull/180501,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/dtensor,"identify whether block is CI, merge conflict, or review gate",2026-05-22T14:26:14Z +pr,184587,Fix torch.compile addmm failure with beta=0 nonbroadcastable bias,https://github.com/pytorch/pytorch/pull/184587,PR blocked,65,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-22T07:46:30Z +issue,179770,[symm_mem] small-M FP8 fused_scaled_matmul_reduce_scatter is slower than unfused baseline on B300/SM100,https://github.com/pytorch/pytorch/issues/179770,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-21T17:04:03Z +issue,177371,LocalTensorMode: all_to_all_single crashes with uneven split sizes,https://github.com/pytorch/pytorch/issues/177371,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-21T17:00:50Z +issue,177815,[distributed][Dynamo] Add get_process_group_all_ranks() for SPMD compilation backends,https://github.com/pytorch/pytorch/issues/177815,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, module: devicemesh, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-21T17:00:43Z +issue,177427,[RFC] Communication primitives for non-contiguous tensors,https://github.com/pytorch/pytorch/issues/177427,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, module: fsdp, module: nccl, module: pipelining",categorize and decide next maintainer owner/action,2026-05-21T17:00:43Z +issue,177309,[ROCm][NCCL watchdog] Cross-thread stream-capture mode restrictions in hipEventQuery/hipEventSynchronize cause false watchdog failures,https://github.com/pytorch/pytorch/issues/177309,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: rocm, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-21T17:00:00Z +issue,177059,Way to have DTensor ops go through torch_function rather than go directly through torch_dispatch?,https://github.com/pytorch/pytorch/issues/177059,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-21T16:59:36Z +issue,175145,[torch.disitributed] Persistent Deadlock when overlapping NCCL P2P and Collectives from separate Python Threads,https://github.com/pytorch/pytorch/issues/175145,needs triage,65,"triage signal present; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-21T16:57:40Z +pr,183922,Unskip test_vector_norm on AArch64,https://github.com/pytorch/pytorch/pull/183922,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-21T08:20:29Z +pr,181060,[fft] Validate _fft_c2r last_dim_size to fix heap-buffer overflow,https://github.com/pytorch/pytorch/pull/181060,PR blocked,65,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/mps, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-21T05:51:18Z +issue,175469,[DTensor] ExportedProgram.run_decompositions() fails with AssertionError: out is not NotImplemented,https://github.com/pytorch/pytorch/issues/175469,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms, oncall: export",categorize and decide next maintainer owner/action,2026-05-20T17:18:22Z +issue,176877,[RFC] Adding Distributed Support to OpenReg,https://github.com/pytorch/pytorch/issues/176877,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: openreg, module: privateuse1, oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-20T17:15:23Z +issue,174669,`TypeError: cannot pickle code objects` with Python 3.13,https://github.com/pytorch/pytorch/issues/174669,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",categorize and decide next maintainer owner/action,2026-05-20T17:15:07Z +issue,175724,[Distributed][NCCL] Request for Backport: SIGSEGV in HeartbeatMonitor due to thread-unsafe getenv/setenv race (#167523),https://github.com/pytorch/pytorch/issues/175724,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-20T17:15:06Z +issue,176276,Path Traversal in torch.distributed.FileStore allows writing files outside intended directory,https://github.com/pytorch/pytorch/issues/176276,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: distributed, oncall: distributed infra",categorize and decide next maintainer owner/action,2026-05-20T17:14:44Z +issue,173680,Large CPU overhead with FSDP2,https://github.com/pytorch/pytorch/issues/173680,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: fsdp, module: performance, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-20T17:14:06Z +issue,174960,Feature Request: Activation offloading with async prefetch in FSDP,https://github.com/pytorch/pytorch/issues/174960,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-20T17:13:21Z +issue,174276,[DTensor] Avoid forced redistribution in layer_norm strategy,https://github.com/pytorch/pytorch/issues/174276,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-20T17:12:45Z +pr,184151,[CUDA] Fix float32 std/var M2 overflow by using double accumulation,https://github.com/pytorch/pytorch/pull/184151,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-20T15:57:32Z +pr,180602,Enable ruff rule PLW1510: require explicit check= in subprocess calls,https://github.com/pytorch/pytorch/pull/180602,PR blocked,65,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-20T15:55:18Z +issue,183888,torch.compile Inductor fails on unfold followed by in-place abs_,https://github.com/pytorch/pytorch/issues/183888,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2026-05-20T11:50:23Z +pr,182055,[Inductor] Fix flaky test_allow_reuse_disable_if_exceed_peak by checking reuse marker,https://github.com/pytorch/pytorch/pull/182055,PR blocked,65,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-20T07:44:18Z +issue,180995,ConvTranspose + BatchNorm fusion broken for `groups != 1`,https://github.com/pytorch/pytorch/issues/180995,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-05-19T19:34:38Z +issue,182641,[v2.11.0] Cannot jvp through FSDP2,https://github.com/pytorch/pytorch/issues/182641,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-19T17:21:39Z +issue,72482,Improve torch::deploy documentation,https://github.com/pytorch/pytorch/issues/72482,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: deploy, oncall: package/deploy",categorize and decide next maintainer owner/action,2026-05-19T15:26:27Z +issue,183702,`grouped fully_shard(list) fails when a grouped module is reused later in the same iteration`,https://github.com/pytorch/pytorch/issues/183702,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-18T19:46:44Z +issue,184078,add_params metric_dict values not shown in tensorboard hparams tab,https://github.com/pytorch/pytorch/issues/184078,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2026-05-17T07:20:32Z +pr,173459,Add dtype mismatch validation for mm/bmm in meta registrations (#171190),https://github.com/pytorch/pytorch/pull/173459,PR blocked,65,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-16T08:53:35Z +pr,180118,[dynamo] Skip pad_packed_sequence during tracing,https://github.com/pytorch/pytorch/pull/180118,PR blocked,65,"failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-16T04:18:07Z +issue,183842,docs.pytorch.org/memory_viz has Unhandled version 5 error,https://github.com/pytorch/pytorch/issues/183842,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-05-15T03:50:31Z +pr,183087,[profiler] Synchronize all active devices on profiler exit,https://github.com/pytorch/pytorch/pull/183087,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-14T17:08:37Z +pr,182406,Fix misleading error when caching_allocator_alloc size overflows ssize_t,https://github.com/pytorch/pytorch/pull/182406,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-13T12:47:59Z +pr,177511,[export] preserve original traceback when augmenting export error messages,https://github.com/pytorch/pytorch/pull/177511,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-05-10T05:35:42Z +issue,182382,[RFC] [DTensor] Support for One-Sided Communication in DTensor,https://github.com/pytorch/pytorch/issues/182382,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dtensor, module: symm_mem, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-08T00:09:11Z +issue,182661,[v2.11.0] Cannot jvp through DDP,https://github.com/pytorch/pytorch/issues/182661,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",categorize and decide next maintainer owner/action,2026-05-07T17:12:08Z +issue,181181,FP8 Per-Head Quantized Scaled Dot Product Attention,https://github.com/pytorch/pytorch/issues/181181,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-05-06T17:36:53Z +issue,182653,CPU FPE crash in `torch.nn.quantized.functional.conv2d` (qnnpack),https://github.com/pytorch/pytorch/issues/182653,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-05-06T14:35:23Z +issue,125171,datettime.now() is not supported by Dynamo,https://github.com/pytorch/pytorch/issues/125171,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dynamo, oncall: export",categorize and decide next maintainer owner/action,2026-05-01T23:42:47Z +issue,181865,CUDA trace callbacks leak across tests in a shared process,https://github.com/pytorch/pytorch/issues/181865,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: cuda, oncall: profiler",categorize and decide next maintainer owner/action,2026-04-30T17:35:39Z +issue,178044,Compilation fails with Apple Clang 17 due to ambiguous py::make_tuple return types,https://github.com/pytorch/pytorch/issues/178044,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-04-29T23:50:06Z +issue,177258,AArch64 Unit Test Failure - test/jit/test_freezing.py TestFreezing.test_module_with_shared_type_instances,https://github.com/pytorch/pytorch/issues/177258,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: arm, module: tests, oncall: jit",categorize and decide next maintainer owner/action,2026-04-29T19:47:25Z +issue,173988,[DTensor] Implement decomposed sharding propagation for SDPA operations,https://github.com/pytorch/pytorch/issues/173988,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dtensor, oncall: distributed",categorize and decide next maintainer owner/action,2026-04-29T16:37:26Z +issue,173990,[DTensor] Propagate 'grad_dtype' attribute in 'to_local' and 'from_local' operations,https://github.com/pytorch/pytorch/issues/173990,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dtensor, oncall: distributed",categorize and decide next maintainer owner/action,2026-04-29T16:37:21Z +issue,174732,[DTensor][aot_eager] compiling sharded embedding with dynamic seqlen fails,https://github.com/pytorch/pytorch/issues/174732,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dtensor, oncall: distributed",categorize and decide next maintainer owner/action,2026-04-29T16:37:15Z +issue,177959,"dist.new_group(..., use_local_synchronization=True) hangs with Gloo for overlapping groups, while NCCL succeeds on the same participant-consistent creation order",https://github.com/pytorch/pytorch/issues/177959,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2026-04-29T16:36:56Z +issue,144663,Support loading and executing a ExportedProgram from torch.export in C++ environment,https://github.com/pytorch/pytorch/issues/144663,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-04-28T18:44:32Z +issue,181473,`torch.jit.ScriptModule.register_full_backward_hook` raises RuntimeError due to `_is_full_backward_hook` attribute type cast failure,https://github.com/pytorch/pytorch/issues/181473,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-04-27T01:57:51Z +issue,181480,DISABLED LiteInterpreterUpgraderTest.DivTensorV2,https://github.com/pytorch/pytorch/issues/181480,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-04-25T05:23:14Z +issue,181231,`jit.script` sequence unpacking breaks when using square brackets,https://github.com/pytorch/pytorch/issues/181231,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-04-23T08:57:25Z +issue,176288,memory profiler data dump streaming-friendly format,https://github.com/pytorch/pytorch/issues/176288,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-04-22T02:57:05Z +issue,177239,[Question] JIT Path Deprecation — Maintenance Timeline and Removal Plans,https://github.com/pytorch/pytorch/issues/177239,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-04-21T05:33:51Z +issue,180307,PackageImporter.load_pickle() race can silently deserialize the wrong tensor across concurrent loads,https://github.com/pytorch/pytorch/issues/180307,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2026-04-14T02:17:25Z +issue,153949,Profiler does not capture steps on Blackwell,https://github.com/pytorch/pytorch/issues/153949,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-04-13T17:23:00Z +issue,141668,[inductor][cpu] AMP CPP wrapper performance regression in 2024-11-25 nightly release,https://github.com/pytorch/pytorch/issues/141668,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2026-04-09T21:54:46Z +issue,167631,`jit.export` analoge for `torch.export`,https://github.com/pytorch/pytorch/issues/167631,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-04-09T20:30:46Z +issue,179111,The smallest values of E4M3 for two-level quantization is incorrect,https://github.com/pytorch/pytorch/issues/179111,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-04-09T11:46:04Z +issue,179176,Support unbacked in export Dim API.,https://github.com/pytorch/pytorch/issues/179176,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-04-02T19:50:24Z +issue,179109,Request to support fp8 x4 packed dtypes in PT,https://github.com/pytorch/pytorch/issues/179109,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-04-02T03:51:02Z +issue,177825,Internal Assert Failure in `torch.ao.nn.quantized.GroupNorm` with incorrectly sized bias tensor,https://github.com/pytorch/pytorch/issues/177825,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-03-26T00:38:43Z +issue,177508,[inductor][cpu] torch_multimodal_clip AMP multiple thread performance regression in 2026-03-16 nightly release,https://github.com/pytorch/pytorch/issues/177508,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: cpu inductor,categorize and decide next maintainer owner/action,2026-03-24T06:31:55Z +issue,178087,"[Profiler] Expose moduleHierarchy, activityType, and other KinetoEvent methods to Python",https://github.com/pytorch/pytorch/issues/178087,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-03-23T22:17:21Z +issue,177476,Crash in torch.ao.nn.quantized.functional.interpolate when input tensor has a 0-sized dimension,https://github.com/pytorch/pytorch/issues/177476,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-03-16T15:47:31Z +issue,124502,[export] Llama3 export with dynamic shapes fails with constraint violations,https://github.com/pytorch/pytorch/issues/124502,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dynamic shapes, oncall: export",categorize and decide next maintainer owner/action,2026-03-15T21:02:11Z +issue,177249,AArch64 Unit Test Failure - test_ops_jit.py test_variant_consistency_jit_linalg_lstsq_cpu_complex64,https://github.com/pytorch/pytorch/issues/177249,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: arm, oncall: jit",categorize and decide next maintainer owner/action,2026-03-12T11:35:05Z +issue,177247,AArch64 Unit Test Failure - test_jit_autocast,https://github.com/pytorch/pytorch/issues/177247,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: arm, oncall: jit",categorize and decide next maintainer owner/action,2026-03-12T11:32:47Z +issue,175708,O(N * B) loop in _C._jit_pass_lint (potentially O(N^2)),https://github.com/pytorch/pytorch/issues/175708,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-03-11T15:46:38Z +issue,175293,TL-parse does not work for export.,https://github.com/pytorch/pytorch/issues/175293,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-03-09T23:28:36Z +issue,176148,"export runtime error: Tensor specified that the operator does not mutate the argument, but this seems to be empirically wrong.",https://github.com/pytorch/pytorch/issues/176148,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-03-09T23:28:22Z +issue,124328,"PreDispatch tracing doesn't produce correct ""val"" for Sparse Fake tensors.",https://github.com/pytorch/pytorch/issues/124328,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: pt2-dispatcher, oncall: export",categorize and decide next maintainer owner/action,2026-03-09T23:28:12Z +issue,166572,[Regression] torch.export.ExportedProgram.run_decompositions Starts to Break in Torch 2.8 with Mutable Buffer,https://github.com/pytorch/pytorch/issues/166572,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-03-09T23:28:06Z +issue,174608,AOTInductor compile crashes with CUDA illegal memory access on exported SAM3 model with math SDPA,https://github.com/pytorch/pytorch/issues/174608,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-03-09T23:24:39Z +issue,175520,Question about torch._dynamo.export and nn.GRU support,https://github.com/pytorch/pytorch/issues/175520,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: dynamo, oncall: export",categorize and decide next maintainer owner/action,2026-03-02T23:38:19Z +issue,33692,JIT tracing check fails with boolean tensor modifications,https://github.com/pytorch/pytorch/issues/33692,needs triage,65,triage signal present; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-03-02T01:23:15Z +issue,175790,Only first CUDA profile of the process is correct?,https://github.com/pytorch/pytorch/issues/175790,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-02-25T21:08:29Z +issue,175493,FX GraphModule saving with TorchScript submodules,https://github.com/pytorch/pytorch/issues/175493,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: fx, oncall: jit",categorize and decide next maintainer owner/action,2026-02-23T00:49:39Z +issue,174555,Profiler chrome traces express record_function and profiler step incorrectly when using with_stack=True,https://github.com/pytorch/pytorch/issues/174555,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-02-19T14:12:26Z +issue,162878,CUDA 13.0 Segmentation Fault in test/test_numba_integration.py::TestNumbaIntegration::test_array_adaptor,https://github.com/pytorch/pytorch/issues/162878,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: ci, module: cuda, oncall: releng",categorize and decide next maintainer owner/action,2026-02-16T09:00:23Z +issue,175045,Consolidate or retire pytorch/almalinux-builder Docker images after MAGMA deprecation,https://github.com/pytorch/pytorch/issues/175045,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: binaries, module: ci, oncall: releng",categorize and decide next maintainer owner/action,2026-02-15T18:20:03Z +issue,174488,[cd][testing] Add almalinux/9-base test to CD pipeline,https://github.com/pytorch/pytorch/issues/174488,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: binaries, module: cuda, oncall: releng",categorize and decide next maintainer owner/action,2026-02-06T20:54:51Z +issue,167623,torch.export: support external weights (.pt2 format),https://github.com/pytorch/pytorch/issues/167623,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-02-05T14:52:00Z +issue,174375,[CUDA] flamegraph.pl download fails on cross-device filesystems,https://github.com/pytorch/pytorch/issues/174375,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2026-02-05T13:09:33Z +issue,166205,Enable Graph Capture & Profiler Integration for PrivateUse1 Backends,https://github.com/pytorch/pytorch/issues/166205,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: privateuse1, oncall: profiler",categorize and decide next maintainer owner/action,2026-02-05T03:07:29Z +issue,174273,using profiler on jvp function with record_shapes=True causes memory leaks.,https://github.com/pytorch/pytorch/issues/174273,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2026-02-04T10:55:27Z +issue,173546,[vllm] [release 2.10] V1 Test e2e + engine : test_draft_model_correctness[False-args0],https://github.com/pytorch/pytorch/issues/173546,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: vllm, oncall: releng",categorize and decide next maintainer owner/action,2026-02-02T19:29:22Z +issue,173492,torch.jit.script failed for inplace bitwise_and,https://github.com/pytorch/pytorch/issues/173492,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-01-27T15:55:51Z +issue,173311,Issue: Remove unnecessary restriction on num_thresholds in TensorBoard PR curve summary,https://github.com/pytorch/pytorch/issues/173311,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: tensorboard, oncall: profiler",categorize and decide next maintainer owner/action,2026-01-26T18:27:19Z +issue,146152,[torch.export] Cannot export TorchVision fasterrcnn_mobilenet_v3_large_fpn,https://github.com/pytorch/pytorch/issues/146152,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-01-23T19:06:23Z +issue,173075,UniformQuantizationObserverBase computes wrong scale and zero_point,https://github.com/pytorch/pytorch/issues/173075,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-01-22T17:25:30Z +issue,171689,"After enabling XPU Graph, I cannot collect profile data during the replay phase.",https://github.com/pytorch/pytorch/issues/171689,needs triage,65,"PyTorch issue lacks triaged label; component/oncall labels: module: xpu, oncall: profiler",categorize and decide next maintainer owner/action,2026-01-21T06:23:27Z +issue,172082,torch.ao.quantization.observer work on CPU (very slow!) due to using registered buffers on CPU,https://github.com/pytorch/pytorch/issues/172082,needs triage,65,PyTorch issue lacks triaged label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-01-13T18:01:15Z +pr,160883,Fallback to eager AC when backend=eager for TagActivationCheckpoint,https://github.com/pytorch/pytorch/pull/160883,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2025-11-11T15:21:08Z +pr,155672,Fixed NLLLoss 1D input crash with torch.compile,https://github.com/pytorch/pytorch/pull/155672,PR blocked,65,failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2025-08-17T08:53:54Z +issue,154354,integer convolution,https://github.com/pytorch/pytorch/issues/154354,needs triage,65,"PyTorch issue lacks triaged label; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, oncall: quantization",route to design/research decision before patch planning,2025-05-26T20:02:11Z +issue,61210,Jitted function cannot be pickled,https://github.com/pytorch/pytorch/issues/61210,needs triage,65,PyTorch issue lacks triaged label; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-03T19:30:59Z +issue,49961,rfc: automating the switching of inputs to the device of the params,https://github.com/pytorch/pytorch/issues/49961,needs triage,65,"PyTorch issue lacks triaged label; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: rpc, module: ux, oncall: distributed",route to design/research decision before patch planning,2022-07-06T10:49:56Z +issue,55577,Source location range issue for dictionary,https://github.com/pytorch/pytorch/issues/55577,needs triage,65,"PyTorch issue lacks triaged label; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2021-10-20T01:50:08Z +issue,51138,[TorchScript Usability] tensor printout format differs from eager mode,https://github.com/pytorch/pytorch/issues/51138,needs triage,65,PyTorch issue lacks triaged label; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-07-17T04:41:20Z +issue,45745,Native CUDA-Aware MPI support for MVAPICH2 and other MPI libraries,https://github.com/pytorch/pytorch/issues/45745,needs triage,65,"PyTorch issue lacks triaged label; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",route to design/research decision before patch planning,2020-10-08T20:26:16Z +issue,111901,Multiprocess. DataLoader worker is killed by signal: Segmentation fault.,https://github.com/pytorch/pytorch/issues/111901,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: dataloader",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-30T09:49:01Z +issue,159454,libtorch_cpu.so link libm.so issue:GLIBC_2.29 not found,https://github.com/pytorch/pytorch/issues/159454,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-29T03:50:24Z +issue,45198,DataLoader with cv2 and some numpy/cv2 import order causes workers to not work,https://github.com/pytorch/pytorch/issues/45198,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multithreading, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-29T23:25:36Z +issue,66247,Floating point exception in mkl_vml_serv_GetMinN () on a specific computer,https://github.com/pytorch/pytorch/issues/66247,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash, module: mkl",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-29T23:24:38Z +issue,31758,DataLoader: Segmentation Fault (core dumped),https://github.com/pytorch/pytorch/issues/31758,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: dataloader",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-29T23:16:25Z +issue,162343,[BUG] `Aborted (core dumped)` on RTX 5090 with Nightly Build when running TTS inference,https://github.com/pytorch/pytorch/issues/162343,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: wsl",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-17T17:20:57Z +issue,162797,Segmentation fault in mkl_blas_avx512_dgemv_t_intrinsics,https://github.com/pytorch/pytorch/issues/162797,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: intel, module: mkl",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-13T07:48:28Z +issue,161865,SEGFAULT in libomp.dylib after torch.matmul on M4 Max with MPS backend,https://github.com/pytorch/pytorch/issues/161865,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash, module: macos, module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-10T13:28:19Z +issue,160876,F.silu(torch.tensor(float('-inf'))) returns nan instead of 0 in PyTorch 2.7.0+cu128,https://github.com/pytorch/pytorch/issues/160876,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-19T19:24:32Z +issue,143780,Looking for valid compiling option for extension based on torch-2.1.0+cpu.cxx11.abi,https://github.com/pytorch/pytorch/issues/143780,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-09T16:53:07Z +issue,157704,Seg faults on macos / OSX,https://github.com/pytorch/pytorch/issues/157704,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-07T15:05:05Z +issue,156723,Segmentation fault on Apple Macbook Pro M4 24GB RAM and 1TB flash dasd during forward() with AutoModelForSequenceClassification using CPU,https://github.com/pytorch/pytorch/issues/156723,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-30T02:48:07Z +issue,156613,[CUDA 12.9 | sm_120] Build fails on RowwiseScaledMM.cu with PTXAS error (PyTorch 2.8.0a0),https://github.com/pytorch/pytorch/issues/156613,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: crash, module: cuda, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-23T16:21:46Z +issue,155693,torch typing: register_buffer -> device,https://github.com/pytorch/pytorch/issues/155693,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, module: typing",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-13T03:52:05Z +issue,154014,Floating point exception when autocast is enabled,https://github.com/pytorch/pytorch/issues/154014,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-02T17:31:31Z +issue,153937,Dynamic compilation fails with torch 2.7,https://github.com/pytorch/pytorch/issues/153937,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-23T17:27:53Z +issue,152102,Segmentation fault with # FIXME: copy.deepcopy() is not defined on nn.module,https://github.com/pytorch/pytorch/issues/152102,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-24T15:52:02Z +issue,53895,"terminate called after throwing an instance of 'c10::Error' what(): isTuple() INTERNAL ASSERT FAILED at ""/home/wenda/libtorch/include/ATen/core/ivalue_inl.h"":927, please report a bug to PyTorch. Expected Tuple but got GenericList",https://github.com/pytorch/pytorch/issues/53895,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-17T11:59:34Z +issue,149941,SEGV in static destructors when Vulkan is enabled,https://github.com/pytorch/pytorch/issues/149941,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: vulkan",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-03-25T15:08:16Z +issue,22259,using multi thread lead to gpu stuck with GPU-util 100%,https://github.com/pytorch/pytorch/issues/22259,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-02-13T07:27:42Z +issue,140230,Inconsistent Model Loading with `weights_only=True` for Custom Models,https://github.com/pytorch/pytorch/issues/140230,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-19T21:28:23Z +issue,134198,[Intel CPU] Same CPU operations on Torch 2.3+ many times slower those on Torch 2.1,https://github.com/pytorch/pytorch/issues/134198,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: intel, module: performance, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-28T02:05:32Z +issue,126963,Segmentation fault (core dumped) when using pytorch Conv layers,https://github.com/pytorch/pytorch/issues/126963,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cudnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-07T01:45:42Z +issue,129304,`python3 setup.py bdist_wheel` tries to write to /usr/local/... during build,https://github.com/pytorch/pytorch/issues/129304,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-30T05:33:01Z +issue,26165,Memory leak in multithreading environment when loading checkpoint,https://github.com/pytorch/pytorch/issues/26165,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: multiprocessing, module: multithreading",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-24T20:48:35Z +issue,122478,6470 segmentation fault (core dumped),https://github.com/pytorch/pytorch/issues/122478,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-18T15:31:20Z +issue,130708,"Converting a numpy array of size larger than 32,768 to a tensor causes a segmentation fault",https://github.com/pytorch/pytorch/issues/130708,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: macos, module: numpy",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-17T03:50:47Z +issue,127275,"Python sometimes crashes inexplicably, with ""/var/log/kernel. log"" displaying index errors",https://github.com/pytorch/pytorch/issues/127275,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-29T09:05:45Z +issue,122308,Segmentation fault about torch::nn::Conv2d in `libtorch2.0.1+cuda11.8`,https://github.com/pytorch/pytorch/issues/122308,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-24T12:33:45Z +issue,125664,Aborted (core dumped) in torch.fft.irfftn/hfftn/ihfftn with malloc(): corrupted top size,https://github.com/pytorch/pytorch/issues/125664,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: fft",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-09T03:11:03Z +issue,124656,Importing polars before torch causes a segfault,https://github.com/pytorch/pytorch/issues/124656,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-04-23T18:01:07Z +issue,122935,Pytorch 2.1 / 2.2 uses 4x more memory during data loading than Pytorch 2.4 or 1.13,https://github.com/pytorch/pytorch/issues/122935,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: memory usage, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-04-02T21:31:32Z +issue,121235,Issue with sentence_transformers pointing to torch tensors manipulations,https://github.com/pytorch/pytorch/issues/121235,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-04-01T13:53:34Z +issue,118088,Segmentation fault (SIGSEGV) in c10::detail::getNonDeterministicRandom(bool),https://github.com/pytorch/pytorch/issues/118088,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: random",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-02-08T07:54:27Z +issue,117494,Kernel crashed when importing torch,https://github.com/pytorch/pytorch/issues/117494,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-16T14:48:12Z +issue,26120,Process fails with assertion error in magma-cuda100,https://github.com/pytorch/pytorch/issues/26120,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: dependency bug, module: linear algebra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-12-27T05:43:28Z +issue,109819,ValueError: only one element tensors can be converted to Python scalars,https://github.com/pytorch/pytorch/issues/109819,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: regression,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-06T19:13:40Z +issue,91633,Segmentation fault when running torch.nn.functional.fractional_max_pool3d on torch 1.13.1,https://github.com/pytorch/pytorch/issues/91633,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-08-21T06:19:03Z +issue,105872,PyTorch 2.0.x `CUDA error: operation not supported` when `Tensor.to` a different device,https://github.com/pytorch/pytorch/issues/105872,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-31T17:14:56Z +issue,99200,Unused `import torch` followed by `cuml.NearestNeighbors` leads to nondeterministic segfault (during Python process exit?),https://github.com/pytorch/pytorch/issues/99200,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-04-21T13:31:06Z +issue,89817,"Basic math operations produce a ""floating point exception""",https://github.com/pytorch/pytorch/issues/89817,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-28T15:09:37Z +issue,96779,Segmentation fault (core dumped) during Torch finetuning (at random step),https://github.com/pytorch/pytorch/issues/96779,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-16T00:59:51Z +issue,70245,"SEGFAULT on ""import torch""",https://github.com/pytorch/pytorch/issues/70245,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-01-23T17:10:01Z +issue,66033,Crashed with `terminating with uncaught exception of type std::__1::system_error: condition_variable wait failed: Invalid argument`,https://github.com/pytorch/pytorch/issues/66033,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash, module: macos, module: multithreading",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-01-13T09:32:06Z +issue,91533,Cannot cast float64 to float32,https://github.com/pytorch/pytorch/issues/91533,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: macos, module: numpy",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-01-03T14:31:18Z +issue,89714,Segfault on torch.nn.functional.one_hot with large tensor on Python 3.9,https://github.com/pytorch/pytorch/issues/89714,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-11-29T10:26:30Z +issue,87209,"Reproducible ""CUDA error: an illegal memory access was encountered""",https://github.com/pytorch/pytorch/issues/87209,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-11-02T16:57:55Z +issue,87446,ipykernel crash importing torch after scipy in .ipynb file,https://github.com/pytorch/pytorch/issues/87446,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-10-21T14:59:30Z +issue,76543,Segfault in ~PyFunctionPreHook,https://github.com/pytorch/pytorch/issues/76543,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-08-15T21:19:49Z +issue,81532,forward program terminated from __cxa_pure_virtual,https://github.com/pytorch/pytorch/issues/81532,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpp, module: crash",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-07-25T13:32:27Z +issue,71149,Memory leak in distributions.multivariate_normal.MultivariateNormal,https://github.com/pytorch/pytorch/issues/71149,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-02-10T03:16:46Z +issue,68920,Breakes with -OO flag in script,https://github.com/pytorch/pytorch/issues/68920,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: regression,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-12-02T19:57:59Z +issue,62308,Error while running FINETUNING TORCHVISION MODELS,https://github.com/pytorch/pytorch/issues/62308,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: vision",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-08-04T07:49:55Z +issue,62331,broken pipe error,https://github.com/pytorch/pytorch/issues/62331,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-07-28T15:40:15Z +issue,60835,"Pytorch V1.1 runs OK, but Pytorch v1.5 to 1.9 run wrong",https://github.com/pytorch/pytorch/issues/60835,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: regression",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-28T13:47:35Z +issue,48555,segment fault for Image model training by four GPUs,https://github.com/pytorch/pytorch/issues/48555,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-04-06T10:04:17Z +issue,54578,cublasSgemmStridedBatched failure when calling grad of grad,https://github.com/pytorch/pytorch/issues/54578,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-26T16:24:49Z +issue,40425,Segment Fault after use cusolverDnDestroy() with torch1.5,https://github.com/pytorch/pytorch/issues/40425,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: dependency bug",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-06-29T16:14:00Z +issue,36871,Stochasticity for DistributedDataParallel on CPU but not on GPU,https://github.com/pytorch/pytorch/issues/36871,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: determinism, oncall: distributed",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-06-25T12:12:17Z +issue,35792,Python 3 build PyTorch got core dump,https://github.com/pytorch/pytorch/issues/35792,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-04-01T13:34:28Z +issue,31362,sklearn and pytorch incompatibility issue,https://github.com/pytorch/pytorch/issues/31362,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-02-25T08:42:27Z +issue,28319,Seg-fault in LayerNormKernelImpl,https://github.com/pytorch/pytorch/issues/28319,high-priority/blocker,64,"high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-12-13T15:22:28Z +issue,29498,"""malloc(): memory corruption (fast)"", action=3) at malloc.c",https://github.com/pytorch/pytorch/issues/29498,high-priority/blocker,64,high-priority/blocker/regression label present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-11-18T18:19:19Z +pr,189024,[Dynamo] Support exception attribute access (StopIteration/AttributeError/NameError),https://github.com/pytorch/pytorch/pull/189024,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T02:17:19Z +pr,189318,Bump pip from 26.0.1 to 26.1.2 in /.ci/docker,https://github.com/pytorch/pytorch/pull/189318,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-07-09T02:16:34Z +pr,188597,[ROCm] Add initial support for gfx1250,https://github.com/pytorch/pytorch/pull/188597,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/binaries, ciflow/docker, ciflow/h100","identify whether block is CI, merge conflict, or review gate",2026-07-09T02:14:29Z +pr,189173,torchfuzz: detach batch_norm running stats,https://github.com/pytorch/pytorch/pull/189173,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-09T02:13:41Z +pr,189111,[torchcomms hash update] update the pinned torchcomms hash,https://github.com/pytorch/pytorch/pull/189111,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T02:13:38Z +pr,184218,Support decorator/curried calling convention for torch.utils.checkpoint.checkpoint,https://github.com/pytorch/pytorch/pull/184218,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T02:13:23Z +pr,189328,Add retry with backoff to download_mnist.py,https://github.com/pytorch/pytorch/pull/189328,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-09T02:05:04Z +pr,181680,[triton hash update] update the pinned triton hash,https://github.com/pytorch/pytorch/pull/181680,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:54:30Z +pr,189217,[vision hash update] update the pinned vision hash,https://github.com/pytorch/pytorch/pull/189217,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:54:24Z +pr,187288,[vllm hash update] update the pinned vllm hash,https://github.com/pytorch/pytorch/pull/187288,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:54:21Z +pr,188262,[CI] fix OpenBLAS build for non-AArch64,https://github.com/pytorch/pytorch/pull/188262,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:54:09Z +pr,179651,[torchtitan hash update] update the pinned torchtitan hash,https://github.com/pytorch/pytorch/pull/179651,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:54:03Z +pr,189083,[profiler][cupti] Drive periodic cuptiActivityFlushAll from the decode thread,https://github.com/pytorch/pytorch/pull/189083,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:56Z +pr,189185,[profiler][cupti] Make CuptiMonitor a singleton configured via a global configure(),https://github.com/pytorch/pytorch/pull/189185,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:53Z +pr,188621,"[profiler][cupti] Generate CUPTI field-id catalogs from the ABI as Field(id, ctype)",https://github.com/pytorch/pytorch/pull/188621,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:52Z +pr,188100,Support callable msg lambda in TestCase.assert* methods beyond assertEqual(),https://github.com/pytorch/pytorch/pull/188100,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:44Z +pr,115316,Automated submodule update: FBGEMM,https://github.com/pytorch/pytorch/pull/115316,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:38Z +pr,189175,torchfuzz: fix clamp_min/max emitting None bound,https://github.com/pytorch/pytorch/pull/189175,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:33Z +pr,188299,Lazily build error messages for assert*() calls across tests to reduce overhead,https://github.com/pytorch/pytorch/pull/188299,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/h100-symm-mem","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:28Z +pr,189192,[MPS] Add `geqrf` and refactor `linalg_qr`,https://github.com/pytorch/pytorch/pull/189192,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/mps","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:27Z +pr,188573,Cleanup dead code across tests,https://github.com/pytorch/pytorch/pull/188573,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:25Z +pr,189264,"[ROCm][CI] Make rocm-nightly workflow fully mi300-specific (name, ciflow tag, filename)",https://github.com/pytorch/pytorch/pull/189264,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:23Z +pr,187465,[Symm_mem] Support expandable segements inside PyTorch Symmetric memory for NCCL backend,https://github.com/pytorch/pytorch/pull/187465,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:15Z +pr,189302,[ROCm][CI] Make ROCM_SO_FILES a single shared source of truth for wheel bundling + preload,https://github.com/pytorch/pytorch/pull/189302,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:12Z +pr,189314,flex_gemm: recognize 2-D block local-reduce contracts,https://github.com/pytorch/pytorch/pull/189314,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:11Z +pr,189315,flex_gemm: store 2-D block local reductions,https://github.com/pytorch/pytorch/pull/189315,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:10Z +pr,187699,Add a debug log when we skip sync_decision_cross_ranks (#187699),https://github.com/pytorch/pytorch/pull/187699,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:09Z +pr,189316,flex_gemm: support mx block-scale stores,https://github.com/pytorch/pytorch/pull/189316,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:08Z +pr,189266,[ROCm][CI] Fix py3.15 ROCm manywheel numpy source build (pkg-config python3),https://github.com/pytorch/pytorch/pull/189266,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:02Z +pr,189291,Add an accurate Metal erfc for MPS (eager + inductor),https://github.com/pytorch/pytorch/pull/189291,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:01Z +pr,188862,[inductor] Fix bool minimum/maximum crash in index propagation (#188862) (#188862),https://github.com/pytorch/pytorch/pull/188862,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:01Z +pr,188643,[AOTI] Add external constants API for zero-copy weight sharing (#188643),https://github.com/pytorch/pytorch/pull/188643,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:52:59Z +pr,181726,[xpu][1/4]Implement scaled_mm_v2 for MXFP8/MXFP4/NVFP4 on XPU,https://github.com/pytorch/pytorch/pull/181726,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:51:44Z +pr,189051,[dynamo] Reject keyword args to set()/frozenset() constructors,https://github.com/pytorch/pytorch/pull/189051,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:51:36Z +pr,189325,[Dynamo] Fix self-referential bw_compiler on backend re-entry,https://github.com/pytorch/pytorch/pull/189325,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:32:21Z +pr,179286,Use C++20 concepts where it improves readability,https://github.com/pytorch/pytorch/pull/179286,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:05:38Z +pr,189249,[Dynamo] Remove dead code in tvm and common backends,https://github.com/pytorch/pytorch/pull/189249,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:04:57Z +pr,189311,[wip] Reduce hackery for Dynamo appeasement,https://github.com/pytorch/pytorch/pull/189311,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:04:10Z +pr,186055,Add CUDA graph support for torch.while_loop,https://github.com/pytorch/pytorch/pull/186055,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:04:09Z +pr,189088,[SymmMem] Place the signal pad at the front of symmetric memory allocations,https://github.com/pytorch/pytorch/pull/189088,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200-symm-mem, ciflow/h100-symm-mem","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:04:08Z +pr,189305,"Reapply ""Remove CUDA sync in torch.combinations to improve `torch.compile` support (#186595)""",https://github.com/pytorch/pytorch/pull/189305,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:04:00Z +pr,183328,[special] Add arbitrary-order modified Bessel functions,https://github.com/pytorch/pytorch/pull/183328,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T00:43:43Z +pr,189319,Enable flake8-bugbear B007 (unused loop control variable),https://github.com/pytorch/pytorch/pull/189319,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/h100-symm-mem, ciflow/inductor, ciflow/inductor-pallas","identify whether block is CI, merge conflict, or review gate",2026-07-09T00:09:57Z +pr,181781,[dynamo] turn canonicalize_output_graph_node_order on by default,https://github.com/pytorch/pytorch/pull/181781,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:50:03Z +pr,188137,[inductor] Decompose shard_dim_alltoall via regular decomposition,https://github.com/pytorch/pytorch/pull/188137,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:49:43Z +pr,187468,[dynamo] add break_graph_if_unsupported to LOAD_ATTR,https://github.com/pytorch/pytorch/pull/187468,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:28Z +pr,187469,[dynamo] extract UDOV hooks: lookup_instance_dict and call_getattr_fallback,https://github.com/pytorch/pytorch/pull/187469,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:14Z +pr,181780,[dynamo] set canonicalize_output_graph_node_order=True for export tests,https://github.com/pytorch/pytorch/pull/181780,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:22:50Z +pr,189190,flex_gemm: extend feed-main to trailing pointwise and axis-1 fragments,https://github.com/pytorch/pytorch/pull/189190,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:12:53Z +pr,185359,Add CE multicast low-contention all-gather,https://github.com/pytorch/pytorch/pull/185359,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T22:47:54Z +pr,187458,[MPS] Fuse cross_entropy into a native Metal kernel,https://github.com/pytorch/pytorch/pull/187458,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T22:47:52Z +pr,189295,Pin flash-attention checkout in FA3 ABI-stable H100 smoke test,https://github.com/pytorch/pytorch/pull/189295,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100,"identify whether block is CI, merge conflict, or review gate",2026-07-08T22:36:47Z +pr,189257,[rocm] Re-enable conv2d backward parametrized test on ROCm by using fp16 on NAVI,https://github.com/pytorch/pytorch/pull/189257,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-rocm-mi200, ciflow/inductor-rocm-mi300, ciflow/inductor-rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-07-08T22:19:46Z +pr,189080,[inductor] Generalize reinplacing to functional ATen ops with an in-place variant,https://github.com/pytorch/pytorch/pull/189080,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T22:03:33Z +pr,186790,Remove c10/util/Array.h,https://github.com/pytorch/pytorch/pull/186790,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:50:46Z +pr,189277,Skip redundant contiguity rescan in empty_tensor_restride (robust) (#189277),https://github.com/pytorch/pytorch/pull/189277,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:50:21Z +pr,175900,[Inductor] Fix Enum serialization in user-defined Triton kernel codegen (#175900),https://github.com/pytorch/pytorch/pull/175900,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:50:16Z +pr,189220,Strip the arc experiment from workflow check_experiments,https://github.com/pytorch/pytorch/pull/189220,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor-pallas,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:36:08Z +pr,189202,[CuteDSL][RMSNorm] use const_data_ptr for alignment check,https://github.com/pytorch/pytorch/pull/189202,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:23:13Z +pr,189153,Bump onnx from 1.21.0 to 1.22.0 in /.ci/docker,https://github.com/pytorch/pytorch/pull/189153,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:09:36Z +pr,177757,[MPS] support for SparseCSR tensor and its unary operators,https://github.com/pytorch/pytorch/pull/177757,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:06:41Z +pr,188824,Port doctest tests from cpython test_generators.py,https://github.com/pytorch/pytorch/pull/188824,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:06:06Z +pr,157149,[generator] Close all open generators in compile_subgraph,https://github.com/pytorch/pytorch/pull/157149,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:06:05Z +pr,187744,[Dynamo] Add a per-generator exception stack,https://github.com/pytorch/pytorch/pull/187744,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:06:02Z +pr,187690,[Dynamo] Refactor generator methods,https://github.com/pytorch/pytorch/pull/187690,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T21:06:01Z +pr,189131,Skip building unused XNNPACK microkernels-all; guard GCC14 workaround,https://github.com/pytorch/pytorch/pull/189131,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:53:11Z +pr,184447,[inductor][autotuning] add incremental autotune foundation,https://github.com/pytorch/pytorch/pull/184447,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:50:06Z +pr,183684,[inductor] extract _bench_launchers and _finalize_autotune_winner helpers,https://github.com/pytorch/pytorch/pull/183684,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:49:59Z +pr,183683,[inductor] extract _iter_compile_results generator with optional thread pool,https://github.com/pytorch/pytorch/pull/183683,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:49:54Z +pr,183682,"[inductor] extract _bundle_compile_result, _all_failed_fallback, _maybe_put_static_autotuner",https://github.com/pytorch/pytorch/pull/183682,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:49:49Z +pr,181826,[inductor][autotuning] extract _should_coordesc_tune cached property (#181826),https://github.com/pytorch/pytorch/pull/181826,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:49:37Z +pr,189289,[TEST] Refactor test_c10d_pypg.py with hw_classification,https://github.com/pytorch/pytorch/pull/189289,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:49:21Z +pr,189263,Remove the Miniconda base env from the CD builder images,https://github.com/pytorch/pytorch/pull/189263,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries, ciflow/docker","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:48:54Z +pr,189206,[Inductor] register `target_code` in `input_codes` tracker,https://github.com/pytorch/pytorch/pull/189206,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:35:27Z +pr,189186,"Extend ""Fix positional arg overflow in TritonBenchmarkRequest when template inputs are deduplicated"" to CUTLASS backend",https://github.com/pytorch/pytorch/pull/189186,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:27:22Z +pr,189247,Register _foreach_clone XPU dispatch,https://github.com/pytorch/pytorch/pull/189247,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:26:40Z +pr,186056,Support input mutations in CUDA graph control flow.,https://github.com/pytorch/pytorch/pull/186056,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:26:01Z +pr,189044,[inductor] Handle symbolic float scalars in FX wrapper metadata hook and reference analysis (#189044),https://github.com/pytorch/pytorch/pull/189044,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:25:58Z +pr,106149,Automated submodule update: kineto,https://github.com/pytorch/pytorch/pull/106149,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:25:54Z +pr,189197,make post-processing in codegen mandatory,https://github.com/pytorch/pytorch/pull/189197,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:25:29Z +pr,189246,[DO NOT MERGE] Test L3 CRCR,https://github.com/pytorch/pytorch/pull/189246,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/crcr/crcr-test,"identify whether block is CI, merge conflict, or review gate",2026-07-08T20:23:09Z +pr,189042,[dynamo] Add specialize_args option to assume_constant_result,https://github.com/pytorch/pytorch/pull/189042,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T20:16:02Z +pr,183910,[Test] Make test/nn/test_embedding.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/183910,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T19:50:46Z +pr,185251,[Test] Move device-agnostic grouped_gemm tests to new test_matmul.py,https://github.com/pytorch/pytorch/pull/185251,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:50:28Z +pr,188440,Simplify atomic_add_float using atomic_ref,https://github.com/pytorch/pytorch/pull/188440,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/android, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:22:53Z +pr,187307,Fix the potential global range overflow issue,https://github.com/pytorch/pytorch/pull/187307,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:13:28Z +pr,189076,[inductor] null-guard size/alignment asserts for optional outputs,https://github.com/pytorch/pytorch/pull/189076,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:13:05Z +pr,180250,Add setup_.py shim for legacy build commands,https://github.com/pytorch/pytorch/pull/180250,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T19:12:48Z +pr,174699,Add CuTeDSL-based scaled grouped MM kernel,https://github.com/pytorch/pytorch/pull/174699,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/docker, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:45:25Z +pr,189276,Update third_party/kineto submodule to cb6e5f7,https://github.com/pytorch/pytorch/pull/189276,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T18:10:06Z +pr,184365,Memory aware partition selection for partitioned scatter fx pass,https://github.com/pytorch/pytorch/pull/184365,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-perf-test-nightly, ciflow/inductor-perf-test-nightly-rocm-mi350, ciflow/inductor-rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:04:47Z +pr,187508,c10d: register nccl2 backend,https://github.com/pytorch/pytorch/pull/187508,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-08T17:00:06Z +pr,186918,[Test] Add test hardware classification for test classes,https://github.com/pytorch/pytorch/pull/186918,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-08T16:17:52Z +pr,188181,Move __dict__ view (get_dict_vt) onto base VariableTracker,https://github.com/pytorch/pytorch/pull/188181,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:17:37Z +pr,189178,Add PyTorch AI policy,https://github.com/pytorch/pytorch/pull/189178,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T16:15:18Z +pr,185918,[Test] Make autocast op-list tests device-agnostic via registry,https://github.com/pytorch/pytorch/pull/185918,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T15:57:15Z +pr,189103,Support dist.split_group on CPU-only (gloo) process groups,https://github.com/pytorch/pytorch/pull/189103,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:35:21Z +pr,188650,[SymmMem] Allocate symmetric memory through the CUDACachingAllocator v2,https://github.com/pytorch/pytorch/pull/188650,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200-symm-mem, ciflow/h100-symm-mem, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:26:33Z +pr,185220,[Test] Make test_wrap_inductor_compiled_regions.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/185220,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T14:08:40Z +pr,188736,[xpu][inductor] update Intel Triton commit pin,https://github.com/pytorch/pytorch/pull/188736,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-08T10:34:30Z +pr,178336,[Distributed] Make DDP tests and tensor parallel dependencies backend agnostic,https://github.com/pytorch/pytorch/pull/178336,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-08T10:17:37Z +pr,184355,[3/N][Test] Migrate TestBwdGradients device-agnostic skips to `@skipOps`,https://github.com/pytorch/pytorch/pull/184355,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-08T09:32:56Z +pr,184356,[4/N][Test] Migrate TestFwdGradients device-agnostic skips to `@skipOps`,https://github.com/pytorch/pytorch/pull/184356,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-08T08:41:27Z +pr,176574,Increase dynamo coverage of the `operator` module by adding polyfills,https://github.com/pytorch/pytorch/pull/176574,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T08:08:34Z +pr,186915,[WIP][XPU]Enable inductor cases when issues are fixed,https://github.com/pytorch/pytorch/pull/186915,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-07-08T07:45:36Z +pr,181000,[inductor] Dump Python stacks on CI test subprocess timeout,https://github.com/pytorch/pytorch/pull/181000,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/rocm-mi200, ciflow/rocm-mi300, ciflow/rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:51:27Z +pr,182376,[CUDA][CPU] Unify CPU/GPU relu/max behavior,https://github.com/pytorch/pytorch/pull/182376,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T05:34:02Z +pr,188043,[Test] Make test/test_sort_and_select.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/188043,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T03:59:54Z +pr,189195,Cuda Graph Default Enable (#189195),https://github.com/pytorch/pytorch/pull/189195,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T02:23:24Z +pr,181357,[dynamo] Add polyfill for operator.indexOf,https://github.com/pytorch/pytorch/pull/181357,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T02:18:56Z +pr,176876,[inductor] Add device-type benchmarking dispatch registry,https://github.com/pytorch/pytorch/pull/176876,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T02:18:46Z +pr,188276,[FSDP] Add MORI SDMA all-gather backend and zero-copy output capability,https://github.com/pytorch/pytorch/pull/188276,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T00:54:36Z +pr,183620,Port 2 tests under test/distributed folder to XPU,https://github.com/pytorch/pytorch/pull/183620,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-08T00:49:47Z +pr,189027,[dynamo] Guard os.environ reads with an ambient env-var guard,https://github.com/pytorch/pytorch/pull/189027,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T00:38:03Z +pr,179550,Enable eager tests for inductor-unittest - Enable TestViewOps and TestDevicePrecision on Intel GPU,https://github.com/pytorch/pytorch/pull/179550,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-07T23:20:53Z +pr,179549,Enable eager tests for inductor-unittest - define xfails for xpu in op_db for test_ops.py using device_type list in DecorateInfo,https://github.com/pytorch/pytorch/pull/179549,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-07T23:05:37Z +pr,176689,Enable eager tests for inductor-unittest - enable test_ops_fwd_gradients.py and test_ops_gradients.py on Intel GPU,https://github.com/pytorch/pytorch/pull/176689,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-07T22:44:17Z +pr,176688,Enable eager tests for inductor-unittest - enable test_dtypes and align XPU op dyptes with OpInfo,https://github.com/pytorch/pytorch/pull/176688,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-07T22:41:50Z +pr,186100,"[dynamo, nested graph breaks] enable nested_graph_breaks by default on OSS",https://github.com/pytorch/pytorch/pull/186100,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T22:27:42Z +pr,185524,"[dynamo, nested graph breaks] add strong nested graph break test infrastructure",https://github.com/pytorch/pytorch/pull/185524,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T22:27:31Z +pr,189095,[inductor] Preserve stream annotations across pattern replacements,https://github.com/pytorch/pytorch/pull/189095,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T21:59:07Z +pr,187131,[muon] add multi-tensor (foreach) standard Muon path (#187131),https://github.com/pytorch/pytorch/pull/187131,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-07T21:29:05Z +pr,187133,[muon] add CUDA graph capture for the NS computation (#187133),https://github.com/pytorch/pytorch/pull/187133,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-07T20:51:41Z +pr,187132,[muon] add Gram Newton-Schulz algorithm + batched standard NS for square groups (#187132),https://github.com/pytorch/pytorch/pull/187132,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-07T20:51:40Z +pr,188979,remaining HOPs register_fake,https://github.com/pytorch/pytorch/pull/188979,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-07T20:51:23Z +pr,182145,[ROCm] Reland #181955: Redirect index_add_ to scatter_add_ for large tensors,https://github.com/pytorch/pytorch/pull/182145,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi300,"identify whether block is CI, merge conflict, or review gate",2026-07-07T20:12:10Z +pr,165766,Enable kineto XPU scope profiler extending existing XPU profiler plugin,https://github.com/pytorch/pytorch/pull/165766,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-07T19:13:14Z +pr,188130,[Test] Make test_autocast device-agnostic for Out-of-Tree backends,https://github.com/pytorch/pytorch/pull/188130,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-07T17:33:31Z +pr,184946,[Inductor] Fix split reduction codegen and skip redundant masking for block ptr backends (#184946),https://github.com/pytorch/pytorch/pull/184946,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:14:10Z +pr,188766,[CP] Fix FlexCP + load balancing enabled + aot_fx_trace issue,https://github.com/pytorch/pytorch/pull/188766,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T17:06:29Z +pr,187164,[ROCm] unskip export tests on ROCm,https://github.com/pytorch/pytorch/pull/187164,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi200, ciflow/rocm-mi300, ciflow/rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-07-07T14:16:38Z +pr,187315,[xpu][3/4] inductor: route MX scaled_mm_v2 fallback through v2 aten kernel,https://github.com/pytorch/pytorch/pull/187315,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-07T10:42:24Z +pr,187931,[XPU Graph] Refine XPU graph capture detection in XPUCachingAllocator,https://github.com/pytorch/pytorch/pull/187931,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-07T06:47:39Z +pr,187549,[xpu][test]testing: enable tf32_on_and_off() for XPU devices,https://github.com/pytorch/pytorch/pull/187549,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-07T05:13:31Z +pr,187876,Fix mark dynamic half bounded range,https://github.com/pytorch/pytorch/pull/187876,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T03:07:05Z +pr,188050,Add FX graph cache custom guard replay test,https://github.com/pytorch/pytorch/pull/188050,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T02:04:37Z +pr,189107,Introduce Inductor lowering for _philox_uniform,https://github.com/pytorch/pytorch/pull/189107,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T01:54:01Z +pr,188000,Vectorize LayerNormSecondPass null affine cases,https://github.com/pytorch/pytorch/pull/188000,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-07T01:53:56Z +pr,188312,[cpu] Address clang-tidy warnings in vec(256/512)_qint.h,https://github.com/pytorch/pytorch/pull/188312,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-07T01:53:50Z +pr,182199,[xpu][test] Port distributed _shard tests cases on Intel GPUs,https://github.com/pytorch/pytorch/pull/182199,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-07T01:48:19Z +pr,180608,[5.5/11][aoti] Add V2 runtime test via ctypes,https://github.com/pytorch/pytorch/pull/180608,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T01:08:42Z +pr,180885,CUDA: unrolled kernels for upsample_linear1d forward/backward when grid is small,https://github.com/pytorch/pytorch/pull/180885,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-07T01:08:40Z +pr,180283,[cuBLAS][cuBLASLt] Thread-local cuBLAS(Lt) workspace maps,https://github.com/pytorch/pytorch/pull/180283,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/h100-symm-mem, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-07-07T00:07:43Z +pr,187891,device_mesh: fall back to new_group when parent backend lacks split s…,https://github.com/pytorch/pytorch/pull/187891,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-06T23:07:52Z +pr,182183,[PT2][export] Add unsafe_allow_callable_serialization for scoped callable serialization,https://github.com/pytorch/pytorch/pull/182183,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T22:51:30Z +pr,183759,[Testing] Expansion of non-contiguous tensor ops testing to cover more complicated and real-world applications,https://github.com/pytorch/pytorch/pull/183759,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T22:34:52Z +pr,188451,[cpu] Address clang-tidy warnings in vec128_float_neon.h,https://github.com/pytorch/pytorch/pull/188451,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:53:05Z +pr,188311,[cpu] Address clang-tidy warnings in vec(256/512)_int.h,https://github.com/pytorch/pytorch/pull/188311,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:52:58Z +pr,187675,"[inductor] Add fuse_region to set fusions,memory reuse boundaries",https://github.com/pytorch/pytorch/pull/187675,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:29:48Z +pr,188571,C++ meta kernel for torch.cat (symint aware),https://github.com/pytorch/pytorch/pull/188571,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:27:56Z +pr,152806,[invoke_subgraph] Force the output stride to be same as eager,https://github.com/pytorch/pytorch/pull/152806,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:13:35Z +pr,187778,Flatten all_to_all_nd copy to fix narrow-row throughput,https://github.com/pytorch/pytorch/pull/187778,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:13:04Z +pr,183705,[SymmMem] Normalize device index in empty_strided_p2p (#183705),https://github.com/pytorch/pytorch/pull/183705,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:12:51Z +pr,186370,inductor: validate square mix-order reductions by common reads,https://github.com/pytorch/pytorch/pull/186370,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:12:50Z +pr,185109,Remove custom FindCUDAToolkit.cmake module,https://github.com/pytorch/pytorch/pull/185109,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:12:49Z +pr,187015,remove spmd_types dependency from nightly wheel,https://github.com/pytorch/pytorch/pull/187015,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:12:18Z +pr,189062,[profiler][cupti] Make CUPTI monitor buffer_size / flush_period_s config-only,https://github.com/pytorch/pytorch/pull/189062,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:11:27Z +pr,187007,Use distinct cache variables for linker option checks,https://github.com/pytorch/pytorch/pull/187007,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:10:52Z +pr,187009,Apply libnop clang warning suppression to TensorPipe subtargets,https://github.com/pytorch/pytorch/pull/187009,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-06T20:10:50Z +pr,187961,[inductor] Add compile-time per-subkernel autotuning for combo kernels,https://github.com/pytorch/pytorch/pull/187961,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T17:18:21Z +pr,176293,[CI] aarch64 manywheel and cuda image build fixes,https://github.com/pytorch/pytorch/pull/176293,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries, ciflow/docker, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T15:59:45Z +pr,188822,[Inductor][Host-side TMA] Support tensordesc<> in the static Triton launcher,https://github.com/pytorch/pytorch/pull/188822,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T15:48:29Z +pr,180208,Update slow tests,https://github.com/pytorch/pytorch/pull/180208,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/slow, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T14:46:19Z +pr,189005,fix(fx): filter internal frames in record_stack_traces to resolve broken stack traces (#130861),https://github.com/pytorch/pytorch/pull/189005,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-06T14:32:35Z +pr,183276,[Inductor][Refactor] 2/N Introduce shared heuristic registry with codegen_heuristic support,https://github.com/pytorch/pytorch/pull/183276,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-06T12:16:15Z +pr,183277,[Inductor][Refactor] 3/N Move pointwise heuristics from triton_heuristics.py into triton_codegen/,https://github.com/pytorch/pytorch/pull/183277,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-06T12:16:14Z +pr,182577,[Inductor] Enable combo_kernel_per_subkernel_blocks by default,https://github.com/pytorch/pytorch/pull/182577,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T09:15:13Z +pr,184955,[Inductor] Exclude indirect-indexing nodes from combo kernels,https://github.com/pytorch/pytorch/pull/184955,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-06T09:15:06Z +pr,185797,[Test] Refactor test/test_bmm_outer_product.py,https://github.com/pytorch/pytorch/pull/185797,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-06T08:27:23Z +pr,182729,[DTensor] Preserve placement for aten.clone.default,https://github.com/pytorch/pytorch/pull/182729,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T01:09:44Z +pr,182727,[shard prop] P(*).clone()->P(*),https://github.com/pytorch/pytorch/pull/182727,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T01:09:43Z +pr,188067,[ROCm] fix wait instructions,https://github.com/pytorch/pytorch/pull/188067,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-05T23:05:01Z +pr,182138,Update torchcomms.txt commit hash,https://github.com/pytorch/pytorch/pull/182138,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-distributed, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-05T20:49:09Z +pr,188517,Fix/180346 dtensor clip grad mixed mesh,https://github.com/pytorch/pytorch/pull/188517,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-05T20:36:01Z +pr,188516,Fix/187177 localresponsenorm validation,https://github.com/pytorch/pytorch/pull/188516,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-05T20:36:00Z +pr,182530,index_add_/index_reduce_: add env-var kill switch for VecSize=4 vectorized path,https://github.com/pytorch/pytorch/pull/182530,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-05T18:51:23Z +pr,170571,[CUDA][cuBLASLt] addmm - Lt path improvements,https://github.com/pytorch/pytorch/pull/170571,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/nightly","identify whether block is CI, merge conflict, or review gate",2026-07-05T15:53:02Z +pr,181743,Remove skipXPU from test_binary_ufuncs,https://github.com/pytorch/pytorch/pull/181743,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-05T14:59:27Z +pr,187897,fix: Add explicit null-buffer guards and shared-memory fallback for Intel Macs with discrete GPU (MPS),https://github.com/pytorch/pytorch/pull/187897,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-05T14:29:19Z +pr,187550,[test][inductor] Fix stale `xfail` and add SM12 skip.,https://github.com/pytorch/pytorch/pull/187550,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-05T09:39:14Z +pr,186906,Update trace_rules for torch.accelerator,https://github.com/pytorch/pytorch/pull/186906,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-04T23:59:10Z +pr,187061,[cuda graphs] Automatic nn.Module FQN annotation in Inductor cudagraph trees,https://github.com/pytorch/pytorch/pull/187061,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-04T23:34:08Z +pr,182529,Use NCCLCommProvider for TorchComm + NCCL symmetric memory,https://github.com/pytorch/pytorch/pull/182529,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-04T20:49:04Z +pr,183776,C++ FakeTensor,https://github.com/pytorch/pytorch/pull/183776,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-04T13:07:15Z +pr,187732,[DDP] Fix static_graph=True + no_sync() gradient accumulation regression,https://github.com/pytorch/pytorch/pull/187732,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-04T11:57:08Z +pr,178500,[autograd] Support Mapping and OrderedDict inputs for autograd.grad and autograd.backward,https://github.com/pytorch/pytorch/pull/178500,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-04T08:36:05Z +pr,182134,[dynamo_wrapped] Remove stale dynamo_expected_failures for torch_np error-path tests,https://github.com/pytorch/pytorch/pull/182134,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-04T07:27:38Z +pr,182362,[PyTorch][AOTI] Add timing instrumentation to AOTI model loading pipeline,https://github.com/pytorch/pytorch/pull/182362,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-03T23:50:44Z +pr,182310,Add Triton scaled-mm v2 blockwise recipes,https://github.com/pytorch/pytorch/pull/182310,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-03T22:50:18Z +pr,182146,[caffe2] Add defensive fallback in get_logging_handler for unknown destinations (#182146),https://github.com/pytorch/pytorch/pull/182146,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-03T20:59:18Z +pr,176239,[dtensor][partial] just_knob error message for fixing .item() on partials,https://github.com/pytorch/pytorch/pull/176239,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-03T20:59:09Z +pr,188646,[inductor] Add tile_n=192 and missing cluster configs to NVGEMM supplement set,https://github.com/pytorch/pytorch/pull/188646,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-03T20:15:05Z +pr,180083,[FSDP] Make FSDP v1 tests device-agnostic for PrivateUse1 backends,https://github.com/pytorch/pytorch/pull/180083,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem,"identify whether block is CI, merge conflict, or review gate",2026-07-03T16:59:22Z +pr,187691,[BE]: Update fmtlib to 12.2.0,https://github.com/pytorch/pytorch/pull/187691,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-03T16:42:16Z +pr,188789,[xpu] Add IPC memory handle sharing support to XPUCachingAllocator,https://github.com/pytorch/pytorch/pull/188789,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-03T11:56:08Z +pr,183323,Adding conversion from PyObject to torch::stable::tensor,https://github.com/pytorch/pytorch/pull/183323,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-07-03T11:00:36Z +pr,181727,[xpu][2/4]Implement scaled_mm_v1 for MXFP8/MXFP4/NVFP4 on XPU,https://github.com/pytorch/pytorch/pull/181727,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-03T07:59:30Z +pr,186902,Generalize dynamo RNG-related part to be device-agnostic,https://github.com/pytorch/pytorch/pull/186902,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-03T07:06:47Z +pr,185115,[Test] Enable TestSDPAGpuOnly on XPU and skip cases unsupported by torch-xpu-ops,https://github.com/pytorch/pytorch/pull/185115,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-03T02:42:08Z +pr,175472,Add FA4 fp8 implementation to SDPA,https://github.com/pytorch/pytorch/pull/175472,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-03T01:01:47Z +pr,182920,Add backward pass for cuDNN FP8 per-tensor attention,https://github.com/pytorch/pytorch/pull/182920,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-03T01:01:35Z +pr,179224,spaceship operator for comparisons in Dict.h,https://github.com/pytorch/pytorch/pull/179224,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-03T00:20:37Z +pr,188439,[CUDA] [CI] fix potential unit test failure after cuda graph xfail (resource leak),https://github.com/pytorch/pytorch/pull/188439,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T23:15:38Z +pr,188827,[AI Codemod][PerfAICT-General] Use SymDimVector to avoid heap allocs in select_symint,https://github.com/pytorch/pytorch/pull/188827,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T20:29:49Z +pr,188826,[AI Codemod][PerfAICT-General] Use SymDimVector to avoid heap allocs in select_symint,https://github.com/pytorch/pytorch/pull/188826,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T19:51:26Z +pr,188502,Fix/188034 autograd setup context leak,https://github.com/pytorch/pytorch/pull/188502,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-02T18:49:55Z +pr,179436,Run clang-tidy with readability-container-contains,https://github.com/pytorch/pytorch/pull/179436,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-07-02T18:37:45Z +pr,173390,Add interactive glossary with hover tooltips,https://github.com/pytorch/pytorch/pull/173390,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-02T18:18:30Z +pr,188755,cudagraph_trees: honor mark_static_address outputs,https://github.com/pytorch/pytorch/pull/188755,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T16:46:48Z +pr,187860,fix DDE in meta__transformer_encoder_layer_fwd,https://github.com/pytorch/pytorch/pull/187860,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T16:39:26Z +pr,188771,[inductor] Rewind block_ptr across outer loop in nested reductions (#188771),https://github.com/pytorch/pytorch/pull/188771,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T16:35:35Z +pr,186599,Introduce isInitialized&set_rng_state in torch.accelerator,https://github.com/pytorch/pytorch/pull/186599,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T14:23:58Z +pr,181041,[hammer] Make HSTU manual recompute policies compatible with PT2 compile,https://github.com/pytorch/pytorch/pull/181041,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T14:16:47Z +pr,174323,[effectful ops] get tokens working with AC,https://github.com/pytorch/pytorch/pull/174323,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-02T14:16:33Z +pr,170396,[typing] Fix return type annotations for various functions,https://github.com/pytorch/pytorch/pull/170396,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T11:35:51Z +pr,185114,[Test] Refractor TestSDPACudaOnly into TestSDPAGpuOnly with CUDA-only assumptions removed,https://github.com/pytorch/pytorch/pull/185114,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-02T07:39:39Z +pr,185113,[Test] Decouple test_transformers.py from specific hardware backends,https://github.com/pytorch/pytorch/pull/185113,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-02T07:39:38Z +pr,187938,[Inductor] Support non-Triton fused kernels in the FX backend via a high-order-operator,https://github.com/pytorch/pytorch/pull/187938,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T07:07:50Z +pr,180050,[FSDPV2] Make FSDP v2 and composable tests device-agnostic for PrivateUse1 backends,https://github.com/pytorch/pytorch/pull/180050,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem,"identify whether block is CI, merge conflict, or review gate",2026-07-02T05:49:23Z +pr,183306,[AI Codemod][ClaudeCodeFbandroidWarningsAsErrors] [3] Bulk run of CodemodConfigClaudeCodeFbandroidWarningsAsErrors (10 units) (#183306),https://github.com/pytorch/pytorch/pull/183306,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T04:50:02Z +pr,186122,[ROCm] Enable test_inverse_errors_large & test_lu_solve_batched_many_batches linalg tests,https://github.com/pytorch/pytorch/pull/186122,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi200, ciflow/rocm-mi300, ciflow/rocm-nightly, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T03:29:51Z +pr,186771,Introduce _isInBadFork and refine torch.random,https://github.com/pytorch/pytorch/pull/186771,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T01:50:27Z +pr,186816,Remove deprecated torch.chain_matmul,https://github.com/pytorch/pytorch/pull/186816,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T01:47:46Z +pr,186899,[2/N] Replace module-level RNG-related APIs with torch.accelerator,https://github.com/pytorch/pytorch/pull/186899,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T01:39:12Z +pr,186968,dynamo: enforce target_values is None|Sequence.,https://github.com/pytorch/pytorch/pull/186968,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T01:26:05Z +pr,188744,[dynamo] fix lazy module initialization with symbolic shapes,https://github.com/pytorch/pytorch/pull/188744,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T01:18:24Z +pr,188745,"[dynamo, nested graph breaks] fix empty nn.Module hook dict reconstruction",https://github.com/pytorch/pytorch/pull/188745,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T01:17:42Z +pr,188581,[dynamo] Add polyfills for operator.contains and operator.call,https://github.com/pytorch/pytorch/pull/188581,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-02T00:52:13Z +pr,185700,[OpenReg] Add compiler integration doc for OOT accelerators,https://github.com/pytorch/pytorch/pull/185700,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-02T00:42:37Z +pr,175011,[S615805][RecSysCheckpointing][PyTorch] Add strides to TensorProperties (#175011),https://github.com/pytorch/pytorch/pull/175011,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T00:10:07Z +pr,188616,[AI Codemod][PerfAICT-General] Skip per-operand has_storage scan for non-PrivateUse1 devices in build,https://github.com/pytorch/pytorch/pull/188616,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-02T00:09:49Z +pr,188563,[AI Codemod][PerfAICT-General] Enable auto-vectorization of at::vec::convert via __restrict and indexed access,https://github.com/pytorch/pytorch/pull/188563,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-01T21:46:16Z +pr,188165,"[cutedsl] Pass current compute capability e.g., SM 10.3 -> `sm_103a` to CuTeDSL compile subprocess",https://github.com/pytorch/pytorch/pull/188165,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T05:14:17Z +pr,187224,Duplicate concurrent alignment fixups,https://github.com/pytorch/pytorch/pull/187224,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T05:13:59Z +pr,178710,Add native torch.cuda.gds.save/load for GPUDirect Storage,https://github.com/pytorch/pytorch/pull/178710,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-01T03:50:58Z +pr,188316,[ROCm] Cap HIP stream pool per priority; add reservable streams,https://github.com/pytorch/pytorch/pull/188316,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-30T21:58:01Z +pr,187205,Add get_device_name and memory snapshot accelerator APIs,https://github.com/pytorch/pytorch/pull/187205,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T21:57:59Z +pr,176894,[cuDNN][Convolution] Expose cuDNN `allow_reduced_precision_reduction` numerical note to fix `float16` tests,https://github.com/pytorch/pytorch/pull/176894,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T20:13:21Z +pr,188501,Fix/147170 deterministic partition ordering,https://github.com/pytorch/pytorch/pull/188501,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T19:03:15Z +pr,188498,Fix/grammar with the same parameters,https://github.com/pytorch/pytorch/pull/188498,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T19:02:00Z +pr,188532,Fix/threshold inplace validation,https://github.com/pytorch/pytorch/pull/188532,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T19:01:00Z +pr,188499,Fix/typo paramater,https://github.com/pytorch/pytorch/pull/188499,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-30T18:54:09Z +pr,188339,Fix/typo overridden applier,https://github.com/pytorch/pytorch/pull/188339,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-30T17:56:08Z +pr,188465,[Doc] Document GIL and allocation constraints for PU1 custom allocators,https://github.com/pytorch/pytorch/pull/188465,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T15:36:37Z +pr,182051,[docs] Link CUDA stream semantics from Tensor.to/copy_ non_blocking docstrings,https://github.com/pytorch/pytorch/pull/182051,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T15:20:48Z +pr,182986,[Eager] [Bitwise-Equivalence] Implement an INNER_TREE sum reduction (#182986),https://github.com/pytorch/pytorch/pull/182986,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-30T08:48:01Z +pr,188535,[test stats] Sanitize empty-file pytest rows at the upload boundary,https://github.com/pytorch/pytorch/pull/188535,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T08:47:24Z +pr,180907,[shard prop] tag more ops,https://github.com/pytorch/pytorch/pull/180907,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-30T05:55:38Z +pr,187150,Remove unused CUDA 12.4 docker build paths,https://github.com/pytorch/pytorch/pull/187150,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor-pallas","identify whether block is CI, merge conflict, or review gate",2026-06-30T01:41:04Z +pr,181944,[cuBLAS][cuBLASLt] Shared default workspace size across cuBLAS and Lt,https://github.com/pytorch/pytorch/pull/181944,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-29T21:04:40Z +pr,188444,[Test] Refactor tensor_creation_ops and foreach.py to be device-agnostic.,https://github.com/pytorch/pytorch/pull/188444,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T21:04:11Z +pr,188340,Fix/indices or section typo,https://github.com/pytorch/pytorch/pull/188340,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T21:02:55Z +pr,188296,linear_cross_entropy: support probability targets with reduction='none',https://github.com/pytorch/pytorch/pull/188296,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps, ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-29T21:02:34Z +pr,188341,Fix/184845 pad docs,https://github.com/pytorch/pytorch/pull/188341,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T21:02:09Z +pr,188385,[fx] _iterate_exprs: treat device/dtype/layout/memory_format as symbol-free,https://github.com/pytorch/pytorch/pull/188385,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-29T20:24:36Z +pr,182011,[WIP][PoC] Enable Arm Optimized Routines for AArch64+Linux,https://github.com/pytorch/pytorch/pull/182011,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/pull, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-29T19:13:21Z +pr,188294,Fix GradientEdge None-grad error message formatting,https://github.com/pytorch/pytorch/pull/188294,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T16:27:59Z +pr,182851,[dynamo] Fix free-threading race in dict recursive-tag watcher callback,https://github.com/pytorch/pytorch/pull/182851,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/tsan","identify whether block is CI, merge conflict, or review gate",2026-06-29T16:27:23Z +pr,188295,Validate queue_callback input is callable,https://github.com/pytorch/pytorch/pull/188295,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T16:27:19Z +pr,187922,[TEST] Refactor test_embedding.py with hw_classification,https://github.com/pytorch/pytorch/pull/187922,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T03:35:32Z +pr,175809,[cuDNN] Migrate to new frontend API for batchnorm,https://github.com/pytorch/pytorch/pull/175809,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T01:29:15Z +pr,177924,[triton hash update] update the pinned triton hash,https://github.com/pytorch/pytorch/pull/177924,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-28T23:52:32Z +pr,178283,Bump torch_tpu pin,https://github.com/pytorch/pytorch/pull/178283,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-pallas","identify whether block is CI, merge conflict, or review gate",2026-06-28T20:49:52Z +pr,173914,[CI] [CUDA] Add CUDA ARM CI image,https://github.com/pytorch/pytorch/pull/173914,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-28T20:49:51Z +pr,179820,[ATen][Native][CUDA] Use dynamic loading in CUDNN FrontEnd,https://github.com/pytorch/pytorch/pull/179820,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries,"identify whether block is CI, merge conflict, or review gate",2026-06-28T19:57:15Z +pr,187711,Strict pyrefly checking: round 1 (2/N),https://github.com/pytorch/pytorch/pull/187711,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-28T19:22:51Z +pr,181767,"[shard prop] autogen infra for inplace,out variants",https://github.com/pytorch/pytorch/pull/181767,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-27T22:50:21Z +pr,181724,S390X: update binaries labels,https://github.com/pytorch/pytorch/pull/181724,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-27T18:54:05Z +pr,181623,[torch.library] Add pallas_op for JAX/Pallas kernel registration,https://github.com/pytorch/pytorch/pull/181623,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-27T18:54:01Z +pr,174771,Fix `sample_kwargs` in `polygamma` op info,https://github.com/pytorch/pytorch/pull/174771,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-27T14:08:20Z +pr,187390,Widen aot_autograd annotations to match runtime values,https://github.com/pytorch/pytorch/pull/187390,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-27T09:18:37Z +pr,187692,Skip mobilenetv2_100 eager-nondeterministic accuracy check,https://github.com/pytorch/pytorch/pull/187692,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-27T09:18:31Z +pr,181530,[caffe2] Fix HIP build: cccl_counting_iterator needs explicit type,https://github.com/pytorch/pytorch/pull/181530,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-27T07:33:50Z +pr,178906,Fix batch_isend_irecv to support ProcessGroup subclasses,https://github.com/pytorch/pytorch/pull/178906,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-27T07:33:45Z +pr,188161,Demote changing CUDAGraph-managed inputs to copied inputs,https://github.com/pytorch/pytorch/pull/188161,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-27T02:16:25Z +pr,188103,Enable recursive dict-tag guard optimization by default on 3.12+ non-free-threaded,https://github.com/pytorch/pytorch/pull/188103,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-27T02:15:46Z +pr,188101,Require an installed dict watcher for the recursive dict-tag fast path on 3.12+,https://github.com/pytorch/pytorch/pull/188101,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-27T02:15:44Z +pr,181414,Update FBGEMM submodule to v1.6.0,https://github.com/pytorch/pytorch/pull/181414,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-27T01:21:39Z +pr,188298,[aot_autograd] Add PySourceBuilder for runtime-wrapper codegen,https://github.com/pytorch/pytorch/pull/188298,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-26T23:49:39Z +pr,180990,[pytorch] Fix symmetric quantization scale to match C++ ChooseQuantizationParams (#180990),https://github.com/pytorch/pytorch/pull/180990,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-26T22:53:26Z +pr,188289,[mtia] Make torch.distributions test suite portable across acelerators,https://github.com/pytorch/pytorch/pull/188289,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-26T22:23:50Z +pr,188215,[dtensor] Resolve ProcessGroup from mesh in-graph under compile_on_one_rank,https://github.com/pytorch/pytorch/pull/188215,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-26T19:03:04Z +pr,188093,Fix record_stack_traces: filter fx-internal frames from recorded trac…,https://github.com/pytorch/pytorch/pull/188093,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-26T18:32:23Z +pr,188180,Reuse full-range splits for persistent CSE,https://github.com/pytorch/pytorch/pull/188180,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-26T18:09:09Z +pr,187875,[AOTI] Refactor model container checkout ownership (#187875),https://github.com/pytorch/pytorch/pull/187875,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-26T17:30:50Z +pr,188063,Enable Blackwell FP8 grouped GEMM,https://github.com/pytorch/pytorch/pull/188063,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-26T16:56:20Z +pr,182948,[CUDA] Fix reentrant deadlock in torch.cuda._lazy_call,https://github.com/pytorch/pytorch/pull/182948,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-rocm, ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-26T15:20:31Z +pr,178751,Delay common_utils import to pytest_sessionstart to prevent faulthandler race condition,https://github.com/pytorch/pytorch/pull/178751,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-06-26T14:18:26Z +pr,186243,"[dynamo, nested graph breaks] enable NGB on dynamo-wrapped tests",https://github.com/pytorch/pytorch/pull/186243,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-26T03:29:55Z +pr,181398,[spmd_types] local_map,https://github.com/pytorch/pytorch/pull/181398,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-26T03:29:20Z +pr,188188,[DTensor] local_map normalizes shard dim checking in_placements,https://github.com/pytorch/pytorch/pull/188188,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T23:48:01Z +pr,186926,Remove sleef from test_aoti_abi_check linkage,https://github.com/pytorch/pytorch/pull/186926,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-25T22:47:19Z +pr,186947,[functorch] Add native_group_norm_backward.multiple_grads plumbing,https://github.com/pytorch/pytorch/pull/186947,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-25T22:08:41Z +pr,188179,Use tiling-aware hints for persistent reductions,https://github.com/pytorch/pytorch/pull/188179,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T22:05:36Z +pr,186946,[mps] Add kernel for native_group_norm_backward.multiple_grads,https://github.com/pytorch/pytorch/pull/186946,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-25T21:30:26Z +pr,186414,"native_group_norm: Handle non-contiguous tensors, rather than throwing",https://github.com/pytorch/pytorch/pull/186414,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-25T20:54:28Z +pr,184568,Always reset Dynamo if it was imported,https://github.com/pytorch/pytorch/pull/184568,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-25T20:14:25Z +pr,175242,Improve FX partitioner complexity,https://github.com/pytorch/pytorch/pull/175242,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-25T18:44:25Z +pr,178318,[Inductor] Add xmask unswitch for dynamic pointwise kernels,https://github.com/pytorch/pytorch/pull/178318,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T13:15:40Z +pr,181307,Set device in forward input tests,https://github.com/pytorch/pytorch/pull/181307,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-25T11:51:05Z +pr,187784,[TokenSwitch] Add expert-major dispatch layout,https://github.com/pytorch/pytorch/pull/187784,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T10:06:26Z +pr,186129,[dtensor] fix conv1d output placement and group scaling,https://github.com/pytorch/pytorch/pull/186129,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T08:14:44Z +pr,186877,Use C++20 std::numbers in c10 MathConstants,https://github.com/pytorch/pytorch/pull/186877,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-25T07:44:16Z +pr,186211,[Inductor] Skip nonfinite preservation for safe scaled softmax,https://github.com/pytorch/pytorch/pull/186211,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T03:16:55Z +pr,176920,Add distributed runtime autotuning for Triton kernels,https://github.com/pytorch/pytorch/pull/176920,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T02:16:53Z +pr,184924,[Test] [CUDA] Skip test_variant_consistency_jit for grid_sampler_2d,https://github.com/pytorch/pytorch/pull/184924,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-24T21:45:19Z +pr,184824,[Reland] Port D104346887/PR 182675 for index_add fast path (#184824),https://github.com/pytorch/pytorch/pull/184824,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-24T21:08:00Z +pr,188094,[profiler][cupti] Remove unused Python reference decoder (CuptiMonitorBuffer),https://github.com/pytorch/pytorch/pull/188094,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-24T21:07:35Z +pr,187797,Add linalg SVD operator benchmarks,https://github.com/pytorch/pytorch/pull/187797,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-24T20:34:10Z +pr,186777,[1/N] Replace module-level RNG-related APIs with torch.accelerator,https://github.com/pytorch/pytorch/pull/186777,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T19:04:55Z +pr,187208,[BE][Ez]: Use CPP20 rvalue overload for ostringstream,https://github.com/pytorch/pytorch/pull/187208,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-24T18:46:10Z +pr,187823,Fix regex for extracting gist URLs from logs,https://github.com/pytorch/pytorch/pull/187823,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-24T18:35:16Z +pr,186600,Introduce manual_seed&seed to torch.accelerator,https://github.com/pytorch/pytorch/pull/186600,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-24T18:20:51Z +pr,185213,[Test] Make dynamo test_streams.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/185213,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-24T16:23:50Z +pr,188058,[inductor] Avoid disjoint horizontal fusion when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/188058,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T16:15:51Z +pr,188056,[inductor] Avoid disjoint horizontal fusion when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/188056,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T16:15:45Z +pr,188059,[inductor] Block only peak-crossing fusions in the peak-memory guards,https://github.com/pytorch/pytorch/pull/188059,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T16:15:42Z +pr,187842,[inductor] Avoid cross-extern fusion when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/187842,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T15:24:18Z +pr,188055,[inductor] Evaluate compile-time ProcessGroup attrs in overlap estimates,https://github.com/pytorch/pytorch/pull/188055,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T15:24:11Z +pr,188057,[inductor] Evaluate compile-time ProcessGroup attrs in overlap estimates,https://github.com/pytorch/pytorch/pull/188057,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T15:24:08Z +pr,187832,[FX][Dynamo] Add tests for nn_module_stack and source_fn_stack metadata contracts,https://github.com/pytorch/pytorch/pull/187832,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-24T15:23:57Z +pr,184867,[symbolic shapes] Avoid direct replacement cycles,https://github.com/pytorch/pytorch/pull/184867,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-24T14:47:23Z +pr,187619,Remove Dynamo guard from custom op kernels,https://github.com/pytorch/pytorch/pull/187619,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-24T14:39:30Z +pr,181396,Print and validate full CUDA version in smoke test,https://github.com/pytorch/pytorch/pull/181396,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries,"identify whether block is CI, merge conflict, or review gate",2026-06-24T12:19:52Z +pr,180455,"[AARCH64] Enable MKLDNN backend for AArch64 for INT8 Matmul, using torch op _int_mm.dtype",https://github.com/pytorch/pytorch/pull/180455,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-24T06:45:05Z +pr,187471,[inductor][mtia] Strip AMD-only autotune kwargs from MTIA Triton compile configs (#187471),https://github.com/pytorch/pytorch/pull/187471,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T04:33:18Z +pr,180644,Relax align_random_eager test tolerances for Philox float precision,https://github.com/pytorch/pytorch/pull/180644,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-24T03:54:04Z +pr,182193,Tighten any_index_expr_overflows_int32 overflow check,https://github.com/pytorch/pytorch/pull/182193,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-24T03:01:44Z +pr,187520,[profiler][cupti] Comms monitoring tests,https://github.com/pytorch/pytorch/pull/187520,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-24T02:55:15Z +pr,187519,[profiler][cupti] Comms monitoring: CommsObserver + comm hook + plugins (CollTrace replacement),https://github.com/pytorch/pytorch/pull/187519,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-24T02:55:14Z +pr,187517,[profiler][cupti] CUDA-graph correlation utils: node-id remap + side-stream events,https://github.com/pytorch/pytorch/pull/187517,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-24T02:55:14Z +pr,187518,[profiler][cupti] In-process NCCL metadata plugin (ncclProfiler_v6),https://github.com/pytorch/pytorch/pull/187518,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-24T02:55:13Z +pr,186655,[profiler][cupti] Benchmarks: columnar-decode roofline + monitor distortion,https://github.com/pytorch/pytorch/pull/186655,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-24T02:55:11Z +pr,187328,Add fusion provenance to profiler timelines,https://github.com/pytorch/pytorch/pull/187328,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T22:47:00Z +pr,187367,Fix typos in comments and docstrings,https://github.com/pytorch/pytorch/pull/187367,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-23T22:41:56Z +pr,183468,[Inductor] Fix linter from tf32 fix,https://github.com/pytorch/pytorch/pull/183468,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-23T22:01:56Z +pr,187965,[inductor] Avoid disjoint horizontal fusion when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/187965,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T21:34:11Z +pr,187962,[inductor] Evaluate compile-time ProcessGroup attrs in overlap estimates,https://github.com/pytorch/pytorch/pull/187962,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T21:34:05Z +pr,187964,[inductor] Evaluate compile-time ProcessGroup attrs in overlap estimates,https://github.com/pytorch/pytorch/pull/187964,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T21:33:50Z +pr,187548,Update ck aiter gfx1250,https://github.com/pytorch/pytorch/pull/187548,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-23T20:56:50Z +pr,185491,Cleanup unused tensor variables,https://github.com/pytorch/pytorch/pull/185491,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-06-23T18:36:01Z +pr,187809,Make DTensor logging and export tests accelerator-agnostic,https://github.com/pytorch/pytorch/pull/187809,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor,"identify whether block is CI, merge conflict, or review gate",2026-06-23T14:25:23Z +pr,187357,Convert _scaled_grouped_mm_v2 to a structured operator,https://github.com/pytorch/pytorch/pull/187357,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-23T13:29:24Z +pr,187529,perf(dynamo): add guard lookup token plan,https://github.com/pytorch/pytorch/pull/187529,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-23T08:56:22Z +pr,187642,Add parameter-contiguous all-gather (symm_mem.all_gather_offset),https://github.com/pytorch/pytorch/pull/187642,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-23T03:28:37Z +pr,186880,Add FlexAttention shared memory compile hints,https://github.com/pytorch/pytorch/pull/186880,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T01:09:16Z +pr,187750,Tune all_gather_offset for skewed buckets,https://github.com/pytorch/pytorch/pull/187750,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-22T23:54:08Z +pr,187776,Speed up all_gather_offset writes with a shared unrolled copy helper,https://github.com/pytorch/pytorch/pull/187776,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-22T23:28:03Z +pr,176754,[CUDA Graph] Thread-safe RNG state for concurrent graph captures,https://github.com/pytorch/pytorch/pull/176754,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-22T23:27:13Z +pr,181518,[Inductor] enable MSVC precompiled headers for cpp-wrapper,https://github.com/pytorch/pytorch/pull/181518,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-22T23:14:35Z +pr,174395,Add optional out argument to F.scaled_mm,https://github.com/pytorch/pytorch/pull/174395,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-22T22:12:29Z +pr,181265,TEMP,https://github.com/pytorch/pytorch/pull/181265,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T21:25:13Z +pr,187134,[dynamo] Reuse ACT-traced graphs for resolved tensor inputs,https://github.com/pytorch/pytorch/pull/187134,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T19:50:06Z +pr,187785,[TokenSwitch] Add symmetric-memory zero-copy dispatch/combine,https://github.com/pytorch/pytorch/pull/187785,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem,"identify whether block is CI, merge conflict, or review gate",2026-06-22T15:16:43Z +pr,181221,[DCP] Fix flaky test_async_process_executor by resetting global state between tests (#181221),https://github.com/pytorch/pytorch/pull/181221,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-22T11:23:21Z +pr,181178,[Inductor] Allow max_autotune_gemm on Jetson Orin AGX,https://github.com/pytorch/pytorch/pull/181178,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T03:48:42Z +pr,186332,Improved common device test module for xpu,https://github.com/pytorch/pytorch/pull/186332,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-06-22T02:08:33Z +pr,181150,update copy_misaligned to copy_if_misaligned,https://github.com/pytorch/pytorch/pull/181150,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-21T22:52:20Z +pr,180529,[inductor] Add TORCHINDUCTOR_PERSISTENT_AUTOTUNE_DIR for persistent autotune cache (#180529),https://github.com/pytorch/pytorch/pull/180529,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-21T22:52:19Z +pr,178800,debug server tutorial,https://github.com/pytorch/pytorch/pull/178800,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-21T21:58:30Z +pr,181023,[PP] Support extra loss_fn kwargs in pipeline schedules,https://github.com/pytorch/pytorch/pull/181023,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-21T20:56:46Z +pr,181010,[inductor][comms] allow compute to borrow memory from pg_alloc mempool,https://github.com/pytorch/pytorch/pull/181010,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-21T09:48:13Z +pr,181009,[inductor][comms] pg_alloc config for strategies,https://github.com/pytorch/pytorch/pull/181009,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-21T09:48:12Z +pr,181008,[inductor][comms] memory budget for pg allocs,https://github.com/pytorch/pytorch/pull/181008,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-21T09:48:11Z +pr,186137,[dynamo] Swap @disable / @allow_in_graph order on invoke_subgraph_wrapper_unboxed,https://github.com/pytorch/pytorch/pull/186137,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-21T08:51:50Z +pr,187712,Strict pyrefly checking: round 1 (3/N),https://github.com/pytorch/pytorch/pull/187712,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-21T07:29:31Z +pr,180535,"[AC remat] Per-region independent recomputation, remove user phase annotations",https://github.com/pytorch/pytorch/pull/180535,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-21T06:14:17Z +pr,180947,[CUDACachingAllocator] Add test for expandable segment OOM stat,https://github.com/pytorch/pytorch/pull/180947,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-20T23:52:44Z +pr,180858,[inductor] symm memm ag/rs with cuStreamWrite/WaitValue32 sync,https://github.com/pytorch/pytorch/pull/180858,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-20T14:16:47Z +pr,180930,[Inductor] Lower MAX_COMPLEX_POINTWISE_CAT from 8 to 5,https://github.com/pytorch/pytorch/pull/180930,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-20T01:26:37Z +pr,180665,"[FSDP2] Fix HSDP + CPUOffloadPolicy grad corruption race, add stream-behavior tests",https://github.com/pytorch/pytorch/pull/180665,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-19T23:52:43Z +pr,178338,[inductor] add option to autotune at fusion time for mm,https://github.com/pytorch/pytorch/pull/178338,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-19T08:21:12Z +pr,187062,[inductor] Expose compile_worker_malloc_conf to tune the compile-worker allocator (#187062),https://github.com/pytorch/pytorch/pull/187062,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-19T05:59:00Z +pr,185173,Run cpp/test_jit and cpp/test_lazy sequentially in test_libtorch_jit,https://github.com/pytorch/pytorch/pull/185173,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-19T02:38:39Z +pr,174940,[BE] Delete old Metal ops,https://github.com/pytorch/pytorch/pull/174940,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:02:01Z +pr,178804,Merge clang18 Python-version Docker images into one multi-env container,https://github.com/pytorch/pytorch/pull/178804,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:01:58Z +pr,178812,[BE][MPS] Delete stale skips from common_modules.py,https://github.com/pytorch/pytorch/pull/178812,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:01:57Z +pr,180181,[CI] Split benchmark docker builds into three-phase build,https://github.com/pytorch/pytorch/pull/180181,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:01:55Z +pr,182038,[MPS] Migrate backward to MPP on MacOS-26,https://github.com/pytorch/pytorch/pull/182038,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:01:51Z +pr,184404,[BE] Exclude `c10/metal` from HEADER_ONLY wraps,https://github.com/pytorch/pytorch/pull/184404,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:01:48Z +pr,186844,Add owner headers to docs/lint/benchmark/vllm workflows,https://github.com/pytorch/pytorch/pull/186844,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:01:41Z +pr,187545,[BE] Use `std::numbers` in `c10/util/MathConstants.h`,https://github.com/pytorch/pytorch/pull/187545,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-06-19T01:01:33Z +pr,187419,Add numerics-debugging skill,https://github.com/pytorch/pytorch/pull/187419,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-18T22:27:01Z +pr,182074,[TEST][CUDA] Bump reflection_pad_2d decomp test to use `bfloat16` threshold,https://github.com/pytorch/pytorch/pull/182074,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-18T22:06:23Z +pr,181906,"[lint][RAWTHROW][3/5] aten, c10/util, distributed, inductor",https://github.com/pytorch/pytorch/pull/181906,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-18T19:35:00Z +pr,181904,[lint][RAWTHROW][1/5] Small patterns sweep: annotate + fix 22 exclude groups,https://github.com/pytorch/pytorch/pull/181904,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-18T19:34:59Z +pr,181907,[lint][RAWTHROW][4/5] Live JIT: python + runtime,https://github.com/pytorch/pytorch/pull/181907,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-06-18T19:28:04Z +pr,181905,"[lint][RAWTHROW][2/5] torch/csrc autograd, utils, profiler, fx",https://github.com/pytorch/pytorch/pull/181905,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-06-18T19:25:32Z +pr,180641,Prune binary size from flash,https://github.com/pytorch/pytorch/pull/180641,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-18T17:36:11Z +pr,170679,Fix FileBaton leaving dangling lock files that cause compilation hang…,https://github.com/pytorch/pytorch/pull/170679,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-18T12:58:30Z +pr,178150,[12/12] Enforce C++20 minimum in header guards (#178150),https://github.com/pytorch/pytorch/pull/178150,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-18T09:57:27Z +pr,175914,[Export] Fix unflatten crash when mutation intermediate has list args,https://github.com/pytorch/pytorch/pull/175914,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-17T21:17:39Z +pr,180442,fix(mps): skip nograph linear on AMD RDNA2 GPUs when input_features>=256,https://github.com/pytorch/pytorch/pull/180442,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-17T19:19:22Z +pr,187391,[export][serde] Serialize SymBoolArgument in graph input/output specs (#187391),https://github.com/pytorch/pytorch/pull/187391,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-17T05:10:33Z +pr,187430,Add Kaiser-Bessel Derived window,https://github.com/pytorch/pytorch/pull/187430,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-17T05:09:59Z +pr,181167,remove size oblivious reasoning core,https://github.com/pytorch/pytorch/pull/181167,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-17T05:09:46Z +pr,180722,Friendler error messages,https://github.com/pytorch/pytorch/pull/180722,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-17T03:48:46Z +pr,180528,Add dynamo_timed wrappers for AOT autograd compilation stages (#180528),https://github.com/pytorch/pytorch/pull/180528,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T20:31:38Z +pr,184487,dynamo: round-trip torch.cuda.stream ctx mgr across graph breaks,https://github.com/pytorch/pytorch/pull/184487,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-16T19:42:01Z +pr,187369,[Test] Add test hardware classification for test_complex.py and test_kernel_launch_checks.py,https://github.com/pytorch/pytorch/pull/187369,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-16T18:51:59Z +pr,183638,[inductor] Fuse NVFP4 nested-reduction packing,https://github.com/pytorch/pytorch/pull/183638,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T18:50:10Z +pr,178004,[ROCm] enable compiler-driven loop unrolling,https://github.com/pytorch/pytorch/pull/178004,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi300, ciflow/rocm-mi355, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-16T17:27:05Z +pr,176657,[CUDAGraph] Detect and error on freeing parent-capture tensors inside CUDA graph conditional nodes,https://github.com/pytorch/pytorch/pull/176657,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-16T16:35:47Z +pr,179702,[torch][inductor] Add dump_python_module config for AOT single-pass mode (#179702),https://github.com/pytorch/pytorch/pull/179702,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T15:36:46Z +pr,178167,triangular_solve_stub: expand leading dim support beyond dense and non-overlapping inputs,https://github.com/pytorch/pytorch/pull/178167,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-16T12:23:46Z +pr,181570,Fix batched LU factorization grain_size heuristic for large matrices (#181570),https://github.com/pytorch/pytorch/pull/181570,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-16T08:18:26Z +pr,177087,[torch][inductor] Get shape information for extern/Aten kernels with enable_kernel_profile (#177087),https://github.com/pytorch/pytorch/pull/177087,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T03:48:30Z +pr,179699,[torch][inductor] Add KernelContextGuard profiling to all kernel codegen paths (#179699),https://github.com/pytorch/pytorch/pull/179699,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T00:30:54Z +pr,179587,[pytorch] Fix resource loading for PAR compatibility using importlib.resources (#179587),https://github.com/pytorch/pytorch/pull/179587,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T00:05:33Z +pr,186102,[inductor] Fast paths in sizevars for concrete integer comparisons,https://github.com/pytorch/pytorch/pull/186102,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T22:52:58Z +pr,186101,[inductor] Avoid O(n) set union in decide_inplace_update,https://github.com/pytorch/pytorch/pull/186101,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T22:52:28Z +pr,186097,[inductor] Add early exits to can_fusion_increase_peak_memory,https://github.com/pytorch/pytorch/pull/186097,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T22:51:36Z +pr,186096,[inductor] Skip combined tiling computation for <=2D pointwise fusion,https://github.com/pytorch/pytorch/pull/186096,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T22:51:32Z +pr,187002,Use C++20 instead of c10::llvm bit helpers in c10,https://github.com/pytorch/pytorch/pull/187002,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/vllm,"identify whether block is CI, merge conflict, or review gate",2026-06-15T15:31:05Z +pr,187168,Fix accuracy degradation for baddbmm on CPU for float tensors,https://github.com/pytorch/pytorch/pull/187168,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-15T11:37:59Z +pr,171219,[DTensor] Support Partial to _StridedShard,https://github.com/pytorch/pytorch/pull/171219,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T10:36:36Z +pr,171338,[DTensor] Fully support uneven _StridedShard redistribution with device order,https://github.com/pytorch/pytorch/pull/171338,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T10:36:34Z +pr,171023,[DTensor] Support Shard to _StridedShard through all-to-all,https://github.com/pytorch/pytorch/pull/171023,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T10:36:33Z +pr,171178,[DTensor] Refactor padding/unpadding with CollectivePaddingContext,https://github.com/pytorch/pytorch/pull/171178,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T10:36:28Z +pr,186762,Fix typo in quantization documentation,https://github.com/pytorch/pytorch/pull/186762,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-15T10:35:42Z +pr,179127,Skip FBGEMM on Arm64,https://github.com/pytorch/pytorch/pull/179127,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-15T04:36:48Z +pr,179642,Cache cuSPARSELt matmul plan/descriptors in _cslt_sparse_mm (#179642),https://github.com/pytorch/pytorch/pull/179642,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-14T18:55:58Z +pr,178014,add missing to() operator which is called in benchmark (#178014),https://github.com/pytorch/pytorch/pull/178014,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-14T17:59:57Z +pr,180183,autograd: handle kMkldnn and kJagged layouts in record_stream_any_impl,https://github.com/pytorch/pytorch/pull/180183,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-14T16:59:38Z +pr,180167,Clarify that spectral_norm does not compute the true operator norm for conv layers,https://github.com/pytorch/pytorch/pull/180167,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-14T16:59:36Z +pr,187069,[torch][autograd] Fix Python refcount leaks in autograd C++ glue,https://github.com/pytorch/pytorch/pull/187069,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-14T16:30:46Z +pr,185958,torch/headeronly/util/TypeList.h: simplify contains and find_if with C++20 fold expressions and if constexpr,https://github.com/pytorch/pytorch/pull/185958,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-14T01:33:13Z +pr,185859,ATen/core/TensorBase.h: replace symint enable_if aliases with C++20 requires clauses,https://github.com/pytorch/pytorch/pull/185859,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-14T01:33:09Z +pr,174409,[inductor] Add logging to autotune output showing cache hit,https://github.com/pytorch/pytorch/pull/174409,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-13T21:52:52Z +pr,185875,c10/core/DispatchKeySet.h: add noexcept to non-constexpr query methods,https://github.com/pytorch/pytorch/pull/185875,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-13T20:33:14Z +pr,187142,Fully type-annotate torch/utils/data/graph_settings.py,https://github.com/pytorch/pytorch/pull/187142,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-13T18:40:59Z +pr,180274,[WIP][Triton] update triton pin,https://github.com/pytorch/pytorch/pull/180274,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-13T13:03:15Z +pr,178922,Add CUDA support for Lanczos interpolation mode,https://github.com/pytorch/pytorch/pull/178922,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-13T13:03:11Z +pr,185261,add meta registration for scaled_grouped_mm_v2,https://github.com/pytorch/pytorch/pull/185261,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-13T01:01:30Z +pr,186190,[spmd_types] typecheck FlexAttention,https://github.com/pytorch/pytorch/pull/186190,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-12T22:48:39Z +pr,186653,Make topk deterministic under deterministic algorithms,https://github.com/pytorch/pytorch/pull/186653,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-12T22:15:52Z +pr,177619,more ops,https://github.com/pytorch/pytorch/pull/177619,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-12T22:03:57Z +pr,177618,replicate-only ops,https://github.com/pytorch/pytorch/pull/177618,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-12T22:03:55Z +pr,186848,Use C++20 concepts (1 of N),https://github.com/pytorch/pytorch/pull/186848,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-12T21:00:38Z +pr,183617,[fsdp2] fix mem accumulation of all reduce input buffers for hsdp with mixed precision,https://github.com/pytorch/pytorch/pull/183617,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-12T20:07:44Z +pr,187189,[DO NOT MERGE][ROCm] Bump default trunk ROCm test shards to 20,https://github.com/pytorch/pytorch/pull/187189,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-12T18:47:15Z +pr,186815,Remove deprecated torch.qr,https://github.com/pytorch/pytorch/pull/186815,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-12T18:00:49Z +pr,185366,[ROCm][Distributed] Add synthesized test for sac_milp to ensure numerical correctness,https://github.com/pytorch/pytorch/pull/185366,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi200, ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-12T12:56:54Z +pr,185236,Detect fp32 precision flag leaks in TestCase and fix offenders,https://github.com/pytorch/pytorch/pull/185236,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-11T18:53:33Z +pr,179257,[c10] Fix off-by-one issue in ApproximateClock.cpp,https://github.com/pytorch/pytorch/pull/179257,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-11T17:43:17Z +pr,182964,[inductor][overlap] Wire _low_contention_all_reduce into LC FX pass,https://github.com/pytorch/pytorch/pull/182964,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:54:07Z +pr,182963,[symm_mem] Add _low_contention_all_reduce to symmetric memory,https://github.com/pytorch/pytorch/pull/182963,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:53:59Z +pr,181748,[Inductor][Triton] Enable decompose_k and use Triton for inner BMM on HIP (#181748) (#181748),https://github.com/pytorch/pytorch/pull/181748,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:51:30Z +pr,178300,[Inductor] Retry Triton compilation without pointer_range_32 on failure,https://github.com/pytorch/pytorch/pull/178300,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:51:23Z +pr,175450,[inductor] CUDAGraph P2P pool handling for symm_mem,https://github.com/pytorch/pytorch/pull/175450,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:50:40Z +pr,181007,[inductor][comms] Optional use of process_group alloc_tensor for collectives,https://github.com/pytorch/pytorch/pull/181007,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:48:36Z +pr,179548,[inductor] Add dedicated thread pool for nogil Triton compilation (#179548) (#179548),https://github.com/pytorch/pytorch/pull/179548,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:40:50Z +pr,179142,[Inductor] Add runtime assertion support to FX wrapper backend (#179142),https://github.com/pytorch/pytorch/pull/179142,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:35:11Z +pr,178083,Update inductor expected accuracy files,https://github.com/pytorch/pytorch/pull/178083,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-11T16:27:40Z +pr,177854,[Inductor] Run first cudagraph warmup in eager pool (#177854),https://github.com/pytorch/pytorch/pull/177854,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:26:02Z +pr,175584,fix aten.remainder in inductor,https://github.com/pytorch/pytorch/pull/175584,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T16:20:54Z +pr,171445,Intorduce torch.accelerator.is_graph_available to check if Graph is available,https://github.com/pytorch/pytorch/pull/171445,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-11T16:17:54Z +pr,170609,Add auto-generated cuDNN depthwise conv heuristic,https://github.com/pytorch/pytorch/pull/170609,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-11T14:58:57Z +pr,185053,[Test]Refactor test_custom_ops.py to be device agnostic,https://github.com/pytorch/pytorch/pull/185053,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-11T14:54:37Z +pr,185932,[MPS] Add `lcm` Metal kernel,https://github.com/pytorch/pytorch/pull/185932,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-11T06:58:02Z +pr,182526,[DTensor] register missing pooling backward sharding strategies,https://github.com/pytorch/pytorch/pull/182526,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T05:16:27Z +pr,184825,[DTensor] add sharding strategy for group norm backward,https://github.com/pytorch/pytorch/pull/184825,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T02:25:56Z +pr,186949,Fix for scan autograd issue,https://github.com/pytorch/pytorch/pull/186949,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-11T00:25:57Z +pr,186608,[user-streams][aoti] Add minimal AOTI user stream support (#186608),https://github.com/pytorch/pytorch/pull/186608,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-10T21:53:51Z +pr,175468,[DO NOT MERGE] Test AMD Capacity.,https://github.com/pytorch/pytorch/pull/175468,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor-rocm-mi300, ciflow/periodic-rocm-mi300, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-10T19:23:01Z +pr,113258,[pytree] support PyStructSequence types for Python pytree,https://github.com/pytorch/pytorch/pull/113258,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-10T16:06:12Z +pr,181188,[AOTInductor] Outline per-kernel-call teardown into AOTI_NOINLINE helpers,https://github.com/pytorch/pytorch/pull/181188,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-10T16:02:28Z +pr,186139,Support closed-over data-dependent indexing in scan (forward and autograd),https://github.com/pytorch/pytorch/pull/186139,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-10T15:56:56Z +pr,186882,RFC Change error order,https://github.com/pytorch/pytorch/pull/186882,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-10T06:44:40Z +pr,186010,Surface ROCm GPU device faults in TorchElastic signal-failure messages (#186010),https://github.com/pytorch/pytorch/pull/186010,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-10T06:43:26Z +pr,179400,Use std::in_range for integer range checking,https://github.com/pytorch/pytorch/pull/179400,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries, ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-10T05:02:37Z +pr,180081,[DTensor] Fix GIL-unsafe thread-local py::object in DTensor dispatch logging,https://github.com/pytorch/pytorch/pull/180081,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-10T01:27:12Z +pr,186831,Improve cudagraph stack trace coverage and assertions,https://github.com/pytorch/pytorch/pull/186831,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-10T00:10:48Z +pr,186829,Auto-promote CUDA graph static input copies,https://github.com/pytorch/pytorch/pull/186829,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-09T23:44:06Z +pr,186830,Add CUDA graph disable regions,https://github.com/pytorch/pytorch/pull/186830,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-09T23:44:04Z +pr,186828,Add CUDA graph output and input copy annotations,https://github.com/pytorch/pytorch/pull/186828,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-09T23:44:00Z +pr,186636,c10/core/impl/SizesAndStrides.h: add [[nodiscard]] to size() and isInline(),https://github.com/pytorch/pytorch/pull/186636,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-09T23:43:28Z +pr,178242,[XPU] Add grouped_mm dispatch for XPU via sycl-tla,https://github.com/pytorch/pytorch/pull/178242,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-06-09T13:01:20Z +pr,182013,Rename vec256_tests to vec_tests and fix executable search,https://github.com/pytorch/pytorch/pull/182013,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-09T10:11:09Z +pr,176600,Fix 1C1W tensor's shape and stride,https://github.com/pytorch/pytorch/pull/176600,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-09T07:32:32Z +pr,186634,c10/util/intrusive_ptr.h: replace hand-written comparison operators with operator<=>,https://github.com/pytorch/pytorch/pull/186634,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-09T06:20:33Z +pr,185101,Drop redundant tuple/list/int wrappers around tensor shape and numel,https://github.com/pytorch/pytorch/pull/185101,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-09T05:23:49Z +pr,185668,Fix typos in comments and docstrings across torch,https://github.com/pytorch/pytorch/pull/185668,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-09T04:14:09Z +pr,186609,[user-streams][aoti] Add minimal AOTI user stream support,https://github.com/pytorch/pytorch/pull/186609,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-08T18:33:12Z +pr,186458,Fix SDPA backend detection for autograd cache,https://github.com/pytorch/pytorch/pull/186458,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-08T14:10:03Z +pr,186389,[BE][Ez]: Update string_util APIs to use std::string_view more,https://github.com/pytorch/pytorch/pull/186389,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-08T14:04:31Z +pr,186586,Propagate override_cudagraphs annotation across CM-lifecycle gaps,https://github.com/pytorch/pytorch/pull/186586,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-08T10:29:44Z +pr,186334,Fix torch.kron crash on non-contiguous inputs,https://github.com/pytorch/pytorch/pull/186334,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-06T19:08:10Z +pr,185649,[Test] Make test_multihead_attention device generic for out-of-tree accelerators,https://github.com/pytorch/pytorch/pull/185649,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-05T19:09:58Z +pr,177263,Eliminate the unnecessary logic,https://github.com/pytorch/pytorch/pull/177263,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-05T10:44:50Z +pr,186304,"c10/core/DispatchKeySet: promote Full, FullAfter, Raw to enum class",https://github.com/pytorch/pytorch/pull/186304,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-05T06:35:30Z +pr,186218,Limit expandable IPC receiver VA reservation,https://github.com/pytorch/pytorch/pull/186218,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-05T03:16:09Z +pr,185956,torch/headeronly/core/TensorAccessor.h: replace typedef with using,https://github.com/pytorch/pytorch/pull/185956,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-05T01:46:11Z +pr,186253,[fsdp2] Add SPMD no-typecheck scopes to hooks,https://github.com/pytorch/pytorch/pull/186253,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-04T22:28:07Z +pr,183847,[distributed][docs] Fix group_src docstring in irecv,https://github.com/pytorch/pytorch/pull/183847,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-04T18:45:08Z +pr,185830,"Fix typos in comments, docstrings, and error messages",https://github.com/pytorch/pytorch/pull/185830,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/inductor-pallas, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-04T18:40:31Z +pr,184315,[Test] Make autograd/test_functional.py device-agnostic for out-of-tree backends.,https://github.com/pytorch/pytorch/pull/184315,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-04T11:47:37Z +pr,186048,Dedupe namespaced codegen and annotation parsing in torchgen,https://github.com/pytorch/pytorch/pull/186048,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-04T09:37:25Z +pr,186130,Add _single ProcessGroup methods and deprecate the _base aliases,https://github.com/pytorch/pytorch/pull/186130,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-04T00:09:31Z +pr,170644,[dynamo] Support dict assignment with lazy constant keys,https://github.com/pytorch/pytorch/pull/170644,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-06-03T23:37:38Z +pr,170567,[dynamo] Support dict mutation without a guard,https://github.com/pytorch/pytorch/pull/170567,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-03T23:37:29Z +pr,170092,[dynamo] Support constant ops on unrealized LazyConstantVariables,https://github.com/pytorch/pytorch/pull/170092,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-03T23:37:16Z +pr,186131,Migrate ProcessGroup method usages to the _single names,https://github.com/pytorch/pytorch/pull/186131,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-03T23:30:37Z +pr,186095,[inductor] Cache per-node tiling in fusion candidate evaluation,https://github.com/pytorch/pytorch/pull/186095,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-03T21:13:17Z +pr,186086,[PyTorch] Add missing meta kernel for aten::_sample_dirichlet,https://github.com/pytorch/pytorch/pull/186086,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-03T21:13:12Z +pr,186106,[inductor] Fast paths in sizevars for concrete integer comparisons,https://github.com/pytorch/pytorch/pull/186106,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-03T21:13:11Z +pr,184965,[Inductor] Cap combo-kernel seed autotune configs by size,https://github.com/pytorch/pytorch/pull/184965,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-03T18:27:25Z +pr,186020,[WIP] Reintroduce ptxas bundling for CUDA 13.0 binaries (revert #174716),https://github.com/pytorch/pytorch/pull/186020,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-03T16:34:21Z +pr,179547,[inductor] Add thread-based compilation support for free-threaded Python (#179547) (#179547),https://github.com/pytorch/pytorch/pull/179547,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-03T00:14:38Z +pr,185844,[CuTeDSL] Just use tensor strides directly in `_to_fake_cute_tensor`,https://github.com/pytorch/pytorch/pull/185844,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-02T22:56:35Z +pr,185592,Remove android directory,https://github.com/pytorch/pytorch/pull/185592,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-02T20:51:03Z +pr,184226,[Inductor] Allow mixed pointwise/reduction combo kernels,https://github.com/pytorch/pytorch/pull/184226,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-02T05:19:36Z +pr,185183,[Inductor] Fix combo-kernel num_warps regression on mixed reductions,https://github.com/pytorch/pytorch/pull/185183,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-02T05:19:20Z +pr,185606,Enable TF32 fpmath mode for XPU deconvolution,https://github.com/pytorch/pytorch/pull/185606,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-06-02T05:19:06Z +pr,183390,[SymmMem] Add NCCL implementation of all_to_all_vdev,https://github.com/pytorch/pytorch/pull/183390,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-01T23:47:50Z +pr,185689,Upload docs preview directly to S3 from the OSDC pod when authenticated,https://github.com/pytorch/pytorch/pull/185689,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-01T20:23:56Z +pr,185646,[AI Codemod][AsyncioGetEventLoopMigration] [asyncio-codemod] xplat/caffe2/test,https://github.com/pytorch/pytorch/pull/185646,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-06-01T18:51:45Z +pr,185432,Upgrade pyrefly 1.0.0,https://github.com/pytorch/pytorch/pull/185432,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-01T15:07:27Z +pr,181087,[MPS] RDMA for distributed MPS,https://github.com/pytorch/pytorch/pull/181087,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-01T14:59:43Z +pr,185666,Add ROCm CI workflows for mi210,https://github.com/pytorch/pytorch/pull/185666,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-30T06:58:45Z +pr,185570,tmp,https://github.com/pytorch/pytorch/pull/185570,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-29T04:59:41Z +pr,184253,[DO NOT MERGE][ROCm][CI] Test k8s dpx,https://github.com/pytorch/pytorch/pull/184253,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-29T04:59:10Z +pr,182655,Update ACL/OpenBLAS/manywheel build scripts,https://github.com/pytorch/pytorch/pull/182655,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-29T02:18:09Z +pr,178314,Separate tagging path for AC when non-strict tracing,https://github.com/pytorch/pytorch/pull/178314,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-29T01:57:49Z +pr,178416,[dynamo] Migrate allow_in_graph → nonstrict_trace in torch/ production code,https://github.com/pytorch/pytorch/pull/178416,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-29T01:57:47Z +pr,177283,RFC : WIP warmup compile with fake tensor with dynamic dims then run with real,https://github.com/pytorch/pytorch/pull/177283,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-29T01:57:45Z +pr,176989,Add supports for itertools.batched and itertools.combinations_with_re…,https://github.com/pytorch/pytorch/pull/176989,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-28T16:30:16Z +pr,182256,[MPS] enable metal performance primitives attention,https://github.com/pytorch/pytorch/pull/182256,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-05-28T03:00:04Z +pr,177568,[MTIA] Add dilation support for max_pool2d_with_indices_backward lowering (#177568),https://github.com/pytorch/pytorch/pull/177568,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-pallas, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-28T01:35:09Z +pr,184192,[Test]Refactor test/nn/test_lazy_modules.py to be device- generic by replacing hardcoded device selection with torch.accelerator API,https://github.com/pytorch/pytorch/pull/184192,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-27T14:51:08Z +pr,175746,RISC-V Vector Extension (RVV) support for ATen,https://github.com/pytorch/pytorch/pull/175746,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-27T12:59:11Z +pr,175758,Create a DeviceModule protocol,https://github.com/pytorch/pytorch/pull/175758,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-27T08:01:58Z +pr,184607,"Speed up inductor test infrastructure (~4x collection, ~1.7x execution)",https://github.com/pytorch/pytorch/pull/184607,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-27T04:01:30Z +pr,185097,Migrate deprecated torch.norm calls to torch.linalg.*,https://github.com/pytorch/pytorch/pull/185097,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-05-26T20:05:15Z +pr,185091,Prefer .view() and drop redundant .contiguous() in tensor-creator chains,https://github.com/pytorch/pytorch/pull/185091,PR blocked,63,"failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-26T13:51:40Z +pr,184509,[torch/csrc] Add bulk storage helpers; remove per-byte loops,https://github.com/pytorch/pytorch/pull/184509,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-26T13:31:38Z +pr,184657,[MPS] Tolerate lenient mixed-dtype batch_norm in InstanceNorm test,https://github.com/pytorch/pytorch/pull/184657,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-05-26T13:29:59Z +pr,165837,[OpenReg][Random] Enrich OpenReg random number generator module implementation add focused documentation,https://github.com/pytorch/pytorch/pull/165837,PR blocked,63,failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-26T13:28:08Z +pr,184254,"Fix `replication_pad{2d,3d}` meta accepting negative output dims",https://github.com/pytorch/pytorch/pull/184254,PR blocked,63,failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-25T12:12:57Z +issue,187781,[RFC] Topology-aware nn.Module primitives for memory optimization in torch.compile,https://github.com/pytorch/pytorch/issues/187781,has linked PR,62,"open PR references this issue; needs design/research label present; component/oncall labels: module: compile ux, module: memory usage, module: nn, oncall: pt2",route to design/research decision before patch planning,2026-07-07T16:34:40Z +issue,108744,switch more test cases to use MultithreadTestCase,https://github.com/pytorch/pytorch/issues/108744,has linked PR,62,open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: dtensor,review linked PR state before touching issue,2026-07-02T18:54:27Z +issue,113040,[PT2] [Hardening] Track recompiles alongside graph breaks in our actual/expected comparison CI runs,https://github.com/pytorch/pytorch/issues/113040,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-06-28T20:14:13Z +issue,115016,Enable ruff rule PLW1510 codebase wide,https://github.com/pytorch/pytorch/issues/115016,has linked PR,62,open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: lint,review linked PR state before touching issue,2026-06-27T21:33:27Z +issue,102870,Inductor: delete code that extracts out sizevars by inspecting tensor inputs to find a size that handled it,https://github.com/pytorch/pytorch/issues/102870,has linked PR,62,open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: inductor,review linked PR state before touching issue,2026-06-25T02:16:35Z +issue,108406,enhance documentation around the developer build,https://github.com/pytorch/pytorch/issues/108406,has linked PR,62,open PR references this issue; label says issue is suitable for a scoped external contribution,review linked PR state before touching issue,2026-06-18T01:02:39Z +issue,137285,Docs are little bit outdated for torch logs,https://github.com/pytorch/pytorch/issues/137285,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: logging, oncall: pt2",review linked PR state before touching issue,2026-06-17T11:05:49Z +issue,149534,Addind RoPE to pytorch core,https://github.com/pytorch/pytorch/issues/149534,has linked PR,62,open PR references this issue; needs design/research label present; component/oncall labels: module: nn,route to design/research decision before patch planning,2026-06-01T01:02:28Z +pr,161044,Thread based dataloading workers,https://github.com/pytorch/pytorch/pull/161044,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-05-21T17:05:17Z +pr,183335,Internal,https://github.com/pytorch/pytorch/pull/183335,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-05-18T18:46:39Z +issue,133443,[BE] Deduplicate auto_functionalized and triton_kernel_wrapper_functional,https://github.com/pytorch/pytorch/issues/133443,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: pt2-dispatcher, module: user triton, oncall: pt2",review linked PR state before touching issue,2026-05-15T17:45:27Z +issue,122129,Obscure error: Expected a value of type 'List[int]' for argument 'sizes' but instead found type 'immutable_list',https://github.com/pytorch/pytorch/issues/122129,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: dynamic shapes, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-12T21:07:07Z +pr,176001,[test] Resolve undefined GRAPH_EXECUTOR in test_jit_profiling.py,https://github.com/pytorch/pytorch/pull/176001,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2026-05-11T14:19:27Z +issue,48251,torch.eye(d) is slow and hogs cpu for d >= 182,https://github.com/pytorch/pytorch/issues/48251,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: cpu, module: multithreading, module: performance",review linked PR state before touching issue,2026-04-15T01:15:15Z +issue,179250,Muon: support of dynamic coefficients and normalization,https://github.com/pytorch/pytorch/issues/179250,has linked PR,62,open PR references this issue; needs design/research label present; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2026-04-03T16:58:12Z +issue,135859,"bmm, topk, cholesky, linalg.norm, max with out variants set causing recompilations in torch.compile",https://github.com/pytorch/pytorch/issues/135859,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2026-03-10T16:14:33Z +issue,116396,[dynamo] Missing support for many trivial builtin functions in operator,https://github.com/pytorch/pytorch/issues/116396,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-08T15:46:12Z +issue,92141,Improve the error message in `linalg.eigh` when using large matrices,https://github.com/pytorch/pytorch/issues/92141,has linked PR,62,"open PR references this issue; label says issue is suitable for a scoped external contribution; component/oncall labels: module: 64-bit, module: linear algebra, module: mkl",review linked PR state before touching issue,2026-02-08T19:05:21Z +pr,166813,Add min/max support for barebones uint types,https://github.com/pytorch/pytorch/pull/166813,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2025-12-29T04:45:53Z +pr,146172,Factory function support for NestedTensor,https://github.com/pytorch/pytorch/pull/146172,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,final maintainer merge/release decision,2025-04-08T21:21:14Z +pr,138626,Fix global namespace pollution in ATen/Dispatch.h,https://github.com/pytorch/pytorch/pull/138626,ready for maintainer decision,62,"approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2025-04-05T08:00:58Z +pr,138436,"Replace use of PyTorch 2.0 with torch.compile, and minor edits",https://github.com/pytorch/pytorch/pull/138436,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,final maintainer merge/release decision,2025-01-30T18:36:45Z +pr,138471,Add torch.compiler.config.cache,https://github.com/pytorch/pytorch/pull/138471,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days,final maintainer merge/release decision,2025-01-30T15:10:38Z +pr,132814,Update TwoTensor to no longer inherit requires_grad from inner tensor,https://github.com/pytorch/pytorch/pull/132814,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days,final maintainer merge/release decision,2024-10-15T13:50:45Z +pr,124490,[dynamo] Support ndarray.dtype attribute access,https://github.com/pytorch/pytorch/pull/124490,ready for maintainer decision,62,"approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk",final maintainer merge/release decision,2024-08-06T21:20:16Z +pr,120076,Change ATEN generator argument type to const std::optional&,https://github.com/pytorch/pytorch/pull/120076,ready for maintainer decision,62,"approved with no detected failing/pending checks; PR has not updated for at least 45 days; CI lane labels: ciflow/binaries, ciflow/inductor, ciflow/mps, ciflow/trunk",final maintainer merge/release decision,2024-05-25T23:18:28Z +pr,111673,[quant][bc-breaking] Remove deprecated QConfigDynamic,https://github.com/pytorch/pytorch/pull/111673,ready for maintainer decision,62,approved with no detected failing/pending checks; PR has not updated for at least 45 days,final maintainer merge/release decision,2024-01-03T07:00:57Z +pr,189313,Support aten::cumprod ONNX export via CumProd (opset 26),https://github.com/pytorch/pytorch/pull/189313,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-09T00:21:13Z +pr,186082,Improve SymIntArrayRef concrete-int diagnostics,https://github.com/pytorch/pytorch/pull/186082,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T21:23:11Z +pr,189273,Fix nn.functional.threshold OpInfo reference for strict NumPy uint8 casting,https://github.com/pytorch/pytorch/pull/189273,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-07-08T19:12:52Z +pr,186189,Report dynamo expected-failure XPASSes,https://github.com/pytorch/pytorch/pull/186189,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:17:51Z +pr,189147,Fix unload utility for newer Triton pyds,https://github.com/pytorch/pytorch/pull/189147,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:17:35Z +pr,186373,Disable duck sizing by default,https://github.com/pytorch/pytorch/pull/186373,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:35:48Z +pr,186283,Trace data_ptr through tensor constructors,https://github.com/pytorch/pytorch/pull/186283,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T15:35:09Z +pr,187609,Match CUDA threshold scalar dtype in decomp,https://github.com/pytorch/pytorch/pull/187609,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T07:46:01Z +pr,185082,Preserve lifted SymInts during strict export capture,https://github.com/pytorch/pytorch/pull/185082,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T07:45:02Z +pr,186801,Allow swap_tensors with internal TensorWeakRef,https://github.com/pytorch/pytorch/pull/186801,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/slow, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:02:14Z +pr,185460,Fix CPU perf smoketest baseline drift checks,https://github.com/pytorch/pytorch/pull/185460,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-07T19:21:55Z +pr,181860,[c10] Add InlineEventBase to unify event APIs in DeviceGuardImplInter…,https://github.com/pytorch/pytorch/pull/181860,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-07T19:12:51Z +pr,187452,Polyhedral Optimization Pass for PyTorch Inductor (Phase 1),https://github.com/pytorch/pytorch/pull/187452,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T13:19:20Z +pr,188894,Fix HAS_XPU_AND_TRITON to require triton XPU backend,https://github.com/pytorch/pytorch/pull/188894,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-07T09:01:03Z +pr,185748,[dynamo] Model RemovableHandle hook registration side effects,https://github.com/pytorch/pytorch/pull/185748,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-07T08:22:16Z +pr,185749,Fix Dynamo construction for classes with __getattribute__,https://github.com/pytorch/pytorch/pull/185749,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T07:05:24Z +pr,182768,fix(resize): rollback tensor metadata when storage resize fails,https://github.com/pytorch/pytorch/pull/182768,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-06T06:01:55Z +pr,188995,[dynamo] Allow raw unbacked SymInts in strict mode (FlexAttention BlockMask) (#187547),https://github.com/pytorch/pytorch/pull/188995,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T04:31:33Z +pr,188993,[dynamo] Preserve Event/Stream subclass identity across graph breaks (#188536),https://github.com/pytorch/pytorch/pull/188993,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T04:16:23Z +pr,188992,[inductor][XPU] Fix AMP autocast round-trip fp16 accuracy (#188261),https://github.com/pytorch/pytorch/pull/188992,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T04:16:23Z +pr,177115,Fix opcheck to detect stride mismatches on CPU tensors,https://github.com/pytorch/pytorch/pull/177115,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-07-05T23:50:50Z +pr,188350,Fix Dynamo stream/event variable python_type,https://github.com/pytorch/pytorch/pull/188350,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T10:55:08Z +pr,179443,[FSDP2] Fix HSDP O(n_layers) memory accumulation of reduce-scatter output buffers,https://github.com/pytorch/pytorch/pull/179443,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-30T22:00:07Z +pr,187461,Do Not Review: Polyhedral Optimization,https://github.com/pytorch/pytorch/pull/187461,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-perf-compare, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-29T17:14:26Z +pr,177205,Fix all_gather_into_tensor with stack-shaped output under torch.compile + dynamic shapes,https://github.com/pytorch/pytorch/pull/177205,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-29T10:04:16Z +pr,179625,Enable ruff rule PLW1510 codebase wide,https://github.com/pytorch/pytorch/pull/179625,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-27T01:01:55Z +pr,188253,Add XPU dispatch for linalg.polar,https://github.com/pytorch/pytorch/pull/188253,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-06-26T12:36:49Z +pr,188178,Guard score_mod oob indexing,https://github.com/pytorch/pytorch/pull/188178,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T22:46:40Z +pr,187883,Avoid per-symbol tuple allocation in the C++ symbolic shape guard,https://github.com/pytorch/pytorch/pull/187883,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-25T02:15:49Z +pr,188037,Fix current stream selection,https://github.com/pytorch/pytorch/pull/188037,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-24T07:31:38Z +pr,180955,[ROCm] Normalize AMDSMI UUID matching,https://github.com/pytorch/pytorch/pull/180955,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/rocm-mi300,"identify whether block is CI, merge conflict, or review gate",2026-06-21T21:58:42Z +pr,180716,Fix torch.compile addmm with beta=0 and mismatched bias,https://github.com/pytorch/pytorch/pull/180716,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-21T21:58:41Z +pr,133297,Update cmake,https://github.com/pytorch/pytorch/pull/133297,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-18T16:40:32Z +pr,186797,"Fix mm_plus_mm XPU: guard broken Triton template, enable all mm_plus_mm tests",https://github.com/pytorch/pytorch/pull/186797,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-17T16:02:37Z +pr,187490,[Dynamo] Trace raw unbacked SymInt inputs,https://github.com/pytorch/pytorch/pull/187490,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-17T03:27:22Z +pr,133289,Update native/mkldnn to native/onednn,https://github.com/pytorch/pytorch/pull/133289,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-17T01:25:48Z +pr,180707,Improve cpp_extension and drop setuptools runtime requirement,https://github.com/pytorch/pytorch/pull/180707,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers,"identify whether block is CI, merge conflict, or review gate",2026-06-16T23:41:24Z +pr,180379,[Dynamo] Inline root module forward hooks when compiling modules,https://github.com/pytorch/pytorch/pull/180379,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-14T03:48:23Z +pr,177287,"Default backward_pass_autocast to ""off"" to match eager autocast semantics",https://github.com/pytorch/pytorch/pull/177287,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-12T17:27:50Z +pr,185989,Bump onnx to 1.22.0rc1 to integrate ONNX 1.22.0 release branch,https://github.com/pytorch/pytorch/pull/185989,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-06-09T23:44:57Z +pr,185308,Align subclass output SymInt unwrapping with metadata,https://github.com/pytorch/pytorch/pull/185308,draft/noise,61,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-03T00:16:59Z +pr,183909,changing Sleef_asinhf8_u10 to std impl,https://github.com/pytorch/pytorch/pull/183909,draft/noise,61,PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-05-28T14:56:18Z +issue,167605,[Profiler] About adding the line number of calling stack from parent,https://github.com/pytorch/pytorch/issues/167605,needs triage,59,PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-20T02:13:47Z +issue,158701,"RuntimeError: optimized_plan_ INTERNAL ASSERT FAILED at ""/opt/conda/envs/bld/conda-bld/libtorch_1741644046984/work/torch/csrc/jit/runtime/profiling_graph_executor_impl.cpp"":725, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/158701,needs triage,59,"PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: third_party, oncall: jit",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-21T01:43:34Z +issue,154708,torch.quantized_batch_norm API return different output when input is nan/inf or undefined mathematical calculation method,https://github.com/pytorch/pytorch/issues/154708,needs triage,59,"PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, oncall: quantization",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-26T06:12:05Z +issue,139745,"cmake libtorch_jni.so, bug: 'fp16/psimd.h' file not found",https://github.com/pytorch/pytorch/issues/139745,needs triage,59,"PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: mobile",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-07T03:12:44Z +issue,133227,Issue Exporting PyTorch Model with pytorch3d.ops.ball_query to ONNX: Bug or Compatibility Problem?,https://github.com/pytorch/pytorch/issues/133227,needs triage,59,"PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: export",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-17T18:22:11Z +issue,106846,untimeError: The following operation failed in the TorchScript interpreter. Traceback of TorchScript (most recent call last): RuntimeError: nvrtc: error: invalid value for --gpu-architecture (-arch),https://github.com/pytorch/pytorch/issues/106846,needs triage,59,PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-21T14:50:54Z +issue,97876,Missing torch import in _contextlib.py when using torch.jit._recursive,https://github.com/pytorch/pytorch/issues/97876,needs triage,59,PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-18T13:49:08Z +issue,49157,softmax fail,https://github.com/pytorch/pytorch/issues/49157,needs triage,59,"PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn, module: viewing and reshaping, oncall: jit",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-12T03:07:10Z +issue,19106,"Performance issue with torch.jit.trace(), slow prediction in C++ (CPU)",https://github.com/pytorch/pytorch/issues/19106,needs triage,59,"triage signal present; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: performance, oncall: jit",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-20T02:38:22Z +issue,30651,[TensorBoard] The different order of import SummaryWriter may cause Segmentation fault,https://github.com/pytorch/pytorch/issues/30651,needs triage,59,PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-12-15T01:48:46Z +issue,30525,[Tensorboard] nothing appears in the projector tab,https://github.com/pytorch/pytorch/issues/30525,needs triage,59,"PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-16T15:34:09Z +issue,60930,error for using faster rcnn on c++ & GPU,https://github.com/pytorch/pytorch/issues/60930,needs triage,59,PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-30T00:03:38Z +issue,16592,[JIT] NVRTC unknown error,https://github.com/pytorch/pytorch/issues/16592,needs triage,59,PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-03-16T20:00:44Z +issue,15342,[JIT] Batch op. batch_sum return a mask which is on cpu,https://github.com/pytorch/pytorch/issues/15342,needs triage,59,PyTorch issue lacks triaged label; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-02-10T23:50:56Z +issue,188355,[Inductor][CUDA Graphs] max-autotune attempts to capture torch.linalg.eigh and crashes with cudaErrorStreamCaptureUnsupported,https://github.com/pytorch/pytorch/issues/188355,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: cuda graphs, module: inductor, module: linear algebra, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-29T17:22:27Z +issue,169688,[CPU] Stride 3 bfloat16 convolutions appear incorrect,https://github.com/pytorch/pytorch/issues/169688,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: convolution, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-27T05:30:44Z +issue,132756,JITerator: Torch.prod produces RuntimeError: CUDA driver error: invalid argument,https://github.com/pytorch/pytorch/issues/132756,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: jiterator, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-14T07:20:25Z +issue,181650,"MPS: `add_dense_scalar_cast_float` shader fails with ""Read-only bytes bound to shader argument with write access enabled"" assertion during pyannote.audio inference (M4 Pro, torch 2.11)",https://github.com/pytorch/pytorch/issues/181650,has linked PR,56,open PR references this issue; needs reproduction label present; component/oncall labels: module: mps,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-26T10:07:50Z +issue,176178,[CuTeDSL] grouped_mm illegal address from TMA descriptor init/update shape mismatch,https://github.com/pytorch/pytorch/issues/176178,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: inductor, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-19T18:06:43Z +issue,150262,Fake mode mismatch when doing nested compile + tensor subclass,https://github.com/pytorch/pytorch/issues/150262,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-15T02:08:39Z +issue,149963,[ued][wan] nonstrict_trace does not support None inputs,https://github.com/pytorch/pytorch/issues/149963,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-15T02:03:50Z +issue,148112,MLA with Learnable RoPE Tensors is Broken with Flex Attention,https://github.com/pytorch/pytorch/issues/148112,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-14T22:11:22Z +issue,137009,pre-dispatch tracing can't interpose on detach() calls from the autograd engine.,https://github.com/pytorch/pytorch/issues/137009,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: proxytensor, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-22T18:35:23Z +issue,173046,[Segfault] compilion with eager for the rnn_tanh operator leads to a segfault,https://github.com/pytorch/pytorch/issues/173046,has linked PR,56,open PR references this issue; needs reproduction label present; component/oncall labels: oncall: pt2,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-04T03:27:31Z +issue,148003,"Error : torch/utils/_sympy/interp.py:176] [0/2] failed while executing pow_by_natural([VR1, int_oo], VR[-1, -1]])",https://github.com/pytorch/pytorch/issues/148003,has linked PR,56,open PR references this issue; needs reproduction label present; component/oncall labels: oncall: pt2,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-27T14:18:26Z +issue,170696,[torch.compile] `CatchErrorsWrapper' object has no attribute '__closure__` when wrapping a fullgraph'd function in fullgraph'd module forward,https://github.com/pytorch/pytorch/issues/170696,has linked PR,56,"open PR references this issue; needs reproduction label present; component/oncall labels: module: dynamo, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-16T22:52:32Z +pr,185044,[AI SLOP] Support SymInt list custom op returns,https://github.com/pytorch/pytorch/pull/185044,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-24T07:25:53Z +pr,184756,Fix Dynamo dict equality semantics,https://github.com/pytorch/pytorch/pull/184756,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-22T20:46:16Z +pr,184758,Fix Dynamo defaultdict shallow copy semantics,https://github.com/pytorch/pytorch/pull/184758,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-22T17:43:40Z +pr,184813,Use custom Min Max in Inductor symbolic indexing,https://github.com/pytorch/pytorch/pull/184813,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/inductor-pallas, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-22T16:34:35Z +pr,184489,Forward CXX compiler to nvcc as host compiler for GCC too,https://github.com/pytorch/pytorch/pull/184489,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-22T15:43:28Z +pr,181077,Enable fbgemm on MacOS,https://github.com/pytorch/pytorch/pull/181077,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-22T15:42:12Z +pr,182781,[1/N] Use const pointers and references,https://github.com/pytorch/pytorch/pull/182781,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-22T13:34:55Z +pr,179397,Modernize map usage in ATen,https://github.com/pytorch/pytorch/pull/179397,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/mps, ciflow/trunk, ciflow/vllm","identify whether block is CI, merge conflict, or review gate",2026-05-22T11:55:18Z +pr,184812,Improve Inductor index simplification,https://github.com/pytorch/pytorch/pull/184812,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-22T03:31:02Z +pr,184801,Gate baddbmm expectedFailureDynamic on TEST_Z3,https://github.com/pytorch/pytorch/pull/184801,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-05-22T02:16:57Z +pr,178176,[DO NOT MERGE]Test New Landing Pg,https://github.com/pytorch/pytorch/pull/178176,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-21T23:38:45Z +pr,184617,Fix Dynamo sequence repeat identity semantics,https://github.com/pytorch/pytorch/pull/184617,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-21T22:15:40Z +pr,184755,Fix Dynamo sequence search semantics,https://github.com/pytorch/pytorch/pull/184755,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:53:45Z +pr,184754,Fix Dynamo reversed iterator consumption,https://github.com/pytorch/pytorch/pull/184754,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:50:47Z +pr,184765,Fix Dynamo ordered contains semantics,https://github.com/pytorch/pytorch/pull/184765,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:49:04Z +pr,184766,Fix Dynamo set discard and remove semantics,https://github.com/pytorch/pytorch/pull/184766,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:49:02Z +pr,184635,Fix Dynamo cmp_to_key list sort semantics,https://github.com/pytorch/pytorch/pull/184635,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:48:59Z +pr,184761,Fix Dynamo mutable set dict-key hashing semantics,https://github.com/pytorch/pytorch/pull/184761,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:48:03Z +pr,184759,Fix Dynamo frozenset dict-key hashing semantics,https://github.com/pytorch/pytorch/pull/184759,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:48:02Z +pr,184763,Fix Dynamo dict copy GC tracking semantics,https://github.com/pytorch/pytorch/pull/184763,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:47:57Z +pr,184762,Fix Dynamo list extend iterator semantics,https://github.com/pytorch/pytorch/pull/184762,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:47:56Z +pr,184757,Fix Dynamo dict fromkeys semantics,https://github.com/pytorch/pytorch/pull/184757,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:47:49Z +pr,184616,Fix Dynamo list constructor iteration protocol,https://github.com/pytorch/pytorch/pull/184616,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:47:36Z +pr,184622,Fix Dynamo tuple constructor semantics,https://github.com/pytorch/pytorch/pull/184622,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:47:24Z +pr,184645,Fix Dynamo defaultdict inplace union semantics,https://github.com/pytorch/pytorch/pull/184645,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:46:35Z +pr,184627,Fix Dynamo dict update semantics,https://github.com/pytorch/pytorch/pull/184627,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:46:24Z +pr,184764,Fix Dynamo iterator reduce semantics,https://github.com/pytorch/pytorch/pull/184764,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:11:34Z +pr,184760,Fix Dynamo frozenset subclass dict-key hashing semantics,https://github.com/pytorch/pytorch/pull/184760,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T21:10:39Z +pr,184300,[xplat/caffe2] Add define_aten_lite macros for selective CUDA registration composed on top of pt_operator_registry,https://github.com/pytorch/pytorch/pull/184300,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-21T18:20:49Z +pr,184536,Skip `TestFullyShardSymmMem` for NCCL < 2.28,https://github.com/pytorch/pytorch/pull/184536,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-21T17:39:56Z +pr,184673,Fix Dynamo dict fromkeys semantics,https://github.com/pytorch/pytorch/pull/184673,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T16:02:36Z +pr,184672,Fix Dynamo dict equality semantics,https://github.com/pytorch/pytorch/pull/184672,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T11:24:17Z +pr,184671,Fix Dynamo sequence search semantics,https://github.com/pytorch/pytorch/pull/184671,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T11:24:14Z +pr,184670,Fix Dynamo reversed iterator consumption,https://github.com/pytorch/pytorch/pull/184670,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T11:24:13Z +pr,184669,Fix Dynamo frozenset copy identity semantics,https://github.com/pytorch/pytorch/pull/184669,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-21T11:24:12Z +pr,184674,fixup! Add CPython Dynamo agentic loop plan,https://github.com/pytorch/pytorch/pull/184674,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-21T10:14:30Z +pr,88221,Add torch.tensor replacement and int_tensor prim,https://github.com/pytorch/pytorch/pull/88221,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-20T21:13:38Z +pr,184257,[scatter_add] Reject misaligned data_ptr / row stride in CuTeDSL override cond,https://github.com/pytorch/pytorch/pull/184257,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-20T15:21:15Z +pr,179538,Fix incorrect type annotation in Future.set_exception (#179538),https://github.com/pytorch/pytorch/pull/179538,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-20T12:16:47Z +pr,184389,[ROCm] Fix test_origami autotune comparison after PR #181617 cap,https://github.com/pytorch/pytorch/pull/184389,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/inductor-rocm-mi200, ciflow/inductor-rocm-mi300, ciflow/inductor-rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-05-20T12:07:31Z +pr,181617,"Speed up inductor test infrastructure (~4x collection, ~1.7x execution)",https://github.com/pytorch/pytorch/pull/181617,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-20T12:07:30Z +pr,184401,Revert D105603961,https://github.com/pytorch/pytorch/pull/184401,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-19T22:13:51Z +pr,183592,[ROCm] op benchmark: use noble nightly image,https://github.com/pytorch/pytorch/pull/183592,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/nightly, ciflow/op-benchmark","identify whether block is CI, merge conflict, or review gate",2026-05-19T15:50:40Z +pr,184236,[inductor][overlap] Enable simple_overlap by default,https://github.com/pytorch/pytorch/pull/184236,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-19T15:36:52Z +pr,57772,Deprecate torch.svd and change svd -> linalg_svd,https://github.com/pytorch/pytorch/pull/57772,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-19T12:26:08Z +pr,184307,Update SDPA error diagnostics,https://github.com/pytorch/pytorch/pull/184307,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-19T06:20:24Z +pr,183394,lint: ignore B904 for vendored code,https://github.com/pytorch/pytorch/pull/183394,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-19T01:17:01Z +pr,183651,[OSDC] Migrate h100-distributed.yml to OSDC (ARC) via dial-up pattern,https://github.com/pytorch/pytorch/pull/183651,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-18T17:35:59Z +pr,184194,[inductor] Add simple_overlap pass for defensive comm/compute overlap,https://github.com/pytorch/pytorch/pull/184194,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-18T15:44:01Z +pr,176857,[XPU] test_indexing test_reductions: Remove obsolete XPU skip guards after verification,https://github.com/pytorch/pytorch/pull/176857,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-05-18T09:17:54Z +pr,183953,[DO NOT MERGE] Point CI workflows to test-infra IPv6 branch,https://github.com/pytorch/pytorch/pull/183953,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/binaries, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-18T01:16:18Z +pr,183978,"[docs] Add .. deprecated:: directives to torch.{cuda,cpu}.amp public APIs",https://github.com/pytorch/pytorch/pull/183978,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-17T23:50:17Z +pr,183455,Port D104346887/PR 182675 for index large select (#183455),https://github.com/pytorch/pytorch/pull/183455,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-16T10:01:13Z +pr,183993,"Back out ""Port D104346887/PR 182675 for index large select""",https://github.com/pytorch/pytorch/pull/183993,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-16T04:29:38Z +pr,179096,[Inductor] Add _scaled_mm_v2 lowering,https://github.com/pytorch/pytorch/pull/179096,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:31:44Z +pr,174893,[Inductor][NVGEMM] Enable Epilogue Fusions,https://github.com/pytorch/pytorch/pull/174893,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:31:05Z +pr,181909,[Inductor][Triton] Forward hint_override in fused Triton compile,https://github.com/pytorch/pytorch/pull/181909,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:26:00Z +pr,182730,[DTensor] Preserve placement for aten.clone.default,https://github.com/pytorch/pytorch/pull/182730,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-16T00:03:33Z +pr,183940,[Native DSL] Vendor kernelagent-oink library,https://github.com/pytorch/pytorch/pull/183940,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-15T21:47:45Z +pr,183941,[Native DSL] Oink-based cuteDSL RMSNorm,https://github.com/pytorch/pytorch/pull/183941,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-15T18:53:58Z +pr,182196,[dynamo][dicts] Fix delitem for dicts,https://github.com/pytorch/pytorch/pull/182196,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-15T07:54:17Z +pr,183487,Update XNNPACK submodule to 0b66c9f,https://github.com/pytorch/pytorch/pull/183487,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-15T04:27:59Z +pr,183086,Remove unused 'type: ignore' suppressions in dynamo/inductor/functorch,https://github.com/pytorch/pytorch/pull/183086,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/inductor-pallas","identify whether block is CI, merge conflict, or review gate",2026-05-15T03:01:29Z +pr,180545,fix AOTInductor fails to compile dynamic-shape CUDA gather pattern with ValueError: The argument '((0)) + 48' is not comparable,https://github.com/pytorch/pytorch/pull/180545,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-15T03:00:27Z +pr,183747,Adding new test_c10d_ops_privateuse1.py variant for distributed testing,https://github.com/pytorch/pytorch/pull/183747,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-05-14T23:11:21Z +pr,170807,Adding Glossary,https://github.com/pytorch/pytorch/pull/170807,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-14T16:11:27Z +pr,183262,[CUDA] Expose cudaHostGetDevicePointer in cudart bindings,https://github.com/pytorch/pytorch/pull/183262,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-14T16:06:33Z +pr,182845,[AI Codemod][RlAutotuneFbsourceWearablesMemory] PackedLinearWeightsQnnp::apply_dynamic (qlinear_dynamic.cpp),https://github.com/pytorch/pytorch/pull/182845,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-14T16:03:43Z +pr,182846,fix: safely handle non-standard boolean values in the CUDA kernel,https://github.com/pytorch/pytorch/pull/182846,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-14T16:03:33Z +pr,182179,[CI] Floor known-slow file times when stale per-config data under-shards,https://github.com/pytorch/pytorch/pull/182179,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/rocm-mi200, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-14T05:59:22Z +pr,181281,[dynamo] Refactor IsInstanceBuiltinVariable,https://github.com/pytorch/pytorch/pull/181281,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-14T05:59:16Z +pr,181338,[dynamo] Refactor TypeBuiltinVariable,https://github.com/pytorch/pytorch/pull/181338,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-14T05:58:47Z +pr,181337,[dynamo] Refactor StrBuiltinVariable,https://github.com/pytorch/pytorch/pull/181337,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-14T05:58:46Z +pr,182598,[dynamo] Refactor builtins init static type check using _fn,https://github.com/pytorch/pytorch/pull/182598,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-14T05:58:45Z +pr,179863,Teach CSAN to detect caching allocator memory reuse races,https://github.com/pytorch/pytorch/pull/179863,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-14T05:15:44Z +pr,183611,[inductor] Record output alias for mutates_and_returns_first_arg fallbacks,https://github.com/pytorch/pytorch/pull/183611,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-13T23:28:38Z +pr,178360,Add device/backend agnostic support for distributed c10d/common tests,https://github.com/pytorch/pytorch/pull/178360,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-05-13T23:26:57Z +pr,183384,[inductor] preserve scan carry strides with shape padding,https://github.com/pytorch/pytorch/pull/183384,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-13T21:58:12Z +pr,183550,[AI Codemod][RlAutotuneFbsourceWearablesMemory] PackedLinearWeightsQnnp::apply_dynamic (qlinear_dynamic.cpp),https://github.com/pytorch/pytorch/pull/183550,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-13T18:32:00Z +pr,182398,"Revert ""[dynamo-wrapped] Skip all Dynamo wrapped tests for numpy (#182240)""",https://github.com/pytorch/pytorch/pull/182398,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-13T07:22:42Z +pr,183418,[Avocado][MI350X] Non-overlay compatibility for new AITER/CK overlays (#183418),https://github.com/pytorch/pytorch/pull/183418,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-12T18:10:44Z +pr,183404,test/dynamo: add CPython coverage planning docs,https://github.com/pytorch/pytorch/pull/183404,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-12T14:57:19Z +pr,182274,[docs] Generate doc for values in ScalingType and SwizzleType enum,https://github.com/pytorch/pytorch/pull/182274,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-12T13:55:53Z +pr,177726,Introduce test skip markers for Sandcastle,https://github.com/pytorch/pytorch/pull/177726,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-12T11:27:16Z +pr,162706,Fix accelerator-related skip-decorators for tests,https://github.com/pytorch/pytorch/pull/162706,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-12T11:27:07Z +pr,179139,Report ACL version info,https://github.com/pytorch/pytorch/pull/179139,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-12T09:36:29Z +pr,183035,Fix arange int64 fractional-step zero handling on CPU and CUDA.,https://github.com/pytorch/pytorch/pull/183035,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-11T21:51:59Z +pr,181435,Add operator benchmark monitoring skill for daily regression detection,https://github.com/pytorch/pytorch/pull/181435,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-11T19:05:46Z +pr,177867,Add ASIMD_BF16 Vectorized class specialisation,https://github.com/pytorch/pytorch/pull/177867,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-11T16:41:56Z +pr,181359,S390x: fixes for docker image build,https://github.com/pytorch/pytorch/pull/181359,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-11T15:18:31Z +pr,182900,test: relax tolerance of test_Embedding_discontiguous_cuda,https://github.com/pytorch/pytorch/pull/182900,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-11T14:47:56Z +pr,183254,Add RISC-V native fp16 conversion paths,https://github.com/pytorch/pytorch/pull/183254,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-11T14:42:55Z +pr,181875,[ROCm][release/2.11] Fix test_transformers.py test suite for ROCm,https://github.com/pytorch/pytorch/pull/181875,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-11T11:46:55Z +pr,183105,Fix formatting in extensions.json,https://github.com/pytorch/pytorch/pull/183105,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-05-10T20:02:14Z +pr,178245,Replace c10::call_once with static initialization,https://github.com/pytorch/pytorch/pull/178245,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-05-10T02:45:08Z +pr,174025,Add test for CosineAnnealingLR when T_max = 0,https://github.com/pytorch/pytorch/pull/174025,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-05T15:53:30Z +pr,161254,Add `zero_grad` param for `optimizer.step`,https://github.com/pytorch/pytorch/pull/161254,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-04-29T01:54:20Z +pr,170249,[do not review] Expose API to create ownership token for autograd.Function grad_fn,https://github.com/pytorch/pytorch/pull/170249,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-03-27T18:49:13Z +pr,170250,[do not review] Use new GradientEdge creation in PP,https://github.com/pytorch/pytorch/pull/170250,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-03-24T19:05:48Z +pr,130556,Allow pipeline stage to take non-tensor args and kwargs,https://github.com/pytorch/pytorch/pull/130556,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-03-21T22:38:41Z +pr,167894,[inductor] split reduction even if all reads are broadcasted,https://github.com/pytorch/pytorch/pull/167894,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-03-06T22:02:57Z +pr,166732,Introducing statefulness for single process dataloader,https://github.com/pytorch/pytorch/pull/166732,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-02-26T08:44:11Z +pr,157193,[c10d] Call destroy_process_group upon interpreter exit,https://github.com/pytorch/pytorch/pull/157193,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2026-02-23T23:30:39Z +pr,140613,"Do not merge, do not close",https://github.com/pytorch/pytorch/pull/140613,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-01-11T02:18:04Z +pr,165800,[flex_attention]: adding a heuristic for USE_TMA,https://github.com/pytorch/pytorch/pull/165800,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2025-12-30T18:42:02Z +pr,164128,Consistently use c10_ovrsource in arvr mode everywhere,https://github.com/pytorch/pytorch/pull/164128,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2025-12-29T04:45:55Z +pr,162033,[RFC] Switch all of c10d to use .h instead of .hpp,https://github.com/pytorch/pytorch/pull/162033,PR blocked,55,"failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2025-12-21T03:17:44Z +pr,160888,Allow branching on whether you are in AC region with is_checkpoint_enabled,https://github.com/pytorch/pytorch/pull/160888,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2025-12-05T01:43:53Z +pr,160878,[WIP] Introduce CachingDeviceAllocatorInterface as a base impl,https://github.com/pytorch/pytorch/pull/160878,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2025-11-27T01:35:47Z +pr,160872,[WIP] [3/N] Introduce a generic CachingDeviceAllocatorImpl for cross backend use,https://github.com/pytorch/pytorch/pull/160872,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2025-11-27T01:35:25Z +pr,160638,[WIP] [2/N] Introduce a generic CachingDeviceAllocatorImpl for cross backend use,https://github.com/pytorch/pytorch/pull/160638,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2025-11-27T01:35:02Z +pr,160182,[WIP] [1/N] Introduce a generic CachingDeviceAllocatorImpl for cross backend use,https://github.com/pytorch/pytorch/pull/160182,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2025-11-27T01:34:25Z +pr,148900,[RFC][BE] assume error checking is on by default (#141914),https://github.com/pytorch/pytorch/pull/148900,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,"identify whether block is CI, merge conflict, or review gate",2025-11-22T02:02:11Z +pr,148668,Fix CUPTI lookup to include target directory,https://github.com/pytorch/pytorch/pull/148668,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2025-10-23T14:20:34Z +pr,112441,[RFC] Tensordict integration,https://github.com/pytorch/pytorch/pull/112441,PR blocked,55,failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2025-05-28T21:34:32Z +pr,167224,[FlopCounter] Add skip_unsupported parameter for graceful handling of unsupported HOPs,https://github.com/pytorch/pytorch/pull/167224,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-09T01:53:29Z +pr,188006,[ROCm] Fix torch.pdist correctness on ROCm; re-enable test_pdist_norm_large,https://github.com/pytorch/pytorch/pull/188006,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T23:43:22Z +pr,181895,[CUDA] Fix linspace integer dtype precision mismatch with CPU,https://github.com/pytorch/pytorch/pull/181895,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/h100-symm-mem",wait for contributor update,2026-07-08T21:06:36Z +pr,189132,Register sspaddmm.out for XPU backend,https://github.com/pytorch/pytorch/pull/189132,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T20:39:59Z +pr,184121,Improve persistent tiled reduction defaults,https://github.com/pytorch/pytorch/pull/184121,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-perf-compare, ciflow/torchtitan",wait for contributor update,2026-07-08T19:39:04Z +pr,185150,Preserve OrderedDict outputs in FX tracing,https://github.com/pytorch/pytorch/pull/185150,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T19:12:06Z +pr,184319,[inductor] Align split reductions for tiled pointwise fusion,https://github.com/pytorch/pytorch/pull/184319,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T19:02:41Z +pr,185433,Fix FakeTensorMode handling for Inductor dynamic scalars,https://github.com/pytorch/pytorch/pull/185433,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T18:14:18Z +pr,184858,Support runtime ConstDict lookups for dynamic keys,https://github.com/pytorch/pytorch/pull/184858,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T17:31:22Z +pr,189215,[MPS] Add `largeTensorTest` to `test_group_norm_backward_large_input`,https://github.com/pytorch/pytorch/pull/189215,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-08T17:28:51Z +pr,185984,Fix Dynamo out= resize for internal tensors,https://github.com/pytorch/pytorch/pull/185984,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T17:00:33Z +pr,184963,Fix FakeTensor shallow copy state,https://github.com/pytorch/pytorch/pull/184963,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T16:46:07Z +pr,184223,Run Inductor fallbacks below autograd,https://github.com/pytorch/pytorch/pull/184223,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T16:17:34Z +pr,189038,[MPS] Fix uint8 casting in MPS `copy_` kernels,https://github.com/pytorch/pytorch/pull/189038,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-08T16:12:55Z +pr,184910,Reject direct aten.set in Dynamo,https://github.com/pytorch/pytorch/pull/184910,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T16:05:11Z +pr,185069,Fix AOTI Windows cross-compile package loading,https://github.com/pytorch/pytorch/pull/185069,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T16:01:31Z +pr,186897,Warn when custom op int schema specializes SymInts,https://github.com/pytorch/pytorch/pull/186897,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T15:53:41Z +pr,184380,Avoid fp32 materialization in checkpoint recompute,https://github.com/pytorch/pytorch/pull/184380,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T15:49:47Z +pr,185976,Fix Dynamo autograd.Function backward grad mode tracing,https://github.com/pytorch/pytorch/pull/185976,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T15:18:05Z +pr,186510,Fix Dynamo low-precision addc opmath,https://github.com/pytorch/pytorch/pull/186510,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T15:12:30Z +pr,184345,Use freezing compiler for traced inference graphs,https://github.com/pytorch/pytorch/pull/184345,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T13:19:57Z +pr,186397,Keep constant-false torch._check in compiled graph,https://github.com/pytorch/pytorch/pull/186397,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T11:51:17Z +pr,186420,Run torchrec tests in Dynamo CI,https://github.com/pytorch/pytorch/pull/186420,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T11:35:42Z +pr,172074,Fixes grad_dtype property being lost in fsdp,https://github.com/pytorch/pytorch/pull/172074,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T10:18:22Z +pr,184694,Fix AOTAutograd input overlap guards,https://github.com/pytorch/pytorch/pull/184694,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T09:35:56Z +pr,184573,Fix compile support for CTC loss tensor lengths,https://github.com/pytorch/pytorch/pull/184573,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T09:32:28Z +pr,184602,Preserve frozen ShapeEnv during fake tensor guards,https://github.com/pytorch/pytorch/pull/184602,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T09:05:31Z +pr,184632,Fix CUDA fake strides for mixed dtype pointwise ops,https://github.com/pytorch/pytorch/pull/184632,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T09:01:26Z +pr,184639,Support None returns from torch.cond,https://github.com/pytorch/pytorch/pull/184639,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T08:58:55Z +pr,185496,Fix export autograd.grad saved tensor tracing,https://github.com/pytorch/pytorch/pull/185496,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T08:44:24Z +pr,186444,Fix Adam capturable zero-lr NaNs,https://github.com/pytorch/pytorch/pull/186444,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T07:45:11Z +pr,185585,Fix mark_unbacked stride recompiles,https://github.com/pytorch/pytorch/pull/185585,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T07:13:24Z +pr,183871,[inductor] Match eval BatchNorm CUDA numerics,https://github.com/pytorch/pytorch/pull/183871,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T07:10:13Z +pr,184009,Handle SymPy boolean Inductor graph inputs,https://github.com/pytorch/pytorch/pull/184009,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T06:43:00Z +pr,188966,Add operator.concat/iconcat support to dynamo builtins,https://github.com/pytorch/pytorch/pull/188966,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T06:17:57Z +pr,185064,Add unroll option to scan,https://github.com/pytorch/pytorch/pull/185064,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T06:15:05Z +pr,186764,fix: Use typing_extensions.TypeAliasType for better reexport of `__module__`,https://github.com/pytorch/pytorch/pull/186764,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T05:48:41Z +pr,187908,[fix] torch.where silently overflows fp16 scalars (issue #187429),https://github.com/pytorch/pytorch/pull/187908,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,wait for contributor update,2026-07-08T05:20:33Z +pr,173948,Fixes LazyConvTranspose modules when used with torch.compile(dynamic=True),https://github.com/pytorch/pytorch/pull/173948,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T04:51:20Z +pr,176072,Fix Triton launcher arg mismatch silent failure,https://github.com/pytorch/pytorch/pull/176072,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-08T04:50:40Z +pr,184938,Fix AOTAutograd forward output overwritten by backward mutation,https://github.com/pytorch/pytorch/pull/184938,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T04:26:00Z +pr,183881,Fix online softmax epilogue fusion for T5,https://github.com/pytorch/pytorch/pull/183881,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T04:09:02Z +pr,183944,Fix compiled autograd clear saved tensors semantics,https://github.com/pytorch/pytorch/pull/183944,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T04:01:54Z +pr,183954,Disallow mid-trace reference opaque creation,https://github.com/pytorch/pytorch/pull/183954,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-08T03:56:03Z +pr,186366,Support vmapped compiled autograd.Functions,https://github.com/pytorch/pytorch/pull/186366,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T03:44:03Z +pr,185764,Recommend nonstrict_trace for skipped C extensions,https://github.com/pytorch/pytorch/pull/185764,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T02:39:09Z +pr,183969,Fix CSE across unrelated mutation regions,https://github.com/pytorch/pytorch/pull/183969,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T02:36:28Z +pr,184958,Prevent escaped vjp TensorWrappers from reaching AOTAutograd,https://github.com/pytorch/pytorch/pull/184958,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T02:20:24Z +pr,180689,[ROCm] Handle profiler trace format differences in combo kernel tests,https://github.com/pytorch/pytorch/pull/180689,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/rocm-mi300",wait for contributor update,2026-07-08T02:18:49Z +pr,184961,Fix packed RNN export fake metadata,https://github.com/pytorch/pytorch/pull/184961,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-08T02:15:57Z +pr,184970,Preserve pin_memory in generated view copy kernels,https://github.com/pytorch/pytorch/pull/184970,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T02:05:15Z +pr,184051,Honor custom op layout constraints in fallbacks,https://github.com/pytorch/pytorch/pull/184051,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-08T01:40:54Z +pr,185017,Fix OpOverloadPacket tracing during initialization,https://github.com/pytorch/pytorch/pull/185017,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T00:45:01Z +pr,184481,Guard synthetic-base input alias layouts,https://github.com/pytorch/pytorch/pull/184481,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T00:38:07Z +pr,185035,Make symbolic formatting guard data-dependent scalars,https://github.com/pytorch/pytorch/pull/185035,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T00:23:05Z +pr,185043,Functionalize mutations in deferred local_map,https://github.com/pytorch/pytorch/pull/185043,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-08T00:08:21Z +pr,184944,Make OpaqueBase usable as a pybind base,https://github.com/pytorch/pytorch/pull/184944,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor",wait for contributor update,2026-07-07T23:19:14Z +pr,185124,Fix prims.broadcast_in_dim functionalization,https://github.com/pytorch/pytorch/pull/185124,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T23:15:13Z +pr,185171,Document activation memory budget config,https://github.com/pytorch/pytorch/pull/185171,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-07T22:07:35Z +pr,185274,Fix ci expected accuracy updates without ClickHouse credentials,https://github.com/pytorch/pytorch/pull/185274,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T21:49:29Z +pr,187161,[FSDP] Cache post-forward DeviceMesh to avoid duplicate NCCL communicators per layer,https://github.com/pytorch/pytorch/pull/187161,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T20:54:49Z +pr,185075,Improve FakeTensorMode deepcopy error for real tensors,https://github.com/pytorch/pytorch/pull/185075,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-07T20:32:17Z +pr,182972,added graph break on exception based on if user code would catch it,https://github.com/pytorch/pytorch/pull/182972,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T19:53:45Z +pr,188605,Fix infinite recursion with pybind11 enum conversions,https://github.com/pytorch/pytorch/pull/188605,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T19:50:37Z +pr,185299,Run CSE in AOT inference graphs,https://github.com/pytorch/pytorch/pull/185299,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T19:20:47Z +pr,184052,Preserve CIA ops in AOT joint export,https://github.com/pytorch/pytorch/pull/184052,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T19:06:21Z +pr,185347,Support torch.compile under user FakeTensorMode,https://github.com/pytorch/pytorch/pull/185347,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T18:51:04Z +pr,185353,Normalize export state names for subclass decomposition,https://github.com/pytorch/pytorch/pull/185353,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T18:42:42Z +pr,185892,Improve export device mismatch errors,https://github.com/pytorch/pytorch/pull/185892,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T17:52:17Z +pr,184061,Fix fused moving avg observer functional meta,https://github.com/pytorch/pytorch/pull/184061,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T17:40:44Z +pr,184985,Fix nested compile handling in AOT joint export,https://github.com/pytorch/pytorch/pull/184985,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T17:10:59Z +pr,188994,[dynamic shapes] Preserve algebraic relationships in foreign unbacked SymInt transfer (#188723),https://github.com/pytorch/pytorch/pull/188994,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T17:08:55Z +pr,184312,Fix efficient zero tensor handling in torch.compile,https://github.com/pytorch/pytorch/pull/184312,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-07T17:08:27Z +pr,188948,Fix triangular_solve for sparse CPU tensors on non-MKL platforms,https://github.com/pytorch/pytorch/pull/188948,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T17:06:02Z +pr,184016,Fix Inductor DCE for fallback alias mutations,https://github.com/pytorch/pytorch/pull/184016,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T16:32:58Z +pr,185499,Fix unbacked memo reuse during decomposition retrace,https://github.com/pytorch/pytorch/pull/185499,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T16:25:41Z +pr,184091,Handle cudagraph liveness changes during replay,https://github.com/pytorch/pytorch/pull/184091,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T15:58:50Z +pr,184092,Skip tiny cudagraph partitions in max-autotune,https://github.com/pytorch/pytorch/pull/184092,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-07T15:54:29Z +pr,185690,Optimize AOT input view mutations,https://github.com/pytorch/pytorch/pull/185690,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-pallas",wait for contributor update,2026-07-07T15:06:00Z +pr,189043,Preload full bundled cuDNN set with RTLD_GLOBAL to prevent sublibrary version mismatch,https://github.com/pytorch/pytorch/pull/189043,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/binaries,wait for contributor update,2026-07-07T13:27:21Z +pr,185759,Improve float32 flex attention backward tiling,https://github.com/pytorch/pytorch/pull/185759,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T13:11:26Z +pr,184130,Avoid torch imports in Inductor compile sidecar,https://github.com/pytorch/pytorch/pull/184130,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T12:24:05Z +pr,183904,Fix CPU addcmul tensor-value decomposition numerics,https://github.com/pytorch/pytorch/pull/183904,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-07T11:39:48Z +pr,186522,Trace Tensor.requires_grad_ in pre-dispatch make_fx,https://github.com/pytorch/pytorch/pull/186522,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T11:04:20Z +pr,187633,Enhance documentation around the developer build,https://github.com/pytorch/pytorch/pull/187633,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-07T10:54:24Z +pr,184153,[Inductor] Clone cudagraph-owned grads before replay,https://github.com/pytorch/pytorch/pull/184153,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T10:47:22Z +pr,184161,Fix Inductor lazy negative input loads,https://github.com/pytorch/pytorch/pull/184161,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-pallas, ciflow/torchtitan",wait for contributor update,2026-07-07T10:43:09Z +pr,185899,Fix Dynamo data mutation on functorch tensors,https://github.com/pytorch/pytorch/pull/185899,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T08:50:40Z +pr,184336,fix RPATH in libtorch extraction from wheel,https://github.com/pytorch/pytorch/pull/184336,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/binaries,wait for contributor update,2026-07-07T08:21:16Z +pr,186160,Fix FakeTensor forward AD batch_norm meta rewrapping,https://github.com/pytorch/pytorch/pull/186160,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T06:32:34Z +pr,184621,Fix Dynamo torch.Size from tensor shape inputs,https://github.com/pytorch/pytorch/pull/184621,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T05:17:12Z +pr,184997,Fix TVM backend scalar tensor inputs,https://github.com/pytorch/pytorch/pull/184997,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T03:30:38Z +pr,178493,[c10d][NCCL] avoid reusing scalable init store keys across communicat…,https://github.com/pytorch/pytorch/pull/178493,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-07T02:20:16Z +pr,181106,Fix torch.compile crash with inference_mode and view ops in forward (#180823),https://github.com/pytorch/pytorch/pull/181106,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk",wait for contributor update,2026-07-06T22:13:47Z +pr,188249,[inductor] Fix bool equality value range analysis,https://github.com/pytorch/pytorch/pull/188249,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-06T21:30:13Z +pr,184303,Fix pattern matcher replacement graph cleanup safety,https://github.com/pytorch/pytorch/pull/184303,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-06T18:21:54Z +pr,184310,Fix flex attention shared memory config fallback,https://github.com/pytorch/pytorch/pull/184310,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-06T17:10:03Z +pr,189019,[MPS] Add torch.mps.MetalGraph capture/replay API,https://github.com/pytorch/pytorch/pull/189019,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-06T14:46:00Z +pr,188748,[MPS] Make F.linear honor PYTORCH_MPS_PREFER_METAL,https://github.com/pytorch/pytorch/pull/188748,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-06T13:00:28Z +pr,181665,Fix UBSan in weight_norm CPU kernel for empty input (#181510),https://github.com/pytorch/pytorch/pull/181665,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-06T02:20:21Z +pr,182276,[inductor] Unwrap ShapeAsConstantBuffer in tensor_constructor size,https://github.com/pytorch/pytorch/pull/182276,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-06T01:09:40Z +pr,188952,Fix C2872 'std' ambiguous symbol when nvcc compiles compiled_autograd.h on Windows extension builds,https://github.com/pytorch/pytorch/pull/188952,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-04T21:09:29Z +pr,188473,[MPS] Add complex64 support to linalg.lu_factor (unblocks complex det/slogdet/lu),https://github.com/pytorch/pytorch/pull/188473,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-04T19:12:03Z +pr,187477,Add context manager support to torch.dtype,https://github.com/pytorch/pytorch/pull/187477,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-04T15:13:41Z +pr,178591,fixes:division by zero check for inductor,https://github.com/pytorch/pytorch/pull/178591,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-04T02:27:03Z +pr,177457,[DTensor] Coalesce collectives for multi-arg redistribution,https://github.com/pytorch/pytorch/pull/177457,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-03T21:14:35Z +pr,174718,[inductor] fix allocation with deterministic guard,https://github.com/pytorch/pytorch/pull/174718,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-02T20:30:43Z +pr,188801,[MPS] Leak MetalShaderLibrary bundled singleton to avoid exit-time destructor crashes,https://github.com/pytorch/pytorch/pull/188801,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-02T19:51:02Z +pr,180675,Add test for autograd.Function fused fwd/bwd pattern (issue #180642),https://github.com/pytorch/pytorch/pull/180675,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-02T18:43:42Z +pr,187572,[Metal][ops] add EmbeddingBag offsets validation (#170370),https://github.com/pytorch/pytorch/pull/187572,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-02T15:35:51Z +pr,180353,[MPS] Replace isin MPSGraph implementation with native Metal kernels,https://github.com/pytorch/pytorch/pull/180353,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-07-01T17:56:01Z +pr,185831,Use minimum-norm subgradient for clamp/clamp_min/clamp_max at boundary,https://github.com/pytorch/pytorch/pull/185831,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-07-01T01:54:03Z +pr,188297,"Validate target range [0, 1] in binary_cross_entropy_with_logits",https://github.com/pytorch/pytorch/pull/188297,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-01T01:53:17Z +pr,186542,Support joint export with unused parameter grads,https://github.com/pytorch/pytorch/pull/186542,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-01T01:47:15Z +pr,188092,Fix cartesian_prod returning inconsistent shape for single tensor input,https://github.com/pytorch/pytorch/pull/188092,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,wait for contributor update,2026-06-29T09:22:11Z +pr,187955,Make expandable-segment Handle a move-only RAII owner and exception-safe,https://github.com/pytorch/pytorch/pull/187955,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-29T01:21:43Z +pr,187316,Support save_on_cpu offloading under first-order torch.func transforms,https://github.com/pytorch/pytorch/pull/187316,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-27T05:24:16Z +pr,183980,[cpu][cuda] Extend native_group_norm_backward to handle grad from all outputs,https://github.com/pytorch/pytorch/pull/183980,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/trunk",wait for contributor update,2026-06-25T20:55:12Z +pr,187663,Avoid oversized storage for reduced mse_loss / smooth_l1_loss,https://github.com/pytorch/pytorch/pull/187663,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-25T07:25:40Z +pr,185519,Fix ZeroTensor meta redispatch for nested hessian compile,https://github.com/pytorch/pytorch/pull/185519,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-06-24T21:19:41Z +pr,174862,[FSDP2] Fix to differing gradient computation graphs in different ranks,https://github.com/pytorch/pytorch/pull/174862,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-24T19:51:16Z +pr,185163,[aot_autograd] Handle None in saved tensors for LSTM/RNN backward; enable allow_rnn by default,https://github.com/pytorch/pytorch/pull/185163,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/b200, ciflow/docker, ciflow/dtensor, ciflow/h100",wait for contributor update,2026-06-24T19:14:57Z +pr,186766,fix: [BE] Deduplicate auto_functionalized and triton_kernel_wrapper_functi...,https://github.com/pytorch/pytorch/pull/186766,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-06-23T19:23:30Z +pr,183508,Chain-based dispatch for custom_op fast path,https://github.com/pytorch/pytorch/pull/183508,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,wait for contributor update,2026-06-22T22:53:24Z +pr,172727,Fix torch.topk out variant recompilations by adding meta function for…,https://github.com/pytorch/pytorch/pull/172727,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-06-22T15:22:24Z +pr,187260,Fix static cudnn building for cudnn v9,https://github.com/pytorch/pytorch/pull/187260,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/binaries,wait for contributor update,2026-06-22T00:32:51Z +pr,187247,Type _add_docstr so type checkers see spectral ops as callable,https://github.com/pytorch/pytorch/pull/187247,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-21T16:33:20Z +pr,175490,Fix CPU `NotImplementedError` for bf16/fp16 antialiased upsampling (`interpolate`),https://github.com/pytorch/pytorch/pull/175490,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-20T01:26:27Z +pr,185355,[pipelining] Fix backward hook accumulation on recv buffer activations,https://github.com/pytorch/pytorch/pull/185355,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-19T07:58:02Z +pr,186445,[AOTI] Avoid serializing full storage for shared-storage constants,https://github.com/pytorch/pytorch/pull/186445,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-18T17:12:23Z +pr,187323,Fix Tensor.to() ignoring memory_format for non-contiguous tensors,https://github.com/pytorch/pytorch/pull/187323,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-18T07:16:17Z +pr,187558,Fix torch.sign to propagate NaN instead of returning 0,https://github.com/pytorch/pytorch/pull/187558,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-18T05:05:22Z +pr,182794,[inductor][rocm] Fix AOTInductor autotune int64 Triton kernel test and enable FP8 platform-aware handling,https://github.com/pytorch/pytorch/pull/182794,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk",wait for contributor update,2026-06-17T21:24:04Z +pr,181465,Fix torch.fx.Tracer.record_stack_traces regression (user frames filtered by internal PyTorch frames),https://github.com/pytorch/pytorch/pull/181465,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-17T21:23:24Z +pr,186579,Fix mse_loss/smooth_l1_loss allocating oversized storage for scalar output,https://github.com/pytorch/pytorch/pull/186579,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-17T21:10:39Z +pr,187154,Use TORCH_INSTALL_LIB_DIR instead of hardcoded path 'lib',https://github.com/pytorch/pytorch/pull/187154,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for contributor update,2026-06-17T21:06:49Z +pr,184680,Inductor Lowering of Trunc Div should use div_rn,https://github.com/pytorch/pytorch/pull/184680,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-17T12:01:27Z +pr,186543,Fix Windows debug build issues,https://github.com/pytorch/pytorch/pull/186543,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/xpu",wait for contributor update,2026-06-17T05:11:10Z +pr,185645,Add batched small-matrix CUDA kernel for linalg.eig,https://github.com/pytorch/pytorch/pull/185645,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-16T15:55:48Z +pr,187106,[inductor] decompose broadcast-bias baddbmm into bmm + pointwise add,https://github.com/pytorch/pytorch/pull/187106,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-15T22:15:26Z +pr,185977,Fix index_add out-of-bounds parity between eager and inductor,https://github.com/pytorch/pytorch/pull/185977,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/mps, ciflow/torchtitan",wait for contributor update,2026-06-15T18:47:12Z +pr,175964,Add validation for embedding_bag offsets to prevent illegal memory access,https://github.com/pytorch/pytorch/pull/175964,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-15T10:35:27Z +pr,185157,[Inductor] Fix CPU bool argmin/argmax incorrect indices,https://github.com/pytorch/pytorch/pull/185157,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-15T08:42:07Z +pr,184249,Fix extern autotune tensor kwargs,https://github.com/pytorch/pytorch/pull/184249,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-11T23:28:09Z +pr,186776,Fix __qualname__ for OpOverload and OpOverloadPacket,https://github.com/pytorch/pytorch/pull/186776,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-11T01:17:39Z +pr,185765,[Inductor] Fix correctness discrepancy in torch.minimum and torch.maximum with signed zeros (-0.0 vs 0.0),https://github.com/pytorch/pytorch/pull/185765,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-10T19:47:04Z +pr,182980,Clear exception tracebacks to avoid reference cycles,https://github.com/pytorch/pytorch/pull/182980,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/docker, ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-10T14:48:54Z +pr,186205,[dynamo] Add as_strided_ handler to fix metadata propagation (#185888),https://github.com/pytorch/pytorch/pull/186205,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-08T13:32:56Z +pr,186079,Add test cases for 0-sized tensor padding,https://github.com/pytorch/pytorch/pull/186079,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-05T11:10:05Z +pr,183691,[Test] Add tolerance and precision override mechanisms for parameteried tests,https://github.com/pytorch/pytorch/pull/183691,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,wait for contributor update,2026-06-04T08:32:40Z +pr,185927,Fix clamp boundary subgradients,https://github.com/pytorch/pytorch/pull/185927,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-03T22:01:22Z +pr,182618,[Profiler] Forward declare nlohmann::json in public header,https://github.com/pytorch/pytorch/pull/182618,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-06-02T01:31:40Z +pr,185361,[pooling] Reject negative divisor_override in avg_pool2d/3d,https://github.com/pytorch/pytorch/pull/185361,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan",wait for contributor update,2026-06-01T01:37:44Z +pr,185263,Fix torch.compile normal tensor std validation,https://github.com/pytorch/pytorch/pull/185263,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-05-31T05:25:25Z +pr,185048,[Bugfix] fix `torch.library._register_fake` crashes when any unrelated `LazyLoader` module in `sys.modules` fails to import,https://github.com/pytorch/pytorch/pull/185048,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-05-30T16:17:35Z +pr,177823,[Inductor] Cap num_warps for non-1x1 Triton conv kernels to fix incorrect results,https://github.com/pytorch/pytorch/pull/177823,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-29T21:44:31Z +pr,183535,[MPS] fix older gpu gen mac matmul failures,https://github.com/pytorch/pytorch/pull/183535,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/mps,wait for contributor update,2026-05-29T16:12:23Z +pr,185256,Fix PendingUnbackedSymbolNotFound for tensor-derived slice bounds,https://github.com/pytorch/pytorch/pull/185256,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-29T15:38:10Z +pr,181526,Replace obsolete TORCHDYNAMO_DYNAMIC_SHAPES env var in torch.compile FAQ,https://github.com/pytorch/pytorch/pull/181526,waiting on contributor,53,review decision requests changes; PR closes or references open issue numbers,wait for contributor update,2026-05-29T01:46:26Z +pr,183023,[Inductor] Filter triton configs by launch geometry in triton heuristics,https://github.com/pytorch/pytorch/pull/183023,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/xpu",wait for contributor update,2026-05-28T19:17:24Z +pr,184531,[Inductor] Tolerance increase for test_bmm_large_batch_dynamic_cuda,https://github.com/pytorch/pytorch/pull/184531,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-rocm-mi200, ciflow/inductor-rocm-mi300, ciflow/inductor-rocm-mi355",wait for contributor update,2026-05-27T12:58:53Z +pr,179830,Fix upsample_nearest3d stride mismatch under torch.compile (#179272),https://github.com/pytorch/pytorch/pull/179830,waiting on contributor,53,"review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/inductor-pallas, ciflow/torchtitan",wait for contributor update,2026-05-26T17:22:28Z +pr,184483,Make DTensor.local_map pytree-aware on in/out_placements,https://github.com/pytorch/pytorch/pull/184483,draft/noise,53,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-20T08:45:10Z +pr,183029,[make_fx] Fix wrong 2nd-order gradients by decomposing detach to alias,https://github.com/pytorch/pytorch/pull/183029,draft/noise,53,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-11T17:39:34Z +pr,175969,[inductor][xpu] Fix convolution_backward accuracy for mixed channels-last / contiguous inputs,https://github.com/pytorch/pytorch/pull/175969,draft/noise,53,"PR is draft; failed checks or blocked/unstable merge state; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-05-11T07:32:58Z +issue,184878,Division by Zero Floating Point Exception in monitor aggregation,https://github.com/pytorch/pytorch/issues/184878,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2026-06-01T17:14:12Z +issue,172011,torch.pixel_shuffle CPU-GPU crash with massive upscale factor,https://github.com/pytorch/pytorch/issues/172011,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2026-01-09T20:15:00Z +issue,128691,Segmentation fault (core dumped) in `torch._remove_batch_dim`,https://github.com/pytorch/pytorch/issues/128691,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2026-01-07T18:55:32Z +issue,150187,at::BlasBackend::Ck does not handle all ROCm BLAS gpus,https://github.com/pytorch/pytorch/issues/150187,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, module: rocm",review early; this can change release or regression risk,2026-01-07T18:05:07Z +issue,68320,Potential race conditions between multiple workers trying to download and cache the same file in torch.hub.load_state_dict_from_url and torch.hub.download_url_to_file <- duplicate dataset/model downloads across DDP workers,https://github.com/pytorch/pytorch/issues/68320,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: hub,review early; this can change release or regression risk,2026-01-02T10:36:36Z +issue,154312,torch.logdet produces incorrect results for singular matrices on CUDA vs CPU,https://github.com/pytorch/pytorch/issues/154312,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda, module: linear algebra, module: nans and infs",review early; this can change release or regression risk,2025-12-30T20:35:27Z +issue,171284,GRU: Segmentation fault (core dumped),https://github.com/pytorch/pytorch/issues/171284,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking, module: rnn",review early; this can change release or regression risk,2025-12-25T12:05:09Z +issue,164402,NCCL-2.28.3 build locally is unusable on H100,https://github.com/pytorch/pytorch/issues/164402,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci, module: nccl, module: regression",review early; this can change release or regression risk,2025-12-24T22:50:02Z +issue,48365,Custom exception for out of memory,https://github.com/pytorch/pytorch/issues/48365,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label,review early; this can change release or regression risk,2025-12-22T16:03:51Z +issue,67237,[docs] Google finds docs pages that give 404,https://github.com/pytorch/pytorch/issues/67237,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: docs,review early; this can change release or regression risk,2025-12-16T22:18:05Z +issue,164957,Something broke `from functorch import make_fx` on main,https://github.com/pytorch/pytorch/issues/164957,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: functorch",review early; this can change release or regression risk,2025-12-11T00:53:30Z +issue,112583,Nesting no_grad in autocast causes backwards graph to be (partially) lost outside of no_grad,https://github.com/pytorch/pytorch/issues/112583,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: autograd",review early; this can change release or regression risk,2025-12-09T20:12:38Z +issue,16897,Implement `numpy.random.choice` equivalent,https://github.com/pytorch/pytorch/issues/16897,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: numpy",review early; this can change release or regression risk,2025-11-23T10:33:22Z +issue,166004,[CUDA] torch.linalg.eigh input size reduced since PyTorch 2.8.0,https://github.com/pytorch/pytorch/issues/166004,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: regression",review early; this can change release or regression risk,2025-11-18T17:43:56Z +issue,157097,Segmentation fault in torch.repeat_interleave,https://github.com/pytorch/pytorch/issues/157097,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2025-11-17T17:07:38Z +issue,146551,libtorch_cuda_linalg.so: undefined symbol: mkl_lapack_dsbrdbn on a source built PyTorch 2.6.0 with USE_STATIC_MKL=1 on CUDA platform,https://github.com/pytorch/pytorch/issues/146551,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkl, module: regression",review early; this can change release or regression risk,2025-11-10T23:21:29Z +issue,141358,PyTorch crashes with `mkl_lapack_dgetrf` if `from_numpy and set_num_threads` are used,https://github.com/pytorch/pytorch/issues/141358,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: intel, module: linear algebra, module: mkl",review early; this can change release or regression risk,2025-11-09T14:40:12Z +issue,165123,https://github.com/pytorch/pytorch/pull/128406 caused ISA-specific symbols leak,https://github.com/pytorch/pytorch/issues/165123,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: crash, module: regression",review early; this can change release or regression risk,2025-11-07T22:32:18Z +issue,158154,`torch.native_channel_shuffle` throws FPE when `groups` > size of the second dimension of the input tensor,https://github.com/pytorch/pytorch/issues/158154,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-11-06T18:50:50Z +issue,163675,Torch CPU wheels no longer available for download from PyTorch website,https://github.com/pytorch/pytorch/issues/163675,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,review early; this can change release or regression risk,2025-11-05T02:06:14Z +issue,124423,`__getitem__` fails to vmap for one dimensional tensors,https://github.com/pytorch/pytorch/issues/124423,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,review early; this can change release or regression risk,2025-11-03T14:53:56Z +issue,163539,conv3d bfloat16 wrong result,https://github.com/pytorch/pytorch/issues/163539,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda, module: cudnn, module: regression",review early; this can change release or regression risk,2025-10-29T07:43:45Z +issue,154887,MPS batch_norm mixed dtype failure,https://github.com/pytorch/pytorch/issues/154887,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps, module: third_party",review early; this can change release or regression risk,2025-10-28T07:40:42Z +issue,164562,Update FlashInfer version used by vLLM tests on PyTorch CI to v0.4.1,https://github.com/pytorch/pytorch/issues/164562,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: vllm",review early; this can change release or regression risk,2025-10-23T22:14:17Z +issue,153203,FPE Crash in torch.fmod with large negative integer modulo value,https://github.com/pytorch/pytorch/issues/153203,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: int overflow, module: python frontend",review early; this can change release or regression risk,2025-10-23T06:53:49Z +issue,157461,vLLM tests failing in torch 2.8rc but passing with torch 2.7,https://github.com/pytorch/pytorch/issues/157461,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: regression, module: vllm",review early; this can change release or regression risk,2025-10-15T20:47:15Z +issue,160296,[inductor][cpu] Performance regression in 2025-08-02 nightly release,https://github.com/pytorch/pytorch/issues/160296,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: performance, module: regression, oncall: cpu inductor",review early; this can change release or regression risk,2025-10-13T07:41:34Z +issue,58005,torch.distributed.nn.all_reduce incorrectly scales the gradient,https://github.com/pytorch/pytorch/issues/58005,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), oncall: distributed",review early; this can change release or regression risk,2025-09-27T19:02:25Z +issue,116928,"Mistral works on (2.0.1+cu117) but ""CUDA out of memory"" on (2.1.2+cu121)",https://github.com/pytorch/pytorch/issues/116928,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: regression",review early; this can change release or regression risk,2025-09-26T23:33:38Z +issue,107699,index.Tensor_out & index_put.out errors or segfaults with indices list containing only null tensors,https://github.com/pytorch/pytorch/issues/107699,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,review early; this can change release or regression risk,2025-09-26T23:31:14Z +issue,119261,Performance Concern about CrossEntropyLoss,https://github.com/pytorch/pytorch/issues/119261,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn, module: performance",review early; this can change release or regression risk,2025-09-26T23:25:54Z +issue,81180,MPS: Placeholder tensor is empty!,https://github.com/pytorch/pytorch/issues/81180,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-09-26T18:51:25Z +issue,155393,crash in torch.histc,https://github.com/pytorch/pytorch/issues/155393,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: crash, module: edge cases",review early; this can change release or regression risk,2025-09-21T11:43:37Z +issue,162884,setup.py develop --cmake no longer works,https://github.com/pytorch/pytorch/issues/162884,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: regression",review early; this can change release or regression risk,2025-09-21T02:42:52Z +issue,146828,"On Linux, passing torch.Generator to multiprocessing.Process crashes for forkserver and spawn start method",https://github.com/pytorch/pytorch/issues/146828,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: random",review early; this can change release or regression risk,2025-09-18T09:13:40Z +issue,161760,Invalid __shared__ read of size 8 bytes in torch.histc,https://github.com/pytorch/pytorch/issues/161760,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: sanitizers",review early; this can change release or regression risk,2025-09-04T13:21:21Z +issue,154424,`Segmentation fault` in `torch.matmul` and `torch.sparse.addmm`,https://github.com/pytorch/pytorch/issues/154424,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: empty tensor, module: error checking",review early; this can change release or regression risk,2025-09-02T06:29:00Z +issue,161874,Floating point exception in torch.nn.quantized.Conv2d.register_full_backward_hook,https://github.com/pytorch/pytorch/issues/161874,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-09-01T06:52:40Z +issue,161358,periodic tests are failing to build torchvision,https://github.com/pytorch/pytorch/issues/161358,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: ci, module: cpp-extensions",review early; this can change release or regression risk,2025-08-26T20:17:49Z +issue,161419,mobilenetv3_large_100 aot_eager accuracy failure,https://github.com/pytorch/pytorch/issues/161419,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review early; this can change release or regression risk,2025-08-26T18:16:14Z +issue,155981,Floating point exception (core dumped) in torch.nn.functional.fold,https://github.com/pytorch/pytorch/issues/155981,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-08-20T15:30:13Z +issue,109014,Cannot install torchmetrics - ERROR 403,https://github.com/pytorch/pytorch/issues/109014,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,review early; this can change release or regression risk,2025-08-13T16:26:32Z +issue,154882,MPS max_pool2d_with_indices failure: destination values and indices length mismatch along axis,https://github.com/pytorch/pytorch/issues/154882,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps, module: third_party",review early; this can change release or regression risk,2025-08-12T16:49:25Z +issue,159713,[MPS Crash] Assertion failed in _getLSTMGradKernelDAGObject using LSTM on macOS with MPS backend,https://github.com/pytorch/pytorch/issues/159713,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps, module: rnn",review early; this can change release or regression risk,2025-08-04T10:45:24Z +issue,121711,torch.func.grad hangs on nested differentiation,https://github.com/pytorch/pytorch/issues/121711,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,review early; this can change release or regression risk,2025-07-29T14:18:38Z +issue,152824,`mypy` stage of `lintrunner -a` has intermittent but continuing crashes,https://github.com/pytorch/pytorch/issues/152824,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: flaky-tests, module: lint",review early; this can change release or regression risk,2025-07-29T14:18:13Z +issue,158728,[CI] Inductor-perf-nighly-macos started to fail to install torchvision,https://github.com/pytorch/pytorch/issues/158728,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: regression",review early; this can change release or regression risk,2025-07-21T22:45:29Z +issue,115347,[Functorch] vmap over index_select expands the output,https://github.com/pytorch/pytorch/issues/115347,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,review early; this can change release or regression risk,2025-07-21T16:59:18Z +issue,153050,Process never ends when sending tensors through multiprocessing queues in Python 3.12+ with filesystem strategy,https://github.com/pytorch/pytorch/issues/153050,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: macos, module: multiprocessing, module: regression",review early; this can change release or regression risk,2025-07-11T00:55:09Z +issue,141836,StrideAPI caused regression in channels-last logic,https://github.com/pytorch/pytorch/issues/141836,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: mps, module: regression",review early; this can change release or regression risk,2025-07-10T16:12:25Z +issue,155201,Importing xgboost before torch + openmp causes seg fault,https://github.com/pytorch/pytorch/issues/155201,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: macos, module: openmp, module: third_party",review early; this can change release or regression risk,2025-07-09T15:41:41Z +issue,152535,optree package status in PyTorch,https://github.com/pytorch/pytorch/issues/152535,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: pytree,review early; this can change release or regression risk,2025-07-09T07:43:18Z +issue,34058,libtorch.so file size is very large,https://github.com/pytorch/pytorch/issues/34058,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp, oncall: mobile",review early; this can change release or regression risk,2025-07-09T02:31:02Z +issue,157098,Floating point exception in torch.nn.functional.conv_transpose3d,https://github.com/pytorch/pytorch/issues/157098,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2025-07-04T01:31:49Z +issue,52663,code linking to libtorch cannot use thrust/cub functions,https://github.com/pytorch/pytorch/issues/52663,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cuda",review early; this can change release or regression risk,2025-06-29T15:38:49Z +issue,142032,"Python 3.12 ""from functorch.einops import rearrange"" fails with ""RuntimeError: First class dim doesn't work with python 3.12""",https://github.com/pytorch/pytorch/issues/142032,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: regression",review early; this can change release or regression risk,2025-06-27T14:35:32Z +issue,148699,CUDA 12.6 Inductor accuracy test failures,https://github.com/pytorch/pytorch/issues/148699,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",review early; this can change release or regression risk,2025-06-20T17:25:31Z +issue,155607,CUDA 12.6->12.8 slow and periodic failures,https://github.com/pytorch/pytorch/issues/155607,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: regression",review early; this can change release or regression risk,2025-06-17T20:26:47Z +issue,143123,MPS incompatibility: Calls into the C++ engine to run the backward pass,https://github.com/pytorch/pytorch/issues/143123,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: mps, module: regression",review early; this can change release or regression risk,2025-06-08T00:40:35Z +issue,149325,☂️ MPS support for large tensors,https://github.com/pytorch/pytorch/issues/149325,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: crash, module: mps",review early; this can change release or regression risk,2025-06-06T15:28:24Z +issue,155293,[MPS] 5D 4Mln rng attempts fail with internal error,https://github.com/pytorch/pytorch/issues/155293,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: crash, module: mps, module: random",review early; this can change release or regression risk,2025-06-06T00:20:54Z +issue,76191,"[Feature request] Exclusive prefix sum, `torch.cumsum(input, dim=0, exclusive=True)`",https://github.com/pytorch/pytorch/issues/76191,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nestedtensor, module: reductions",review early; this can change release or regression risk,2025-06-05T03:18:33Z +issue,152292,`torch==2.6` broke `nn.Module.dtype` typing,https://github.com/pytorch/pytorch/issues/152292,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, module: typing",review early; this can change release or regression risk,2025-06-05T03:13:23Z +issue,146286,Non-blocking GPU to CPU copy of complex numbers with the different conj status produces wrong results,https://github.com/pytorch/pytorch/issues/146286,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: complex, module: correctness (silent)",review early; this can change release or regression risk,2025-05-31T02:29:21Z +issue,99625,Conda Pytorch set processor affinity to the first physical core after fork,https://github.com/pytorch/pytorch/issues/99625,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: dependency bug, module: intel, module: mkl",review early; this can change release or regression risk,2025-05-30T15:14:50Z +issue,154422,`Segmentation fault` in `torch.fx.experimental.partitioner_utils.map_arg`,https://github.com/pytorch/pytorch/issues/154422,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-05-27T18:25:18Z +issue,127622,"Segfault, possibly due to recursion limit",https://github.com/pytorch/pytorch/issues/127622,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: multiprocessing",review early; this can change release or regression risk,2025-05-25T01:58:53Z +issue,152747,torch.compile causes stride mismatch in SDPA with non-contiguous query in torch 2.7,https://github.com/pytorch/pytorch/issues/152747,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, oncall: pt2",review early; this can change release or regression risk,2025-05-22T15:38:08Z +issue,153961,[Windows][CPU] Root cause the VS2022 illegal instruction,https://github.com/pytorch/pytorch/issues/153961,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: windows",review early; this can change release or regression risk,2025-05-22T08:09:15Z +issue,153733,[SOURCE] links for patch releases lead to 2.7.1,https://github.com/pytorch/pytorch/issues/153733,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: doc infra, module: docs",review early; this can change release or regression risk,2025-05-19T17:05:46Z +issue,153231,`torch.native_channel_shuffle` crashes with Floating Point Exception when given large integer parameter,https://github.com/pytorch/pytorch/issues/153231,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: nn",review early; this can change release or regression risk,2025-05-14T18:33:23Z +issue,152913,Segmentation fault (core dumped) in torch.nn.functional.max_unpool2d,https://github.com/pytorch/pytorch/issues/152913,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2025-05-12T20:15:02Z +issue,102904,Unable to checkpoint model and optimizer state when using Hybrid Sharding Strategy,https://github.com/pytorch/pytorch/issues/102904,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed checkpointing",review early; this can change release or regression risk,2025-05-09T00:24:54Z +issue,152979,FPE when using `torch.lcm_` with int32 tensor and int16 scalar,https://github.com/pytorch/pytorch/issues/152979,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2025-05-08T08:28:54Z +issue,2575,ImportError: dlopen: cannot load any more object with static TLS,https://github.com/pytorch/pytorch/issues/2575,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: build, module: crash",review early; this can change release or regression risk,2025-05-07T15:56:54Z +issue,152777,Segmentation fault (core dumped) in torch.nn.functional.alpha_dropout,https://github.com/pytorch/pytorch/issues/152777,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: empty tensor, module: error checking, oncall: quantization",review early; this can change release or regression risk,2025-05-04T19:12:50Z +issue,151680,"undefined reference to `at::native::blas_impl::fp16_gemv_notrans(int, int, float, c10::Half const*, int, c10::Half const*, int, float, c10::Half*, int)'",https://github.com/pytorch/pytorch/issues/151680,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: regression",review early; this can change release or regression risk,2025-04-18T18:10:08Z +issue,139298,CUDNN sdp attention causes loss explosion,https://github.com/pytorch/pytorch/issues/139298,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: sdpa",review early; this can change release or regression risk,2025-04-17T22:03:17Z +issue,19739,Importing open3d after PyTorch causes free(): invalid pointer,https://github.com/pytorch/pytorch/issues/19739,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: pybind",review early; this can change release or regression risk,2025-04-13T20:13:49Z +issue,149201,Segment fault when used with FAISS on arm,https://github.com/pytorch/pytorch/issues/149201,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: crash, module: macos",review early; this can change release or regression risk,2025-04-04T16:20:29Z +issue,150388,double free or corruption (out) in torch.as_strided_scatter,https://github.com/pytorch/pytorch/issues/150388,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-04-02T05:32:38Z +issue,147362,Floating point exception (core dumped) in torch.nn.functional.conv3d,https://github.com/pytorch/pytorch/issues/147362,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mkldnn, module: nn",review early; this can change release or regression risk,2025-04-01T01:56:23Z +issue,149658,unexpected kwarg 'no_python_abi_suffix' when installing stable version of pytorch with sam2,https://github.com/pytorch/pytorch/issues/149658,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions",review early; this can change release or regression risk,2025-03-31T20:27:40Z +issue,149820,`Segmentation fault` when using `torch.sparse.mm` with `torch.sparse_csr_tensor`,https://github.com/pytorch/pytorch/issues/149820,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: sparse",review early; this can change release or regression risk,2025-03-24T19:32:47Z +issue,128819,Environment gating of CUDA_VISIBLE_DEVICES returns a CUDA initialization error.,https://github.com/pytorch/pytorch/issues/128819,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: regression",review early; this can change release or regression risk,2025-03-23T00:14:11Z +issue,141083,ARM build failed with recent XNNPACK update: third_party/XNNPACK/src/reference/unary-elementwise.cc:125:14: error: invalid ‘static_cast’ from type ‘xnn_bfloat16’ to type ‘_Float16’,https://github.com/pytorch/pytorch/issues/141083,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: regression, module: xnnpack",review early; this can change release or regression risk,2025-02-25T18:17:42Z +issue,146798,Torch 2.6 Unexpected Graph Break with SubConfigProxy,https://github.com/pytorch/pytorch/issues/146798,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: graph breaks, module: regression, oncall: pt2",review early; this can change release or regression risk,2025-02-11T19:27:50Z +issue,146857,Segmentation Fault nn/test_convolution.py test_ConvTranspose2d_output_size_downsample_upsample on Aarch64,https://github.com/pytorch/pytorch/issues/146857,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: crash",review early; this can change release or regression risk,2025-02-11T09:17:11Z +issue,92776,segfault when running torch.lu_unpack,https://github.com/pytorch/pytorch/issues/92776,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: linear algebra",review early; this can change release or regression risk,2025-02-10T15:05:21Z +issue,146787,Floating Point Exception in `torch.ops.aten.native_channel_shuffle` with `groups=0`,https://github.com/pytorch/pytorch/issues/146787,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: empty tensor, module: error checking",review early; this can change release or regression risk,2025-02-09T20:17:30Z +issue,146789,Segmentation Fault in `torch.ops.aten.linalg_eigvals` After Invalid `unfold_copy`,https://github.com/pytorch/pytorch/issues/146789,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: linear algebra",review early; this can change release or regression risk,2025-02-09T20:17:01Z +issue,146791,Floating Point Exception in `torch.ops.aten.unfold_backward` with Specific Input,https://github.com/pytorch/pytorch/issues/146791,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-02-09T20:16:28Z +issue,146793,Segmentation Fault in `torch.ops.aten.as_strided` with Large `storage_offset`,https://github.com/pytorch/pytorch/issues/146793,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-02-09T20:14:42Z +issue,146794,Segmentation Fault in `torch.as_strided_copy` with Large `storage_offset`,https://github.com/pytorch/pytorch/issues/146794,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-02-09T20:14:31Z +issue,113663,FP8 types should not participate in type promotion and should have no math ops defined on them,https://github.com/pytorch/pytorch/issues/113663,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),review early; this can change release or regression risk,2025-02-08T02:16:18Z +issue,86120,TransformerEncoder src_key_padding_mask does not work in eval(),https://github.com/pytorch/pytorch/issues/86120,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,review early; this can change release or regression risk,2025-02-07T12:55:25Z +issue,145259,No Range Check for `storage_offset` in `as_strided` Function,https://github.com/pytorch/pytorch/issues/145259,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2025-02-04T21:19:53Z +issue,48559,[FR] torch.load should support loading directly to pinned/shared memory,https://github.com/pytorch/pytorch/issues/48559,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: serialization",review early; this can change release or regression risk,2025-02-03T18:44:54Z +issue,97109,"""Adaptive pool MPS: input sizes must be divisible by output sizes"", I keep getting this error even when I try to adjust for size",https://github.com/pytorch/pytorch/issues/97109,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-02-01T11:10:55Z +issue,144563,Multiple tests not run / run as no-ops by `run_test.py`,https://github.com/pytorch/pytorch/issues/144563,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: tests,review early; this can change release or regression risk,2025-01-30T23:08:34Z +issue,134323,"Fix `torch.stft` and `torch.istft` when using `center=False`, non-rectangular windows and `win_length==hop_length`",https://github.com/pytorch/pytorch/issues/134323,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,review early; this can change release or regression risk,2025-01-30T10:51:56Z +issue,145604,`torch.ops.aten.copy` causes SIGSEGV when handling sparse CSR tensors with invalid metadata,https://github.com/pytorch/pytorch/issues/145604,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: sparse",review early; this can change release or regression risk,2025-01-24T19:33:15Z +issue,145065,SIGFPE error when passing very large kernel_size to `avg_pool1d`,https://github.com/pytorch/pytorch/issues/145065,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2025-01-21T15:40:46Z +issue,145071,Illegal memory access and segmentation fault due to large `storage_offset` in `as_strided`,https://github.com/pytorch/pytorch/issues/145071,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2025-01-21T15:40:35Z +issue,145267,`torch.ops.aten.embedding_dense_backward` Crashes with Out-of-Bounds Indices On CPU,https://github.com/pytorch/pytorch/issues/145267,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: embedding, module: error checking",review early; this can change release or regression risk,2025-01-21T15:03:02Z +issue,92033,Unable to use vmap atop torch.distribution functionality,https://github.com/pytorch/pytorch/issues/92033,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,review early; this can change release or regression risk,2025-01-17T01:05:43Z +issue,143480,Segmentation fault (core dumped) in `conv1d`,https://github.com/pytorch/pytorch/issues/143480,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: error checking, module: nn",review early; this can change release or regression risk,2025-01-14T15:00:04Z +issue,143481,Segmentation fault (core dumped) in `conv3d`,https://github.com/pytorch/pytorch/issues/143481,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: crash, module: error checking, module: nn",review early; this can change release or regression risk,2025-01-14T09:36:55Z +issue,104342,"Segmentation error while using F.cross_entropy with mps(for code that works fine with device= ""cpu"")",https://github.com/pytorch/pytorch/issues/104342,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-01-09T19:27:40Z +issue,103966,'MPS' Issue Running HuggingFace Transformer Pix2Struct Model,https://github.com/pytorch/pytorch/issues/103966,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-01-09T19:26:24Z +issue,102911,PackedSequences on MPS accelerator yields `grad_y` missing or crashes the kernel.,https://github.com/pytorch/pytorch/issues/102911,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-01-09T19:23:50Z +issue,119367,Updating a slice on CPU with a slice from MPS results in an error,https://github.com/pytorch/pytorch/issues/119367,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-01-09T19:04:02Z +issue,96056,MPS device throws error for `F.adaptive_avg_pool2d`,https://github.com/pytorch/pytorch/issues/96056,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-01-09T18:49:22Z +issue,133625,Binary Cross Entropy crashes kernel on apple silicon with singleton dimensions,https://github.com/pytorch/pytorch/issues/133625,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-01-09T18:18:29Z +issue,84489,"For PyTorch Nightly, failure when changing MPS device to CPU after PYTORCH_ENABLE_MPS_FALLBACK occurs.",https://github.com/pytorch/pytorch/issues/84489,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2025-01-09T18:15:02Z +issue,13402,batch_norm doesn't bump version counter of running stats,https://github.com/pytorch/pytorch/issues/13402,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",review early; this can change release or regression risk,2024-12-19T22:02:31Z +issue,143485,Aborted (core dumped) in `mkldnn_rnn_layer`,https://github.com/pytorch/pytorch/issues/143485,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking, module: mkldnn",review early; this can change release or regression risk,2024-12-19T08:52:34Z +issue,143490,Segmentation fault (core dumped) in `replication_pad2d`,https://github.com/pytorch/pytorch/issues/143490,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2024-12-18T22:09:38Z +issue,143483,Segmentation fault (core dumped) in `embedding_bag.padding_idx`,https://github.com/pytorch/pytorch/issues/143483,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: embedding, module: empty tensor, module: error checking",review early; this can change release or regression risk,2024-12-18T16:02:40Z +issue,143482,Segmentation fault (core dumped) in `embedding_backward`,https://github.com/pytorch/pytorch/issues/143482,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: embedding, module: empty tensor, module: error checking",review early; this can change release or regression risk,2024-12-18T16:02:13Z +issue,142209,Segmentation fault (core dumped) in `convolution`,https://github.com/pytorch/pytorch/issues/142209,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: crash, module: edge cases, module: empty tensor",review early; this can change release or regression risk,2024-12-16T18:38:59Z +issue,35600,Increased memory usage in repetitive torch.jit.trace calls,https://github.com/pytorch/pytorch/issues/35600,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2024-12-16T08:30:28Z +issue,142838,Segmentation fault (core dumped) in `fractional_max_pool3d_backward`,https://github.com/pytorch/pytorch/issues/142838,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: pooling",review early; this can change release or regression risk,2024-12-11T23:05:24Z +issue,142837,Segmentation fault (core dumped) in `embedding_dense_backward`,https://github.com/pytorch/pytorch/issues/142837,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: embedding",review early; this can change release or regression risk,2024-12-11T23:05:07Z +issue,142221,[MPS] nansum crashes with scalar input,https://github.com/pytorch/pytorch/issues/142221,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: mps",review early; this can change release or regression risk,2024-12-07T01:43:02Z +issue,141847,Floating point exception (core dumped) in `slow_conv3d_forward`,https://github.com/pytorch/pytorch/issues/141847,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2024-12-03T04:05:56Z +issue,141831,Nearest-neighbour upsampling with `interpolate` silently fails on CUDA when output size exceeds 2^31,https://github.com/pytorch/pytorch/issues/141831,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: nn",review early; this can change release or regression risk,2024-12-02T23:57:22Z +issue,134804,FlopsProfiler double counts sdpa flops,https://github.com/pytorch/pytorch/issues/134804,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: flop counter, module: regression, module: sdpa",review early; this can change release or regression risk,2024-12-02T18:19:12Z +issue,137347,cuDNN dBias error starting on 10/2 nightly,https://github.com/pytorch/pytorch/issues/137347,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: regression, module: sdpa",review early; this can change release or regression risk,2024-12-02T18:19:04Z +issue,122617,[Regression PyTorch 2.1 & 2.2] `torch._transformer_encoder_layer_fwd` outputs on CUDA/NestedTensorCUDA backends significantly diverge from truth for torch>=2.1 (SDP math)),https://github.com/pytorch/pytorch/issues/122617,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: nn",review early; this can change release or regression risk,2024-11-27T21:50:10Z +issue,141218,Floating point exception (core dumped) in `scaled_dot_product_attention`,https://github.com/pytorch/pytorch/issues/141218,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: empty tensor, module: error checking",review early; this can change release or regression risk,2024-11-27T21:33:24Z +issue,141229,FPE in `torch._scaled_dot_product_attention_math`,https://github.com/pytorch/pytorch/issues/141229,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: empty tensor, module: sdpa",review early; this can change release or regression risk,2024-11-27T20:11:51Z +issue,141219,Floating point exception (core dumped) in `_scaled_dot_product_flash_attention_for_cpu`,https://github.com/pytorch/pytorch/issues/141219,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: empty tensor, module: sdpa",review early; this can change release or regression risk,2024-11-27T20:11:50Z +issue,128690,`torch._transform_bias_rescale_qkv`:FPE,https://github.com/pytorch/pytorch/issues/128690,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: nn",review early; this can change release or regression risk,2024-11-27T20:02:11Z +issue,106700,C++ API `torch::nn::MultiheadAttention` Crashes by division by zero,https://github.com/pytorch/pytorch/issues/106700,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: crash, module: nn",review early; this can change release or regression risk,2024-11-27T19:52:28Z +issue,50804,Segmentation fault encountered when using nn.MultiheadAttention with v1.7.1,https://github.com/pytorch/pytorch/issues/50804,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: regression",review early; this can change release or regression risk,2024-11-27T19:48:31Z +issue,141481,Segmentation fault (core dumped) in `segment_reduce`,https://github.com/pytorch/pytorch/issues/141481,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: python frontend",review early; this can change release or regression risk,2024-11-26T08:04:46Z +issue,140945,Segmentation fault (core dumped) in `replication_pad1d_backward`,https://github.com/pytorch/pytorch/issues/140945,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: empty tensor",review early; this can change release or regression risk,2024-11-18T18:45:29Z +issue,140272,There is probably a destructor ordering accident going on with registerPythonModule,https://github.com/pytorch/pytorch/issues/140272,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: crash, module: python frontend",review early; this can change release or regression risk,2024-11-11T23:19:03Z +issue,139987,FPE in torch.nn.functional.pixel_shuffle,https://github.com/pytorch/pytorch/issues/139987,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: nn",review early; this can change release or regression risk,2024-11-08T13:37:06Z +issue,139242,Tensor.index_reduce produces incorrect result,https://github.com/pytorch/pytorch/issues/139242,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,review early; this can change release or regression risk,2024-11-04T18:21:10Z +issue,135885,☂️ 150+ MacOS tests were marked flaky recently,https://github.com/pytorch/pytorch/issues/135885,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: infra, module: macos",review early; this can change release or regression risk,2024-11-01T09:57:47Z +issue,138425,torch.div throws a floating point exception when rounding_mode is `trunc` or `floor`,https://github.com/pytorch/pytorch/issues/138425,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: intel, module: python frontend",review early; this can change release or regression risk,2024-10-22T12:47:28Z +issue,17199,Deadlock with multiprocessing (using fork) and OpenMP / PyTorch should warn after OMP and fork that multithreading may be broken,https://github.com/pytorch/pytorch/issues/17199,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: multiprocessing",review early; this can change release or regression risk,2024-10-11T07:29:28Z +issue,137098,[CI] Multiple tests fail with `Host and device pointer dont match with cudaHostRegister` and are being disabled,https://github.com/pytorch/pytorch/issues/137098,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",review early; this can change release or regression risk,2024-10-05T06:18:10Z +issue,135637,MPS code contains references to undocumented APIs,https://github.com/pytorch/pytorch/issues/135637,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: mps",review early; this can change release or regression risk,2024-10-02T17:35:58Z +issue,136819,Aborted (core dumped) in `torch.smm`/`torch.hspmm`/`torch.hsmm`/`torch.sspaddmm`,https://github.com/pytorch/pytorch/issues/136819,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: sparse",review early; this can change release or regression risk,2024-09-30T17:27:34Z +issue,136704,Segmentation fault (core dumped) in `torch._fft_r2c`/`torch._fft_c2`/`torch._fft_c2r`,https://github.com/pytorch/pytorch/issues/136704,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2024-09-27T02:16:03Z +issue,136709,Segmentation fault (core dumped) in `torch._weight_norm`/`torch._weight_int8pack_mm`,https://github.com/pytorch/pytorch/issues/136709,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2024-09-27T02:15:46Z +issue,136714,Aborted (core dumped) in `torch.linalg.ldl_solve` with double free or corruption (out),https://github.com/pytorch/pytorch/issues/136714,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: linear algebra",review early; this can change release or regression risk,2024-09-27T01:59:42Z +issue,136719,Segmentation fault (core dumped) in `torch.nn.functional.max_pool1d`,https://github.com/pytorch/pytorch/issues/136719,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: python frontend",review early; this can change release or regression risk,2024-09-27T01:57:45Z +issue,136374,Runbook for GCP A100 runners should be updated,https://github.com/pytorch/pytorch/issues/136374,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: infra,review early; this can change release or regression risk,2024-09-23T17:45:11Z +issue,136299,The results of baddbmm on CPU seem to have issues under certain conditions,https://github.com/pytorch/pytorch/issues/136299,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: correctness (silent), module: regression",review early; this can change release or regression risk,2024-09-23T17:15:38Z +issue,136217,Segmentaion fault on Apple M1 caused by libomp on v2.4.1,https://github.com/pytorch/pytorch/issues/136217,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: macos, module: openmp",review early; this can change release or regression risk,2024-09-19T16:53:56Z +issue,122394,[MPS] Severe Performance Degradation with Deterministic Algorithm on macOS 14.4,https://github.com/pytorch/pytorch/issues/122394,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance, module: regression",review early; this can change release or regression risk,2024-09-10T21:52:16Z +issue,82627,PyTorch crashes when running with OpenACC,https://github.com/pytorch/pytorch/issues/82627,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: openmp, module: third_party",review early; this can change release or regression risk,2024-09-08T13:46:39Z +issue,133334,torch._scaled_mm row-wise hits CUDA invalid memory access when M % 256 != 0,https://github.com/pytorch/pytorch/issues/133334,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: floatx (formerly float8)",review early; this can change release or regression risk,2024-08-22T19:44:12Z +issue,80431,CMake Error: File /opt/pytorch/build_variables.bzl does not exist.,https://github.com/pytorch/pytorch/issues/80431,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: docker, module: regression",review early; this can change release or regression risk,2024-08-15T11:43:10Z +issue,132086,Result of MPS memory operation can be corrupted if memory pressure is high,https://github.com/pytorch/pytorch/issues/132086,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: memory usage, module: mps",review early; this can change release or regression risk,2024-08-05T18:27:55Z +issue,70108,Derivative for _ctc_loss_backward,https://github.com/pytorch/pytorch/issues/70108,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: double backwards",review early; this can change release or regression risk,2024-07-24T16:56:00Z +issue,83818,torch.linalg.eigh crashe for matrices of size 2895×2895 or larger on eigen and M1,https://github.com/pytorch/pytorch/issues/83818,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: linear algebra, module: m1",review early; this can change release or regression risk,2024-07-12T20:37:35Z +issue,78829,Certain import order triggers segmentation fault,https://github.com/pytorch/pytorch/issues/78829,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2024-07-10T10:20:41Z +issue,128503,crash@sleef_tryVXE2 () while trying to run torch.compile() BERT model,https://github.com/pytorch/pytorch/issues/128503,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: power, module: sleef",review early; this can change release or regression risk,2024-07-04T07:25:20Z +issue,125247,RuntimeError: could not create a primitive descriptor for a matmul primitive,https://github.com/pytorch/pytorch/issues/125247,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash",review early; this can change release or regression risk,2024-07-02T07:27:40Z +issue,39007,[docs] Urls changed => forum links would become invalid,https://github.com/pytorch/pytorch/issues/39007,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",review early; this can change release or regression risk,2024-06-24T15:23:51Z +issue,128694,Segmentation fault (core dumped) in `torch._fused_moving_avg_obs_fq_helper`,https://github.com/pytorch/pytorch/issues/128694,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2024-06-15T03:11:38Z +issue,128695,Segmentation fault (core dumped) in `torch._weight_norm_interface`,https://github.com/pytorch/pytorch/issues/128695,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: empty tensor",review early; this can change release or regression risk,2024-06-14T14:10:57Z +issue,128692,Segmentation fault (core dumped) in `torch._weight_int4pack_mm`,https://github.com/pytorch/pytorch/issues/128692,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2024-06-14T13:38:42Z +issue,126385,Segmentation Faults loading Models pytorch v2.3.0 Apple M2,https://github.com/pytorch/pytorch/issues/126385,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: macos, module: openmp",review early; this can change release or regression risk,2024-06-10T20:43:46Z +issue,113037,"`TensorIteratorBase::is_scalar` return `false` for empty numpy tensors, but true for empty Torch ones",https://github.com/pytorch/pytorch/issues/113037,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: regression",review early; this can change release or regression risk,2024-06-05T23:01:53Z +issue,127102,segfault when registering op with mismatched schema,https://github.com/pytorch/pytorch/issues/127102,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: custom-operators, module: opcheck",review early; this can change release or regression risk,2024-05-29T00:02:17Z +issue,66782,"Segmentation Fault when importing Torch, 2021 version",https://github.com/pytorch/pytorch/issues/66782,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash",review early; this can change release or regression risk,2024-05-16T17:39:05Z +issue,64794,linalg.lstsq out variant fails internal assert because it uses non-inplace view op for some inputs,https://github.com/pytorch/pytorch/issues/64794,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: linear algebra",review early; this can change release or regression risk,2024-05-14T00:41:00Z +issue,120995,AttributeError: module 'mpmath' has no attribute 'rational',https://github.com/pytorch/pytorch/issues/120995,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: releng",review early; this can change release or regression risk,2024-05-09T18:43:09Z +issue,82068,upsample_bilinear2d() received an invalid combination of arguments,https://github.com/pytorch/pytorch/issues/82068,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: regression",review early; this can change release or regression risk,2024-05-06T17:12:16Z +issue,120622,_native_batch_norm_legit doesn't bump version counter of buffers,https://github.com/pytorch/pytorch/issues/120622,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: codegen, module: correctness (silent), module: norms and normalization",review early; this can change release or regression risk,2024-05-06T15:10:03Z +issue,124644,torch.Tensor.remainder raises a floating point exception when divisor is -1,https://github.com/pytorch/pytorch/issues/124644,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2024-04-29T17:18:50Z +issue,105943,PT2 is not thread safe,https://github.com/pytorch/pytorch/issues/105943,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review early; this can change release or regression risk,2024-04-09T18:37:31Z +issue,123219,[CI] vec_test_all tests are not run in CI,https://github.com/pytorch/pytorch/issues/123219,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cpu, module: intel, module: regression",review early; this can change release or regression risk,2024-04-03T14:30:02Z +issue,2576,CUDA multinomial is limited to 2^24 categories,https://github.com/pytorch/pytorch/issues/2576,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: distributions",review early; this can change release or regression risk,2024-04-01T14:46:25Z +issue,121150,☂️ oneDNN update functional/regression,https://github.com/pytorch/pytorch/issues/121150,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: mkldnn,review early; this can change release or regression risk,2024-03-15T11:35:49Z +issue,101529,Inconsistent performance degradation of 3x3 convolution (torch 2.0.1+cu118),https://github.com/pytorch/pytorch/issues/101529,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: performance, module: regression",review early; this can change release or regression risk,2024-03-02T02:06:59Z +issue,119547,CUDA Memory error while having a lot of free memory,https://github.com/pytorch/pytorch/issues/119547,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: regression",review early; this can change release or regression risk,2024-02-28T12:02:21Z +issue,120597,floating point exception on `torch.fmod` when x is large negative integer and y is `-1`,https://github.com/pytorch/pytorch/issues/120597,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2024-02-26T17:54:09Z +issue,47056,Training slows down and memory usage increases when upgrading from PyTorch 1.6 to 1.7,https://github.com/pytorch/pytorch/issues/47056,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: performance, module: regression",review early; this can change release or regression risk,2024-02-22T21:50:34Z +issue,53625,torch.arange() issue,https://github.com/pytorch/pytorch/issues/53625,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: regression, oncall: distributed",review early; this can change release or regression risk,2024-02-10T14:53:21Z +issue,119000,Performance regression in torch.nn.Linear,https://github.com/pytorch/pytorch/issues/119000,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn, module: performance, module: regression",review early; this can change release or regression risk,2024-02-05T16:48:23Z +issue,117431,Program killed by SIGFPE in the backward pass of torch.sparse.sampled_addmm when input has specific shape,https://github.com/pytorch/pytorch/issues/117431,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: sparse",review early; this can change release or regression risk,2024-01-16T15:11:56Z +issue,78924,"If large enough tensor is being cloned, parallel dataloading hangs on M1 Mac",https://github.com/pytorch/pytorch/issues/78924,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: dataloader, module: deadlock, module: macos",review early; this can change release or regression risk,2024-01-14T18:50:07Z +issue,117254,```at::mse_loss```: Internal assertion failed,https://github.com/pytorch/pytorch/issues/117254,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: loss",review early; this can change release or regression risk,2024-01-11T19:57:09Z +issue,115416,torch.nn.functional.embedding segmentation fault on large negative ids,https://github.com/pytorch/pytorch/issues/115416,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: int overflow",review early; this can change release or regression risk,2023-12-08T23:19:37Z +issue,112043,"AssertionError: graph-captured input #1, of type , is not among original inputs of types…",https://github.com/pytorch/pytorch/issues/112043,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: export,review early; this can change release or regression risk,2023-11-28T17:08:45Z +issue,114529,Bug in element-wise multiplication of `torch.sparse_csr_tensor`s on GPU - 0's in result considered significant - PyTorch 2.1.1,https://github.com/pytorch/pytorch/issues/114529,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, module: sparse",review early; this can change release or regression risk,2023-11-28T14:40:00Z +issue,50037,Torch quantized_lstm_cell op out-of-bounds access,https://github.com/pytorch/pytorch/issues/50037,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, oncall: quantization",review early; this can change release or regression risk,2023-11-17T22:54:35Z +issue,106432,inconsistent dtype of scale and zero_point in observers,https://github.com/pytorch/pytorch/issues/106432,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,review early; this can change release or regression risk,2023-11-17T22:42:45Z +issue,88882,Can't import torch --> OSError related to libcublasLt.so.11,https://github.com/pytorch/pytorch/issues/88882,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: regression",review early; this can change release or regression risk,2023-11-02T13:58:24Z +issue,99719,New feature requested: vmap for torch.histc,https://github.com/pytorch/pytorch/issues/99719,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,review early; this can change release or regression risk,2023-10-12T18:20:37Z +issue,110205,RuntimeError: Expected packed scalar Tensor to be of dimension 1. Got 0 instead.,https://github.com/pytorch/pytorch/issues/110205,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: regression",review early; this can change release or regression risk,2023-10-04T21:39:47Z +issue,107668,torch.dot gives wrong result on Macos,https://github.com/pytorch/pytorch/issues/107668,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: macos",review early; this can change release or regression risk,2023-10-01T20:58:21Z +issue,94544,Abort Caused by Virtual Function,https://github.com/pytorch/pytorch/issues/94544,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: regression",review early; this can change release or regression risk,2023-09-08T02:03:21Z +issue,108532,"Breaking incompatibility with Cuda 12.2, pytorch stable, torchvision",https://github.com/pytorch/pytorch/issues/108532,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash, module: cuda",review early; this can change release or regression risk,2023-09-07T17:37:27Z +issue,107363,Regression in text encoding,https://github.com/pytorch/pytorch/issues/107363,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multithreading, module: performance, module: regression",review early; this can change release or regression risk,2023-08-23T03:26:14Z +issue,107429,Abort when running torch.set_num_interop_threads,https://github.com/pytorch/pytorch/issues/107429,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: single threaded",review early; this can change release or regression risk,2023-08-20T00:54:18Z +issue,39388,"RuntimeError: NCCL error in: /pytorch/torch/lib/c10d/ProcessGroupNCCL.cpp:514, unhandled system error, NCCL version 2.4.8",https://github.com/pytorch/pytorch/issues/39388,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: nccl",review early; this can change release or regression risk,2023-08-18T18:28:22Z +issue,106362,Calling ops.aten.embedding_bag() function got silent crash,https://github.com/pytorch/pytorch/issues/106362,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: nn",review early; this can change release or regression risk,2023-08-14T17:02:56Z +issue,50013,torch.Tensor.repeat is divergent from np.repeat,https://github.com/pytorch/pytorch/issues/50013,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy",review early; this can change release or regression risk,2023-07-27T16:39:10Z +issue,103756,libtorch > 1.9.1 produces segfault on Qt5 gui application exit,https://github.com/pytorch/pytorch/issues/103756,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: crash",review early; this can change release or regression risk,2023-07-25T08:26:45Z +issue,42793,distributed training with c10d does not work with layerdrop in pytorch > 1.4,https://github.com/pytorch/pytorch/issues/42793,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,review early; this can change release or regression risk,2023-07-06T11:00:37Z +issue,103117,Observed regress in DataLoader spawn from PyTorch1.13 to PyTorch2.0,https://github.com/pytorch/pytorch/issues/103117,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: data, module: performance, module: regression",review early; this can change release or regression risk,2023-06-13T18:00:24Z +issue,99509,'Illegal instruction (core dumped)' for gpt-j bf16 generation task using greedy search,https://github.com/pytorch/pytorch/issues/99509,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: intel",review early; this can change release or regression risk,2023-06-07T16:06:24Z +issue,99982,torch.cuda.is_available() crashes python in systems with disabled gpu,https://github.com/pytorch/pytorch/issues/99982,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: edge cases",review early; this can change release or regression risk,2023-06-03T15:16:36Z +issue,98537,Segfault when using torch.ops.* to access de-registered op,https://github.com/pytorch/pytorch/issues/98537,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: dispatch, module: library",review early; this can change release or regression risk,2023-04-28T14:50:07Z +issue,100085,[regression] torch.norm with out dtype bfloat16 cause runtime error,https://github.com/pytorch/pytorch/issues/100085,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization, module: regression",review early; this can change release or regression risk,2023-04-27T17:17:31Z +issue,98136,A Segment Fault can be triggered in torch._grid_sampler_2d_cpu_fallback,https://github.com/pytorch/pytorch/issues/98136,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: nn",review early; this can change release or regression risk,2023-04-14T15:10:18Z +issue,98503,Power VSX vectorization support disabled,https://github.com/pytorch/pytorch/issues/98503,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: power, module: regression",review early; this can change release or regression risk,2023-04-07T17:07:37Z +issue,55375,`DataParallel` (`broadcast_coalesced`) with complex tensors yield real views,https://github.com/pytorch/pytorch/issues/55375,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: complex, module: data parallel, oncall: distributed",review early; this can change release or regression risk,2023-04-04T13:59:09Z +issue,95432,Regression bug in `torch.nn.ReLU6` and `torch.nn.Hardtanh` that `inplace=True` doesn't work in PyTorch 1.10.0~1.13.1,https://github.com/pytorch/pytorch/issues/95432,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: nn",review early; this can change release or regression risk,2023-04-03T18:32:30Z +issue,97086,A crash due to Floating Point Exception can be triggered in torch.index_select,https://github.com/pytorch/pytorch/issues/97086,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2023-03-27T17:12:04Z +issue,96276,A Segment Fault can be triggered in torch.geqrf with an edge case,https://github.com/pytorch/pytorch/issues/96276,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-03-09T12:04:23Z +issue,96275,A Segment Fault can be triggered in torch.pinverse,https://github.com/pytorch/pytorch/issues/96275,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-03-09T12:03:55Z +issue,96277,A Segment Fault can be triggered in torch.adaptive_max_pool1d with an edge case,https://github.com/pytorch/pytorch/issues/96277,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-03-09T12:03:45Z +issue,91754,CUDA error: initialization error,https://github.com/pytorch/pytorch/issues/91754,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nvfuser, module: regression",review early; this can change release or regression risk,2023-03-03T04:07:14Z +issue,54418,Some Numba tests are failing,https://github.com/pytorch/pytorch/issues/54418,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: numba, module: tests",review early; this can change release or regression risk,2023-02-26T10:16:29Z +issue,94808,Memory Corruption in torch.lstm caused by edge cases,https://github.com/pytorch/pytorch/issues/94808,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: rnn",review early; this can change release or regression risk,2023-02-15T16:22:38Z +issue,94698,A segment fault can be triggered in torch.avg_pool1d,https://github.com/pytorch/pytorch/issues/94698,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-02-15T16:22:09Z +issue,94668,A segment fault can be triggered in torch.lstm with edge cases,https://github.com/pytorch/pytorch/issues/94668,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-02-15T16:21:43Z +issue,94669,A segment fault can be triggered in torch.svd,https://github.com/pytorch/pytorch/issues/94669,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-02-15T16:21:26Z +issue,94696,A segment fault can be triggered in torch.max_pool1d_with_indices,https://github.com/pytorch/pytorch/issues/94696,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-02-15T16:19:49Z +issue,94594,A segment fault can be triggered in torch.histogramdd,https://github.com/pytorch/pytorch/issues/94594,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-02-10T17:12:51Z +issue,93846,large number of temporary files generated when using dataloader with num_workers>0,https://github.com/pytorch/pytorch/issues/93846,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: openmp",review early; this can change release or regression risk,2023-02-10T04:29:56Z +issue,68349,"SIGILL, Illegal Instruction from libtorch_cpu.so when callling backward() function in a VM.",https://github.com/pytorch/pytorch/issues/68349,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: dependency bug, module: mkldnn",review early; this can change release or regression risk,2023-02-08T07:39:55Z +issue,94125,A Floating Point Exception can be trigerred in torch._C._nn.slow_conv3d,https://github.com/pytorch/pytorch/issues/94125,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases",review early; this can change release or regression risk,2023-02-07T18:51:10Z +issue,91590,Crashes of linalg.ldl_solve on different edge cases not coming from linalg.ldl_factor,https://github.com/pytorch/pytorch/issues/91590,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: linear algebra",review early; this can change release or regression risk,2023-01-23T16:08:58Z +issue,82894,linalg and lu tests fail when run in parallel on linux cuda,https://github.com/pytorch/pytorch/issues/82894,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda, module: linear algebra",review early; this can change release or regression risk,2023-01-16T10:34:45Z +issue,58414,cuDNN v8 API tracking issue,https://github.com/pytorch/pytorch/issues/58414,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cudnn,review early; this can change release or regression risk,2023-01-11T00:17:59Z +issue,90535,Illegal hardware instruction using torch.nn.Conv2d on aarch64 (Raspberry Pi 4),https://github.com/pytorch/pytorch/issues/90535,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2023-01-09T10:43:02Z +issue,41661,libgcc_s.so.1 must be installed for pthread_cancel to work,https://github.com/pytorch/pytorch/issues/41661,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,review early; this can change release or regression risk,2023-01-05T12:22:29Z +issue,83243,RPC: wait method of Future object return 0 sometimes in rpc framework,https://github.com/pytorch/pytorch/issues/83243,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",review early; this can change release or regression risk,2022-12-27T03:34:00Z +issue,90676,RuntimeError: Error in dlopen: libnvJitLink.so.12: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/90676,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: distributions",review early; this can change release or regression risk,2022-12-12T15:07:55Z +issue,88842,test/test_ops.py is segfaulting on master build with DEBUG assets,https://github.com/pytorch/pytorch/issues/88842,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: testing",review early; this can change release or regression risk,2022-12-06T20:11:14Z +issue,89718,`torch.Tensor.flatten` Trigger Segmentation Fault when trying to provide and output named dim,https://github.com/pytorch/pytorch/issues/89718,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: named tensor",review early; this can change release or regression risk,2022-11-30T10:25:17Z +issue,82072,cant build with USE_VULKAN=1,https://github.com/pytorch/pytorch/issues/82072,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: vulkan, oncall: mobile",review early; this can change release or regression risk,2022-11-23T22:26:42Z +issue,38122,Pybind11 cpp extensions broken with pytorch v1.5.0,https://github.com/pytorch/pytorch/issues/38122,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: docs, module: regression",review early; this can change release or regression risk,2022-11-22T15:05:59Z +issue,83393,Test public bindings in CI gives weird output on error,https://github.com/pytorch/pytorch/issues/83393,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: python frontend, module: tests",review early; this can change release or regression risk,2022-11-15T04:22:12Z +issue,89006,Add smoke-tests for CPP extensions compilations,https://github.com/pytorch/pytorch/issues/89006,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: regression",review early; this can change release or regression risk,2022-11-14T19:19:05Z +issue,73179,Floating point exception in _nnpack_spatial_convolution,https://github.com/pytorch/pytorch/issues/73179,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cpu, module: error checking, module: nnpack",review early; this can change release or regression risk,2022-11-07T22:55:03Z +issue,73174,`LayerNorm` triggers INTERNAL ASSERT,https://github.com/pytorch/pytorch/issues/73174,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: error checking, module: norms and normalization",review early; this can change release or regression risk,2022-11-07T08:22:24Z +issue,88036,A segment fault can be triggered in fbgemm_pack_gemm_matrix_fp16,https://github.com/pytorch/pytorch/issues/88036,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: half, module: third_party",review early; this can change release or regression risk,2022-10-31T19:12:43Z +issue,87283,Missing string parsing for some parameter types in python arg parsing logic,https://github.com/pytorch/pytorch/issues/87283,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: pybind,review early; this can change release or regression risk,2022-10-24T17:10:43Z +issue,86704,Performance tests mnist_hogwild-cpu_memory CPU memory increase by 30%,https://github.com/pytorch/pytorch/issues/86704,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression,review early; this can change release or regression risk,2022-10-17T17:06:45Z +issue,85538,Iterative Global Pruning Cause GPU Memory Leak,https://github.com/pytorch/pytorch/issues/85538,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: nn",review early; this can change release or regression risk,2022-10-08T04:28:05Z +issue,85909,Steam Deck Core Dump,https://github.com/pytorch/pytorch/issues/85909,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: rocm,review early; this can change release or regression risk,2022-10-03T17:27:31Z +issue,80104,Provide error message when thread pool is exhausted in RPC,https://github.com/pytorch/pytorch/issues/80104,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",review early; this can change release or regression risk,2022-09-26T16:07:00Z +issue,85329,Crash in `torch.package.PackageExporter`,https://github.com/pytorch/pytorch/issues/85329,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, oncall: package/deploy",review early; this can change release or regression risk,2022-09-21T19:49:07Z +issue,85215,Floating point exception in gather gradient computation.,https://github.com/pytorch/pytorch/issues/85215,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: error checking",review early; this can change release or regression risk,2022-09-21T19:43:42Z +issue,85217,Segmentation fault in native_batch_norm,https://github.com/pytorch/pytorch/issues/85217,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking, module: norms and normalization",review early; this can change release or regression risk,2022-09-21T19:43:33Z +issue,85214,Segmentation fault in mkldnn_reorder_conv2d_weight and mkldnn_reorder_conv3d_weight,https://github.com/pytorch/pytorch/issues/85214,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: edge cases, module: mkldnn",review early; this can change release or regression risk,2022-09-19T17:25:25Z +issue,84990,Segmentation fault in `torch.futures.collect_all`,https://github.com/pytorch/pytorch/issues/84990,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: error checking",review early; this can change release or regression risk,2022-09-16T17:24:38Z +issue,84181,Session of Google Colab crashes when `torch.utils::SummaryWriter` is called after importing `torchaudio`,https://github.com/pytorch/pytorch/issues/84181,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: tensorboard",review early; this can change release or regression risk,2022-09-12T17:11:56Z +issue,80488,Negative values still produced by torch.nn.functional.kl_div,https://github.com/pytorch/pytorch/issues/80488,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,review early; this can change release or regression risk,2022-08-31T08:31:10Z +issue,84051,model.load_state_dict won't work in Child process if a sufficiently large tensor was padded in the Parent (even if empty padded),https://github.com/pytorch/pytorch/issues/84051,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: multiprocessing",review early; this can change release or regression risk,2022-08-30T10:06:09Z +issue,83070,make_fx + aot_autograd segfaults,https://github.com/pytorch/pytorch/issues/83070,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: functorch, module: fx",review early; this can change release or regression risk,2022-08-22T14:53:58Z +issue,80946,torch.nn.functional.linear fails for multi-dimensional bias from torch 1.12,https://github.com/pytorch/pytorch/issues/80946,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: regression",review early; this can change release or regression risk,2022-08-18T15:03:34Z +issue,83577,libtorch malloc cause coredump,https://github.com/pytorch/pytorch/issues/83577,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: crash, module: static linking",review early; this can change release or regression risk,2022-08-17T14:17:50Z +issue,82813,functionalize and make_fx are not composable resulting in segfault and cuda error,https://github.com/pytorch/pytorch/issues/82813,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: functorch, module: fx",review early; this can change release or regression risk,2022-08-10T16:12:21Z +issue,81195,torch._weight_norm with specified dim returns wrong output,https://github.com/pytorch/pytorch/issues/81195,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn, module: norms and normalization, module: regression",review early; this can change release or regression risk,2022-07-27T16:46:02Z +issue,77284,non-rentrant checkpointing uses same memory as non-checkpointed code,https://github.com/pytorch/pytorch/issues/77284,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, oncall: distributed",review early; this can change release or regression risk,2022-07-25T03:31:50Z +issue,81127,Docker updates cause subsequent builds to fail,https://github.com/pytorch/pytorch/issues/81127,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci,review early; this can change release or regression risk,2022-07-12T16:47:25Z +issue,80337,Illegal Memory Access from nonzero method when Tensor is Too Large,https://github.com/pytorch/pytorch/issues/80337,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: dependency bug, module: edge cases",review early; this can change release or regression risk,2022-06-27T21:45:57Z +issue,80012,static builds are broken by MKL_DNN,https://github.com/pytorch/pytorch/issues/80012,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: regression, module: third_party",review early; this can change release or regression risk,2022-06-22T17:45:34Z +issue,69801,"Dropout2d doesn't drop channels for (C, H, W)",https://github.com/pytorch/pytorch/issues/69801,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: nn",review early; this can change release or regression risk,2022-06-15T20:43:38Z +issue,13918,Pytorch very slow to convert list of numpy arrays into tensors,https://github.com/pytorch/pytorch/issues/13918,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: performance",review early; this can change release or regression risk,2022-06-15T09:00:31Z +issue,79014,RuntimeError: cuDNN error: CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/79014,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: cudnn, module: ddp",review early; this can change release or regression risk,2022-06-13T18:41:56Z +issue,59745,[c10d] Work objects should have a general operator<<,https://github.com/pytorch/pytorch/issues/59745,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",review early; this can change release or regression risk,2022-06-03T21:10:44Z +issue,73186,Segmentation fault in fractional_max_pool3d,https://github.com/pytorch/pytorch/issues/73186,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: error checking, module: pooling",review early; this can change release or regression risk,2022-05-07T18:08:23Z +issue,73185,Segmentation fault in fractional_max_pool2d,https://github.com/pytorch/pytorch/issues/73185,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: error checking, module: pooling",review early; this can change release or regression risk,2022-05-07T18:06:58Z +issue,47953,Linear algebra GPU backend tracking issue [magma/cusolver/cublas],https://github.com/pytorch/pytorch/issues/47953,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",review early; this can change release or regression risk,2022-05-05T13:28:14Z +issue,61223,optimize_for_inference does not handle MKLDNN for BatchNorm1d,https://github.com/pytorch/pytorch/issues/61223,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2022-05-01T09:42:28Z +issue,65815,Overflow error in torch.linalg uncaught,https://github.com/pytorch/pytorch/issues/65815,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: linear algebra, module: multithreading, module: numerical-stability",review early; this can change release or regression risk,2022-04-14T19:49:19Z +issue,73488,After XNNPack update `TestXNNPACKSerDes.test_linear` started to fail,https://github.com/pytorch/pytorch/issues/73488,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, module: rocm, module: xnnpack",review early; this can change release or regression risk,2022-03-31T17:16:55Z +issue,74547,importing open3d before pytorch causes matmul to produce a segfault,https://github.com/pytorch/pytorch/issues/74547,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cublas, module: cuda, module: regression",review early; this can change release or regression risk,2022-03-23T20:33:42Z +issue,38228,pybind11::gil_scoped_release: crash on exit with daemon threads,https://github.com/pytorch/pytorch/issues/38228,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: dependency bug",review early; this can change release or regression risk,2022-03-22T17:47:20Z +issue,74422,Embedding Pytorch in C++ using pybind fails on interpreter shutdown,https://github.com/pytorch/pytorch/issues/74422,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: pybind, module: third_party",review early; this can change release or regression risk,2022-03-22T00:56:08Z +issue,73792,Bug: torch.distributions.mixture_same_distribution._pad_mixture_dimension,https://github.com/pytorch/pytorch/issues/73792,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: distributions",review early; this can change release or regression risk,2022-03-07T19:01:49Z +issue,73349,Large performance difference of loss.backward() between torch-1.9.0 and torch-1.8.0,https://github.com/pytorch/pytorch/issues/73349,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: regression",review early; this can change release or regression risk,2022-02-25T08:57:52Z +issue,35151,Segfault when using misaligned data pointer (from joblib),https://github.com/pytorch/pytorch/issues/35151,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: numpy",review early; this can change release or regression risk,2022-02-24T22:37:54Z +issue,70350,Core dumped with large matmul on aarch64,https://github.com/pytorch/pytorch/issues/70350,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: crash",review early; this can change release or regression risk,2022-02-24T13:06:18Z +issue,15849,DataLoader with option to re-use worker processes,https://github.com/pytorch/pytorch/issues/15849,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,review early; this can change release or regression risk,2022-02-22T13:42:40Z +issue,71227,Move torch::deploy tests to their own workflow job,https://github.com/pytorch/pytorch/issues/71227,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: deploy,review early; this can change release or regression risk,2022-02-08T19:46:28Z +issue,72182,test_fn_fwgrad_bwgrad_special_ndtr_cuda_float64 fails,https://github.com/pytorch/pytorch/issues/72182,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",review early; this can change release or regression risk,2022-02-02T20:00:50Z +issue,72189,test_fn_fwgrad_bwgrad_[trapezoid|trapz]_cuda_complex128 causes CUDA memory exception,https://github.com/pytorch/pytorch/issues/72189,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: crash, module: cuda",review early; this can change release or regression risk,2022-02-02T19:59:44Z +issue,29973,Indexing into tensor order of magnitude slower than numpy,https://github.com/pytorch/pytorch/issues/29973,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: autograd, module: performance",review early; this can change release or regression risk,2022-02-01T15:49:01Z +issue,71391,Calling .backward() inside of an LBFGS closure function throws an exception in Libtorch v1.6.0+,https://github.com/pytorch/pytorch/issues/71391,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: regression",review early; this can change release or regression risk,2022-01-25T19:36:01Z +issue,42452,`device-side assert triggered` - when probability > 1.0,https://github.com/pytorch/pytorch/issues/42452,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",review early; this can change release or regression risk,2022-01-14T02:41:10Z +issue,42487,Support recursive data type in TorchScript,https://github.com/pytorch/pytorch/issues/42487,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2021-12-30T06:42:53Z +issue,69512,Inconsistent behavior of cosine_similarity between fp16 and fp32 inputs,https://github.com/pytorch/pytorch/issues/69512,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: half, module: numerical-stability",review early; this can change release or regression risk,2021-12-13T18:21:56Z +issue,67448,Deadlock in test_multiprocessing_spawn.py,https://github.com/pytorch/pytorch/issues/67448,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: multiprocessing",review early; this can change release or regression risk,2021-11-08T18:25:55Z +issue,67797,Conv2d kernel performance regression on CPU since PyTorch 1.9,https://github.com/pytorch/pytorch/issues/67797,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: mkldnn, module: performance, module: regression",review early; this can change release or regression risk,2021-11-05T19:56:50Z +issue,67247,pytorch_linux_xenial_py3_6_gcc5_4_test may timeout during test_multiprocessing_spawn,https://github.com/pytorch/pytorch/issues/67247,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: tests",review early; this can change release or regression risk,2021-11-02T17:46:10Z +issue,60081,[Bug] numpy is no longer a required dependency,https://github.com/pytorch/pytorch/issues/60081,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy,review early; this can change release or regression risk,2021-11-01T10:47:14Z +issue,59237,Multiplication of `torch.tensor` with `np.array` does the operation with numpy.,https://github.com/pytorch/pytorch/issues/59237,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy,review early; this can change release or regression risk,2021-10-28T13:31:30Z +issue,65366,Exception thrown in final autograd callback (queue_callback) not caught if not on CPU thread: terminate called after throwing an instance of 'python_error',https://github.com/pytorch/pytorch/issues/65366,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: bootcamp, module: crash",review early; this can change release or regression risk,2021-10-11T17:25:21Z +issue,46155,Complex backward returns NaN values,https://github.com/pytorch/pytorch/issues/46155,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cuda",review early; this can change release or regression risk,2021-10-06T15:42:35Z +issue,38910,Memory Leak with Docker GPU,https://github.com/pytorch/pytorch/issues/38910,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: memory usage, module: regression",review early; this can change release or regression risk,2021-10-05T19:20:41Z +issue,52633,CUDA `linalg.norm` matrix order +/-2 and nuclear norm for extreme values may be incorrect,https://github.com/pytorch/pytorch/issues/52633,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: nans and infs",review early; this can change release or regression risk,2021-09-14T16:20:13Z +issue,57178,MaxPool2D Returns Wrong Shape With Ceil_Mode,https://github.com/pytorch/pytorch/issues/57178,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: pooling",review early; this can change release or regression risk,2021-09-13T16:08:34Z +issue,63486,Distributed / MPI / CUDA: Incorrect messages received from isend/irecv in PyTorch 1.9,https://github.com/pytorch/pytorch/issues/63486,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), oncall: distributed",review early; this can change release or regression risk,2021-09-02T15:14:16Z +issue,50344,NumPy Compatibility tracking issue,https://github.com/pytorch/pytorch/issues/50344,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy,review early; this can change release or regression risk,2021-08-31T07:10:26Z +issue,63749,Segmentation fault when using C++/pybind11 module without also importing torch,https://github.com/pytorch/pytorch/issues/63749,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: pybind, module: regression",review early; this can change release or regression risk,2021-08-23T15:50:27Z +issue,50012,torch.split is divergent from np.split,https://github.com/pytorch/pytorch/issues/50012,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy",review early; this can change release or regression risk,2021-08-13T18:36:42Z +issue,59196,torch.zeros memory leak : cpu,https://github.com/pytorch/pytorch/issues/59196,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage,review early; this can change release or regression risk,2021-08-04T11:56:56Z +issue,48558,[FR] tensor ctors should support directly creating in shared memory,https://github.com/pytorch/pytorch/issues/48558,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,review early; this can change release or regression risk,2021-07-07T13:25:52Z +issue,60626,Impossible to raise the limit on number of shared memory tensors,https://github.com/pytorch/pytorch/issues/60626,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,review early; this can change release or regression risk,2021-07-05T23:38:19Z +issue,58370,Add better python operators annotations for IDE type checking,https://github.com/pytorch/pytorch/issues/58370,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: typing,review early; this can change release or regression risk,2021-06-30T08:17:25Z +issue,60628,pip install nightly torch torchaudio and torchvision together will install 0.3.0 version of torchvsion on Ubuntu,https://github.com/pytorch/pytorch/issues/60628,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,review early; this can change release or regression risk,2021-06-28T20:48:47Z +issue,44634,torch.sparse improvements - tracking issue,https://github.com/pytorch/pytorch/issues/44634,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: sparse,review early; this can change release or regression risk,2021-06-28T11:12:20Z +issue,24823,"Problematic handling of NaN and inf in grid_sample, causing segfaults, corrupted CUDA memory, and incorrect results",https://github.com/pytorch/pytorch/issues/24823,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: interpolation",review early; this can change release or regression risk,2021-06-21T15:13:23Z +issue,37449,RuntimeError: CUDA error: an illegal memory access was encountered with channels_last,https://github.com/pytorch/pytorch/issues/37449,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: cudnn, module: dependency bug",review early; this can change release or regression risk,2021-06-14T14:11:26Z +issue,59645,Tensor.type() does not work with meta tensors,https://github.com/pytorch/pytorch/issues/59645,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: meta tensors",review early; this can change release or regression risk,2021-06-08T19:53:52Z +issue,6998,[pytorch] Not handling python reload properly,https://github.com/pytorch/pytorch/issues/6998,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2021-06-06T22:11:41Z +issue,48491,SegmentationFault when pytorch is installed from source.,https://github.com/pytorch/pytorch/issues/48491,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: crash",review early; this can change release or regression risk,2021-05-28T12:02:28Z +issue,48382,hangs indefinitely at os.waitpid(),https://github.com/pytorch/pytorch/issues/48382,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: docs, module: error checking, module: multiprocessing",review early; this can change release or regression risk,2021-05-14T02:34:34Z +issue,40137,Improve RPC test debugability,https://github.com/pytorch/pytorch/issues/40137,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: rpc,review early; this can change release or regression risk,2021-04-26T17:17:43Z +issue,49520,Segfault in torch.bincount,https://github.com/pytorch/pytorch/issues/49520,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash,review early; this can change release or regression risk,2021-04-21T16:09:11Z +issue,51134,segmentation fault in torch.nn.ReplicationPad3d/2d when padding is large,https://github.com/pytorch/pytorch/issues/51134,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: nn, module: padding",review early; this can change release or regression risk,2021-04-14T22:04:11Z +issue,55669,Segmentation fault when loss.backward(),https://github.com/pytorch/pytorch/issues/55669,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",review early; this can change release or regression risk,2021-04-12T17:24:23Z +issue,45136,Multiple torch.load in one file,https://github.com/pytorch/pytorch/issues/45136,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: regression, module: serialization",review early; this can change release or regression risk,2021-04-03T09:22:01Z +issue,50879,Unexpected slow dropout in stacked RNN/LSTM/GRU,https://github.com/pytorch/pytorch/issues/50879,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: rnn",review early; this can change release or regression risk,2021-03-26T01:13:58Z +issue,20497,Statically make `__setstate__` set all attributes/parameters,https://github.com/pytorch/pytorch/issues/20497,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,review early; this can change release or regression risk,2021-03-15T17:48:44Z +issue,53796,foreach ops in autograd,https://github.com/pytorch/pytorch/issues/53796,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,review early; this can change release or regression risk,2021-03-15T17:21:29Z +issue,49877,The same tensor requires more memory on RTX3090,https://github.com/pytorch/pytorch/issues/49877,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",review early; this can change release or regression risk,2021-03-09T21:03:38Z +issue,18300,NCCL backend fails when calling broadcast from different threads,https://github.com/pytorch/pytorch/issues/18300,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",review early; this can change release or regression risk,2021-03-03T08:13:12Z +issue,46491,cannot call rpc.init_rpc twice within a single process,https://github.com/pytorch/pytorch/issues/46491,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",review early; this can change release or regression risk,2021-02-23T02:07:43Z +issue,47844,Incorrect output loss value under specific CUDA version,https://github.com/pytorch/pytorch/issues/47844,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dependency bug, module: linear algebra, module: numerical-stability",review early; this can change release or regression risk,2021-02-11T23:12:17Z +issue,51863,TestVectorizedMemoryAccess.CopyKernel starting to fail after driver to 460.39,https://github.com/pytorch/pytorch/issues/51863,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",review early; this can change release or regression risk,2021-02-10T20:11:19Z +issue,51332,Cant compile from source 1.8.0a0 because 'magma_v2.h' file not found and undefined references to `magma_*',https://github.com/pytorch/pytorch/issues/51332,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: linear algebra, module: magma",review early; this can change release or regression risk,2021-02-08T18:15:40Z +issue,51732,torch.nn.functional.ctc_loss crash(segfault),https://github.com/pytorch/pytorch/issues/51732,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: loss, module: nn",review early; this can change release or regression risk,2021-02-05T19:57:13Z +issue,51083,apex internal assert failed,https://github.com/pytorch/pytorch/issues/51083,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: dependency bug, module: regression",review early; this can change release or regression risk,2021-02-01T18:05:18Z +issue,49758,c10::scalar_to_tensor(...) uses should be audited for performance and type promotion impact,https://github.com/pytorch/pytorch/issues/49758,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",review early; this can change release or regression risk,2021-01-22T18:01:54Z +issue,50038,Torch _remove_batch_dim OP out-of-bounds access,https://github.com/pytorch/pytorch/issues/50038,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: vmap",review early; this can change release or regression risk,2021-01-06T02:25:15Z +issue,42815,Add a guide for what to do when you think there's a memory leak,https://github.com/pytorch/pytorch/issues/42815,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: memory usage",review early; this can change release or regression risk,2020-12-28T18:07:05Z +issue,49459,This error occurs occasionally during the run,https://github.com/pytorch/pytorch/issues/49459,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cudnn",review early; this can change release or regression risk,2020-12-28T17:46:44Z +issue,49392,Uninitialized variable was not detected in ASAN CI config,https://github.com/pytorch/pytorch/issues/49392,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ci,review early; this can change release or regression risk,2020-12-21T18:17:04Z +issue,49473,Pytorch 1.5+: Kernel dies / Segmentation Fault with torch.cuda.mermory_allocated() and torch.cuda.mermory_reserved(),https://github.com/pytorch/pytorch/issues/49473,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: regression",review early; this can change release or regression risk,2020-12-18T02:59:31Z +issue,47116,trying to Installing pytorch on python3.9 via pip results in an non-descript error,https://github.com/pytorch/pytorch/issues/47116,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: releng",review early; this can change release or regression risk,2020-12-11T01:07:44Z +issue,48400,conv_transpose3d returns different result when the input and kernel are mkldnn tensors,https://github.com/pytorch/pytorch/issues/48400,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: dependency bug, module: mkldnn",review early; this can change release or regression risk,2020-12-10T01:27:13Z +issue,38231,Python autograd engine threads never terminate in Python 3.5-3.8,https://github.com/pytorch/pytorch/issues/38231,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: ci, module: crash",review early; this can change release or regression risk,2020-11-16T21:04:05Z +issue,47161,segfault during shutdown with torch1.7,https://github.com/pytorch/pytorch/issues/47161,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: dependency bug, module: pybind, module: third_party",review early; this can change release or regression risk,2020-11-09T18:17:35Z +issue,39061,Some @slowtests are never run in CI,https://github.com/pytorch/pytorch/issues/39061,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: tests,review early; this can change release or regression risk,2020-11-06T19:12:56Z +issue,38623,Providing CUDA tensor to model on CPU causes a crash,https://github.com/pytorch/pytorch/issues/38623,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: rnn",review early; this can change release or regression risk,2020-11-06T19:07:46Z +issue,31558,More dynamic PyTorch APIs,https://github.com/pytorch/pytorch/issues/31558,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: ux,review early; this can change release or regression risk,2020-10-10T08:45:13Z +issue,43005,`torch.distributions.Categorical` crashes with illegal instruction,https://github.com/pytorch/pytorch/issues/43005,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: distributions",review early; this can change release or regression risk,2020-10-08T07:43:38Z +issue,42784,libtorch 1.5 crashes when used on macs when using torch::max without AVX support,https://github.com/pytorch/pytorch/issues/42784,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: crash, module: macos",review early; this can change release or regression risk,2020-10-07T03:49:08Z +issue,40457,DataParallel with Torch 1.5,https://github.com/pytorch/pytorch/issues/40457,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: regression",review early; this can change release or regression risk,2020-09-30T08:20:37Z +issue,43484,pybind11_object_dealloc error,https://github.com/pytorch/pytorch/issues/43484,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: pybind",review early; this can change release or regression risk,2020-09-04T09:23:28Z +issue,35418,PyTorch / libtorch executables fail when built against libcuda stub library,https://github.com/pytorch/pytorch/issues/35418,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",review early; this can change release or regression risk,2020-07-29T13:06:37Z +issue,41226,"The name ""Contiguous"" in `torch::MemoryFormat::Contiguous` can mislead users into assuming contiguous memory",https://github.com/pytorch/pytorch/issues/41226,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: memory format,review early; this can change release or regression risk,2020-07-13T17:09:35Z +issue,5353,Bugs: Score Function approach in REINFORCE for PONG,https://github.com/pytorch/pytorch/issues/5353,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda, module: loss, module: memory usage",review early; this can change release or regression risk,2020-04-20T21:26:45Z +issue,36040,[JIT] Huge delay (1274s vs 0.031s) when running scripted model,https://github.com/pytorch/pytorch/issues/36040,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,review early; this can change release or regression risk,2020-04-17T20:40:23Z +issue,36152,Seg Fault: import vaex with torch,https://github.com/pytorch/pytorch/issues/36152,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: crash",review early; this can change release or regression risk,2020-04-13T17:23:57Z +issue,35597,libtorch_global_deps.so not found.,https://github.com/pytorch/pytorch/issues/35597,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: regression",review early; this can change release or regression risk,2020-03-30T22:24:47Z +issue,32918,Interpolate in the “bicubic” mode with the same shape outputs zeros from second sample onwards,https://github.com/pytorch/pytorch/issues/32918,high-priority/blocker,52,high-priority/blocker/regression label present; old update age or low-urgency/stale label,review early; this can change release or regression risk,2020-03-28T20:52:04Z +issue,28980,Unsafe use of `at::parallel_for` in current codebase,https://github.com/pytorch/pytorch/issues/28980,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: internals, module: multithreading, module: performance",review early; this can change release or regression risk,2019-11-04T18:11:58Z +issue,25518,Assign torch.cuda.FloatTensor to List tensor,https://github.com/pytorch/pytorch/issues/25518,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: cuda",review early; this can change release or regression risk,2019-11-02T07:28:42Z +issue,16953,Seg fault with test_rnn_retain_variables on ppc64le,https://github.com/pytorch/pytorch/issues/16953,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: crash, module: power",review early; this can change release or regression risk,2019-06-17T20:11:52Z +issue,20748,Segmentation fault when use torch::from_blob,https://github.com/pytorch/pytorch/issues/20748,high-priority/blocker,52,"high-priority/blocker/regression label present; old update age or low-urgency/stale label; component/oncall labels: module: abi, module: cpp, module: crash",review early; this can change release or regression risk,2019-05-22T18:57:11Z +pr,189329,[DEBUG] Profile the XPU SYCL device-link bottleneck,https://github.com/pytorch/pytorch/pull/189329,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-09T02:15:47Z +pr,188963,[XPU] Enable XPU testing for 50 test classes across 40 test files,https://github.com/pytorch/pytorch/pull/188963,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-09T02:13:43Z +pr,189229,[DO NOT MERGE][ROCm] Use torch.version.hip to detect ROCm when /opt/rocm is absent,https://github.com/pytorch/pytorch/pull/189229,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/dtensor, ciflow/inductor, ciflow/rocm-nightly","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:50Z +pr,189275,Consolidate CD wheel build pipelines under .ci/wheel/ and unify the numpy pin,https://github.com/pytorch/pytorch/pull/189275,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:40Z +pr,189320,[inductor] Enable nested Inductor configs for invoke_subgraph regions,https://github.com/pytorch/pytorch/pull/189320,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:07Z +pr,189324,[T278829272][Claude Code] Gate TLX flex_attention choices on tlx_mode,https://github.com/pytorch/pytorch/pull/189324,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:05Z +pr,189321,[inductor] Enable per-region cudagraphs for invoke_subgraph via auto graph partition,https://github.com/pytorch/pytorch/pull/189321,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:53:04Z +pr,178393,[DO NOT MERGE] Dynamo 3.15 support,https://github.com/pytorch/pytorch/pull/178393,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-09T01:05:02Z +pr,189299,[MemoryViz] Decimate over-long stacked-area polygons so the summarize…,https://github.com/pytorch/pytorch/pull/189299,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-09T01:04:20Z +pr,179094,[DO NOT MERGE] Add docker image for python 3.15,https://github.com/pytorch/pytorch/pull/179094,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker,"identify whether block is CI, merge conflict, or review gate",2026-07-09T00:21:49Z +pr,189301,[ROCm] Bump AOTriton to 0.13b,https://github.com/pytorch/pytorch/pull/189301,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi300,"identify whether block is CI, merge conflict, or review gate",2026-07-08T23:49:17Z +pr,189092,[dynamo] reduce call_obj_hasattr overrides,https://github.com/pytorch/pytorch/pull/189092,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:53Z +pr,187707,[dynamo] remove GetAttrVariable,https://github.com/pytorch/pytorch/pull/187707,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:45Z +pr,189091,[dynamo] guard getattr() builtin constant-fold fallback,https://github.com/pytorch/pytorch/pull/189091,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:44Z +pr,187531,[dynamo] enable generic attribute mutation on non-UDOV VTs,https://github.com/pytorch/pytorch/pull/187531,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:35Z +pr,187532,[dynamo] fix __getattr__ and __getattribute__ dunder semantics,https://github.com/pytorch/pytorch/pull/187532,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:23:32Z +pr,188469,flex_gemm: support compressed local reduce aux outputs,https://github.com/pytorch/pytorch/pull/188469,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T23:12:49Z +pr,188435,[dynamo] Model uninitialized closure cells in __build_class__,https://github.com/pytorch/pytorch/pull/188435,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T18:44:50Z +pr,189216,[Dynamo] Implement DELETE_DEREF,https://github.com/pytorch/pytorch/pull/189216,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T17:02:56Z +pr,189165,[aot_compile] Add torch.compiler.precompile,https://github.com/pytorch/pytorch/pull/189165,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T16:58:43Z +pr,189237,Add c-shims for aten.unsqueeze and aten.expand,https://github.com/pytorch/pytorch/pull/189237,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-08T14:58:22Z +pr,188016,Update torch-xpu-ops commit pin,https://github.com/pytorch/pytorch/pull/188016,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:52:31Z +pr,189198,[DO NOT MERGE][ROCm] Validate 7.14 skip set on ROCm preview lane with TheRock ROCm deps preloading,https://github.com/pytorch/pytorch/pull/189198,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/dtensor, ciflow/inductor, ciflow/rocm-nightly","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:01:49Z +pr,189218,[DO NOT MERGE][ROCm] Validate 7.14 skip set on TheRock preview lane,https://github.com/pytorch/pytorch/pull/189218,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/dtensor, ciflow/inductor, ciflow/rocm-nightly","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:01:47Z +pr,188720,[ROCm] Enable cholesky_ex via hipSOLVER xgeev (ROCm >= 7.14),https://github.com/pytorch/pytorch/pull/188720,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/rocm-mi200, ciflow/rocm-mi300, ciflow/rocm-nightly","identify whether block is CI, merge conflict, or review gate",2026-07-08T13:01:38Z +pr,189250,[Test][Refactor] Decouple test_dataloader and test_comparison_utils according to the accelerator relevance.,https://github.com/pytorch/pytorch/pull/189250,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T13:01:10Z +pr,188236,[inductor]: Include convolution_backward in decide_layout_opt conv node detection,https://github.com/pytorch/pytorch/pull/188236,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-08T10:30:31Z +pr,189240,[TEST] Refactor test_c10d_logger.py with hw_classification,https://github.com/pytorch/pytorch/pull/189240,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T10:13:21Z +pr,188539,Add support for win_arm64 CUDA builds,https://github.com/pytorch/pytorch/pull/188539,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T10:06:30Z +pr,189227,[xpu][feature]Extend SYCL device-specific math functions to NumericUtils,https://github.com/pytorch/pytorch/pull/189227,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-08T08:40:33Z +pr,189203,[DO NOT MERGE] This is a reference branch to test 7.14,https://github.com/pytorch/pytorch/pull/189203,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor-perf-test-nightly-rocm-mi300, ciflow/inductor-perf-test-nightly-rocm-mi350","identify whether block is CI, merge conflict, or review gate",2026-07-08T06:51:35Z +pr,189084,[DO NOT MERGE] Testing Rocm 7.14,https://github.com/pytorch/pytorch/pull/189084,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor-perf-test-nightly-rocm-mi300, ciflow/inductor-perf-test-nightly-rocm-mi350, ciflow/inductor-rocm-mi200","identify whether block is CI, merge conflict, or review gate",2026-07-08T05:57:52Z +pr,189205,[DO NOT MERGE] Harmless comment just to enable CI,https://github.com/pytorch/pytorch/pull/189205,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor-perf-test-nightly-rocm-mi300, ciflow/inductor-perf-test-nightly-rocm-mi350","identify whether block is CI, merge conflict, or review gate",2026-07-08T04:51:10Z +pr,180186,Try to enable LTO in OSS,https://github.com/pytorch/pytorch/pull/180186,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries,"identify whether block is CI, merge conflict, or review gate",2026-07-08T03:41:17Z +pr,189209,[inductor] Fix with_effects star deps keyed by buffer name instead of operation name,https://github.com/pytorch/pytorch/pull/189209,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T02:23:04Z +pr,182334,[inductor][autotuning] pipeline AsyncCompile + CachingAutotuner (#182334),https://github.com/pytorch/pytorch/pull/182334,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-08T01:01:49Z +pr,163503,[dynamo] prevent recompilation limit exceeded on external_utils.wrap_inline,https://github.com/pytorch/pytorch/pull/163503,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T22:27:47Z +pr,188173,"[dynamo, nested graph breaks] propagate NGB suppression through intermediate inlined functions",https://github.com/pytorch/pytorch/pull/188173,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-07T22:00:17Z +pr,171826,"[dynamo] add break_graph_if_unsupported to COMPARE_OP, BINARY_*, and INPLACE_* instructions",https://github.com/pytorch/pytorch/pull/171826,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-07T20:52:40Z +pr,187650,[Pytorch Test] Add skip_if_lt_x_devices and migrate FSDP autograd tests to be device-agnostic,https://github.com/pytorch/pytorch/pull/187650,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-07T12:35:34Z +pr,181728,[xpu][4/4]Enable MXFP8/MXFP4/NVFP4 tests for XPU,https://github.com/pytorch/pytorch/pull/181728,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-07-07T10:42:46Z +pr,183613,[CD] Tighten ROCm manywheel build pipeline,https://github.com/pytorch/pytorch/pull/183613,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries,"identify whether block is CI, merge conflict, or review gate",2026-07-07T08:22:23Z +pr,189128,Gil acquire bool conversion,https://github.com/pytorch/pytorch/pull/189128,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-07T08:21:34Z +pr,189070,Accept device-qualified gloo backend in monitored_barrier,https://github.com/pytorch/pytorch/pull/189070,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-07T04:25:35Z +pr,189073,Make subgroup name hash salt collective-consistent,https://github.com/pytorch/pytorch/pull/189073,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-07T04:25:34Z +pr,189074,Deduplicate comms when destroying a process group,https://github.com/pytorch/pytorch/pull/189074,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-07T04:25:33Z +pr,189072,Create TorchComms subgroups on the correct device,https://github.com/pytorch/pytorch/pull/189072,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-07T04:25:33Z +pr,189071,Remove new_group to split_group delegation on the TorchComms path,https://github.com/pytorch/pytorch/pull/189071,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-07T04:25:30Z +pr,189069,Route only TorchComms-owned backends through comms,https://github.com/pytorch/pytorch/pull/189069,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-07T01:03:17Z +pr,189068,Register nccl-lazy TorchComms backend,https://github.com/pytorch/pytorch/pull/189068,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-07T00:07:18Z +pr,182864,[donotrvw][broken][WIP] spmd_types + PP: _SpmdTypesMeta and nullcontext decorator fix,https://github.com/pytorch/pytorch/pull/182864,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-06T23:51:34Z +pr,185208,[PoC] uv.lock dependency-groups for runtime + build deps,https://github.com/pytorch/pytorch/pull/185208,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel, ciflow/docker","identify whether block is CI, merge conflict, or review gate",2026-07-06T20:14:39Z +pr,182844,[CI] Pin miniforge to 26.1.1-3 to fix docker builds,https://github.com/pytorch/pytorch/pull/182844,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-06T19:08:18Z +pr,181215,Add cooperative GPU-sharing infrastructure via CUDA process-checkpoint,https://github.com/pytorch/pytorch/pull/181215,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-06T19:08:17Z +pr,188119,[inductor] Enable combo kernel compile time autotuning by default,https://github.com/pytorch/pytorch/pull/188119,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T12:16:07Z +pr,188326,feat(dynamo): backport recursive dict tags,https://github.com/pytorch/pytorch/pull/188326,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T11:07:14Z +pr,188203,[inductor] Fix combo-kernel peak-memory KeyError when reorder_for_peak_memory is disabled,https://github.com/pytorch/pytorch/pull/188203,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T09:15:11Z +pr,189003,[inductor] Coordinate-descent tune compile-time combo kernel blocks,https://github.com/pytorch/pytorch/pull/189003,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T09:15:09Z +pr,189002,[inductor] Don't combo-fuse data-independent producers feeding attention,https://github.com/pytorch/pytorch/pull/189002,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T09:15:09Z +pr,189001,[inductor] Fix flaky combo-kernel peak-memory test,https://github.com/pytorch/pytorch/pull/189001,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-06T09:15:08Z +pr,188886,[AOTI] Support PDL launch attributes via `cuLaunchKernelEx` in CUDA wrappers,https://github.com/pytorch/pytorch/pull/188886,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T07:51:18Z +pr,181324,"[dynamo] Remove TypingVariable, trace more of typing.py",https://github.com/pytorch/pytorch/pull/181324,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T06:01:50Z +pr,181611,[don't merge] test combo kernels,https://github.com/pytorch/pytorch/pull/181611,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-06T02:20:20Z +pr,174503,[Draft]Trying to see if cpp modification would work,https://github.com/pytorch/pytorch/pull/174503,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-05T20:49:00Z +pr,181205,"NOT MERGE, JUST FOR TEST",https://github.com/pytorch/pytorch/pull/181205,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-07-05T07:38:56Z +pr,182143,[OSDC] Migrate periodic.yml to OSDC via dial-up pattern,https://github.com/pytorch/pytorch/pull/182143,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/periodic,"identify whether block is CI, merge conflict, or review gate",2026-07-05T02:19:58Z +pr,182314,[OSDC] Migrate dynamo-unittest.yml to OSDC via dial-up pattern,https://github.com/pytorch/pytorch/pull/182314,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dynamo,"identify whether block is CI, merge conflict, or review gate",2026-07-05T01:09:32Z +pr,180875,[CUDA][cuBLASLt] torch.addmm - improve numerics of bias fusion with reduced precision SplitK algorithms,https://github.com/pytorch/pytorch/pull/180875,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-04T11:55:40Z +pr,182415,[c10d] Apply record_comm names to NCCL launch annotations,https://github.com/pytorch/pytorch/pull/182415,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-07-04T09:18:40Z +pr,182389,Collective buffer detection for DDP and FSDP,https://github.com/pytorch/pytorch/pull/182389,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-04T02:20:07Z +pr,182388,Training hooks and public API for memory component tracking,https://github.com/pytorch/pytorch/pull/182388,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-04T02:20:06Z +pr,182387,Allocator-level memory attribution primitives,https://github.com/pytorch/pytorch/pull/182387,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-04T02:20:05Z +pr,181412,Add Claude-powered operator benchmark triage,https://github.com/pytorch/pytorch/pull/181412,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-03T17:59:34Z +pr,182226,[CI] Install git-core PPA without add-apt-repository,https://github.com/pytorch/pytorch/pull/182226,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-03T16:04:25Z +pr,188887,Enable cpp_wrapper for mixed order reduction,https://github.com/pytorch/pytorch/pull/188887,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-03T14:13:29Z +pr,188858,[DO NOT MERGE] Complex Wrapper Test,https://github.com/pytorch/pytorch/pull/188858,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-03T05:47:09Z +pr,188734,Add tracer kwarg to torch.compile for make_fx tracing,https://github.com/pytorch/pytorch/pull/188734,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-07-03T04:22:06Z +pr,186945,Enable nested Inductor configs for invoke_subgraph regions,https://github.com/pytorch/pytorch/pull/186945,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T22:50:21Z +pr,182810,[ROCm] Implement TORCH_HIP_VERSION,https://github.com/pytorch/pytorch/pull/182810,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/rocm-mi300, ciflow/rocm-mi355, ciflow/rocm-nightly, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-02T19:47:46Z +pr,188787,OneDNN 3.12.2 acceptance test,https://github.com/pytorch/pytorch/pull/188787,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel,"identify whether block is CI, merge conflict, or review gate",2026-07-02T15:31:04Z +pr,182691,Parameterized cuda graph launch 3,https://github.com/pytorch/pytorch/pull/182691,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T22:03:04Z +pr,187909,[DONT MERGE] Test new xpu runners,https://github.com/pytorch/pytorch/pull/187909,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-01T21:29:39Z +pr,188136,[BE] Build all XPU wheels on a single runner,https://github.com/pytorch/pytorch/pull/188136,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel,"identify whether block is CI, merge conflict, or review gate",2026-07-01T18:10:57Z +pr,188662,[ROCm] Updated mi300 runners to have 8 shards,https://github.com/pytorch/pytorch/pull/188662,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-rocm-mi300, ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-07-01T17:27:58Z +pr,188284,[Draft] Fix BUILD_SEPARATE_OPS cyclic dependency — register with xpu_hal.dll,https://github.com/pytorch/pytorch/pull/188284,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-07-01T17:27:47Z +pr,188629,docs: document S3-compatible object storage for distributed checkpoint,https://github.com/pytorch/pytorch/pull/188629,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-01T14:03:19Z +pr,188577,Migrate Build Triton Wheel to OSDC,https://github.com/pytorch/pytorch/pull/188577,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-07-01T06:46:07Z +pr,188259,Add XPU caching allocator checkpoint pool state bindings,https://github.com/pytorch/pytorch/pull/188259,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-07-01T01:30:26Z +pr,188592,vllm wheel 0630,https://github.com/pytorch/pytorch/pull/188592,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T23:51:15Z +pr,187206,[WIP][no-ci] auto_custom_op,https://github.com/pytorch/pytorch/pull/187206,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-30T21:46:47Z +pr,187878,Add c10d Watchdog: a libuv-backed timeout service,https://github.com/pytorch/pytorch/pull/187878,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-30T19:51:50Z +pr,188429,[ROCm][CI] Switch rocm-preview to TheRock wheels,https://github.com/pytorch/pytorch/pull/188429,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/rocm-nightly, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-30T17:53:53Z +pr,188463,[PU1] Add output_size parameter to torch.bincount for eager-meta support.,https://github.com/pytorch/pytorch/pull/188463,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-30T00:05:47Z +pr,187870,compile-on-one-rank: device-agnostic inductor codegen and launcher,https://github.com/pytorch/pytorch/pull/187870,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-29T22:37:25Z +pr,188464,[spmd_types] infer R@FSDP axes from active mesh for empty params,https://github.com/pytorch/pytorch/pull/188464,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-29T21:44:37Z +pr,188448,Add DTensor sharding strategy for searchsorted,https://github.com/pytorch/pytorch/pull/188448,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-29T21:01:41Z +pr,188404,Inductor: Add metadata overlap dependency implementation,https://github.com/pytorch/pytorch/pull/188404,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-29T14:44:45Z +pr,181678,[spmd_types] CI CPU job w/o spmd_types,https://github.com/pytorch/pytorch/pull/181678,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T10:04:20Z +pr,181956,[do-not-review] cudagraph while node support,https://github.com/pytorch/pytorch/pull/181956,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-29T06:07:05Z +pr,187900,[XPU] Migrate CI Docker to Intel OMIX for oneAPI + GPU driver installation,https://github.com/pytorch/pytorch/pull/187900,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-06-29T05:29:36Z +pr,181444,Scaled cublas grouped gemm sm90,https://github.com/pytorch/pytorch/pull/181444,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-28T23:52:40Z +pr,181443,Cublas grouped gemm nvfp4,https://github.com/pytorch/pytorch/pull/181443,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-28T23:52:39Z +pr,181627,[ ROCm] [ DRAFT ] 181288 + 177219,https://github.com/pytorch/pytorch/pull/181627,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/op-benchmark,"identify whether block is CI, merge conflict, or review gate",2026-06-27T18:54:03Z +pr,181628,[ ROCm] [ DRAFT ] 181288 + 177219 + 181615,https://github.com/pytorch/pytorch/pull/181628,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/op-benchmark,"identify whether block is CI, merge conflict, or review gate",2026-06-27T17:50:58Z +pr,177612,[inductor] Add pattern match for fused gate+up GEMM + SiLU on XPU,https://github.com/pytorch/pytorch/pull/177612,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-pallas, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-27T12:55:21Z +pr,181615,[ROCm] CI: skip operator matmul microbenchmark,https://github.com/pytorch/pytorch/pull/181615,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/op-benchmark,"identify whether block is CI, merge conflict, or review gate",2026-06-27T05:48:57Z +pr,181664,[FSDP2] Design notes: options for RS/AR back-edge in HSDP,https://github.com/pytorch/pytorch/pull/181664,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-27T02:21:55Z +pr,186271,[CUDA] Add TunableOp support for scaled GEMM,https://github.com/pytorch/pytorch/pull/186271,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-27T01:21:00Z +pr,182271,[ci][WIP] Cut redundant before-test fanout + trim over-eager periodic crons,https://github.com/pytorch/pytorch/pull/182271,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-27T01:01:49Z +pr,177038,Add cublaslt backend for scaled grouped GEMM,https://github.com/pytorch/pytorch/pull/177038,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-26T23:50:32Z +pr,172682,[NOT FOR MERGE] Initial test of features disabled when doing cooperative reduction,https://github.com/pytorch/pytorch/pull/172682,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-26T20:50:46Z +pr,187981,[not for land] CODA backward epilogue fusion prototype,https://github.com/pytorch/pytorch/pull/187981,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-26T18:08:13Z +pr,184892,xpu: use stride-based conv descriptors and preserve grouped layout mapping,https://github.com/pytorch/pytorch/pull/184892,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-06-26T16:49:16Z +pr,187606,"[inductor] Per-arch ROCm channels_last layout-opt multipliers (gfx942, gfx950)",https://github.com/pytorch/pytorch/pull/187606,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-perf-test-nightly-rocm-mi300, ciflow/inductor-rocm-mi300, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-26T06:23:55Z +pr,188184,Update definition of ROCM_VERSION in cpp_extension.py,https://github.com/pytorch/pytorch/pull/188184,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-25T22:04:23Z +pr,188189,Fix split_group non-member handling regression for mixed backends,https://github.com/pytorch/pytorch/pull/188189,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-25T20:57:34Z +pr,188168,Fix hipBLASLt FP32 precision degradation on MI300X (ROCM-3139),https://github.com/pytorch/pytorch/pull/188168,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-25T18:20:01Z +pr,178862,[FSDP] enable async H2D prefetching for world_size=1,https://github.com/pytorch/pytorch/pull/178862,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-25T17:35:10Z +pr,185258,[ROCm][DO NOT MERGE][DEBUGGING] MI355 hang reproduce and dump subprocess logs,https://github.com/pytorch/pytorch/pull/185258,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-25T04:53:37Z +pr,187882,[RFC] Specialize float attributes of nn.Modules,https://github.com/pytorch/pytorch/pull/187882,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-25T01:17:23Z +pr,187464,[ROCm] [Operator Benchmark] Add gfx950 ROCm nightly CI lane and docker image,https://github.com/pytorch/pytorch/pull/187464,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/op-benchmark","identify whether block is CI, merge conflict, or review gate",2026-06-24T15:35:10Z +pr,187818,[WIP] generalize unitttests,https://github.com/pytorch/pytorch/pull/187818,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-24T13:46:58Z +pr,187739,[ROCm][CI] Misc updates,https://github.com/pytorch/pytorch/pull/187739,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-24T08:36:03Z +pr,187971,[inductor] Add strict_numerics config to emit INNER_TREE reduction ordering for sum,https://github.com/pytorch/pytorch/pull/187971,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-24T06:07:52Z +pr,181199,[spmd types] fully_shard in spmd_types,https://github.com/pytorch/pytorch/pull/181199,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T05:52:05Z +pr,188003,[AMD][hipBLASLt] Enable streak-K work stealing for GEMM operations,https://github.com/pytorch/pytorch/pull/188003,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-24T03:00:52Z +pr,181449,AOTI: Add use_fake_constants mode for zero-weight compilation,https://github.com/pytorch/pytorch/pull/181449,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-24T01:19:07Z +pr,175486,[WIP][inductor] Layout allocator approach for symm_mem graph inputs,https://github.com/pytorch/pytorch/pull/175486,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-23T23:49:13Z +pr,181392,[AC] Add example_inputs param to remat pass and mark recomputed nodes as backward,https://github.com/pytorch/pytorch/pull/181392,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T19:14:45Z +pr,186033,[CD] Enable Python 3.15 & 3.15t Windows wheel builds and Triton XPU Windows build,https://github.com/pytorch/pytorch/pull/186033,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel,"identify whether block is CI, merge conflict, or review gate",2026-06-23T13:24:37Z +pr,181218,FSDP2 chunked-loss ROCm race: stream-level wait_event at post_backward exit,https://github.com/pytorch/pytorch/pull/181218,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/rocm-mi300, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T05:53:04Z +pr,181269,[draft] refactor linux workflows,https://github.com/pytorch/pytorch/pull/181269,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-23T05:53:03Z +pr,187815,[Inductor][XPU] Unfuse addmm for all half-dtype inputs on XPU,https://github.com/pytorch/pytorch/pull/187815,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T02:58:19Z +pr,179149,[Inductor][RFC] Symbolic Analysis of User-Defined Triton Kernels,https://github.com/pytorch/pytorch/pull/179149,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-23T01:21:13Z +pr,187892,"[profiler][cupti] Perfetto-native (.pftrace) export, full chrome-args parity",https://github.com/pytorch/pytorch/pull/187892,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-23T00:28:36Z +pr,181278,[spmd_types][unused] local_map for spmd_types tensors,https://github.com/pytorch/pytorch/pull/181278,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T22:13:56Z +pr,187258,[aot_compile] Add torch.compiler.precompile,https://github.com/pytorch/pytorch/pull/187258,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T20:16:20Z +pr,187840,[inductor] Add allow_peak_memory_increasing_fusion config,https://github.com/pytorch/pytorch/pull/187840,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T19:03:50Z +pr,187841,[inductor] Avoid disjoint horizontal fusion when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/187841,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T19:03:46Z +pr,187839,[inductor] Resolve ProcessGroup attrs in overlap estimates,https://github.com/pytorch/pytorch/pull/187839,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T19:03:41Z +pr,187843,[inductor] Avoid late-output vertical fusion when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/187843,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T19:03:38Z +pr,187846,[inductor] Pick low-pressure buffer reuse when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/187846,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T19:03:37Z +pr,187845,[inductor] Prefer successor-unblocking LPMF when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/187845,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T19:03:30Z +pr,187844,[inductor] Realize reused pointwise add when allow_peak_memory_increasing_fusion is off,https://github.com/pytorch/pytorch/pull/187844,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T17:48:12Z +pr,180986,Enable nested_compile_region for AOTInductor + tensor-subclass params,https://github.com/pytorch/pytorch/pull/180986,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T15:18:58Z +pr,180754,[ROCm] Fix decorator for skipping FP8 tests,https://github.com/pytorch/pytorch/pull/180754,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi200","identify whether block is CI, merge conflict, or review gate",2026-06-22T11:23:18Z +pr,181189,Tensor.record_use: precise cross-stream lifetime for the CUDA caching allocator,https://github.com/pytorch/pytorch/pull/181189,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-22T08:34:15Z +pr,180867,Use SavedTensor for SAC cached tensors to enable saved tensor hooks,https://github.com/pytorch/pytorch/pull/180867,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-21T23:56:02Z +pr,187559,[MPS] Migrate nll_loss to native Metal kernels,https://github.com/pytorch/pytorch/pull/187559,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-21T22:38:31Z +pr,180866,Add save_tensor() API for selective activation checkpointing,https://github.com/pytorch/pytorch/pull/180866,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-21T21:58:40Z +pr,181047,[FSDP2] Introduce StreamHandoff; migrate AR keep-alive,https://github.com/pytorch/pytorch/pull/181047,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-21T11:31:47Z +pr,180933,[spmd_types] fully_shard,https://github.com/pytorch/pytorch/pull/180933,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-20T23:52:43Z +pr,187756,[testintro][DEMO - DO NOT LAND] exercise the test-diff CI comment,https://github.com/pytorch/pytorch/pull/187756,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-20T21:52:03Z +pr,180924,Remove allow_cache_entry_mutation from SAC,https://github.com/pytorch/pytorch/pull/180924,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-20T20:02:56Z +pr,180925,[dynamo] Extract dynamic graph break hints,https://github.com/pytorch/pytorch/pull/180925,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-20T02:25:37Z +pr,187479,Add Python c10d Backend trampoline,https://github.com/pytorch/pytorch/pull/187479,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-19T00:29:35Z +pr,187624,[DRAFT]codeowner updates finer grained groups,https://github.com/pytorch/pytorch/pull/187624,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-18T22:26:02Z +pr,187674,Preserve boxed lifetimes for regional compiled calls,https://github.com/pytorch/pytorch/pull/187674,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-18T19:37:05Z +pr,184886,[inductor] Diagonal-symmetric mm TritonTemplate for Gram matrices,https://github.com/pytorch/pytorch/pull/184886,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-18T17:03:20Z +pr,187435,[inductor] Support control_deps as fusion barriers,https://github.com/pytorch/pytorch/pull/187435,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-18T16:30:22Z +pr,187636,[AOTI] Flip the default value of autotune_at_compile_time to False,https://github.com/pytorch/pytorch/pull/187636,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-18T07:07:07Z +pr,187640,Support in-tree XNNPACK build on Windows ARM64,https://github.com/pytorch/pytorch/pull/187640,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-18T04:46:24Z +pr,187472,[dynamo] don't reconstruct live torch.Stream values as StreamContext across graph breaks,https://github.com/pytorch/pytorch/pull/187472,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-17T00:00:58Z +pr,180684,AOTI: Write tensor data directly to file without Python bytes copy,https://github.com/pytorch/pytorch/pull/180684,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T20:31:37Z +pr,180683,AOTI: Stream external weights to disk instead of in-memory blob,https://github.com/pytorch/pytorch/pull/180683,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T18:53:41Z +pr,180682,AOTI: Only clone mutated buffers in _unlift_graph,https://github.com/pytorch/pytorch/pull/180682,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T18:53:41Z +pr,185612,[WIP][CI][CRCR] Update mergebot rule to adapt CRCR L3,https://github.com/pytorch/pytorch/pull/185612,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-16T15:35:35Z +pr,177648,LinAlgUtils -- update the toolkit to address the MathLibs dispatch with redundant memory allocations/movements.,https://github.com/pytorch/pytorch/pull/177648,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-16T12:23:27Z +pr,186752,[Inductor][Triton] Add host-side TMA support for B200 GEMM templates,https://github.com/pytorch/pytorch/pull/186752,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-16T08:19:37Z +pr,186756,[Inductor][Host-side TMA GEMM] Enable host-side TMA for TMA epilogue store,https://github.com/pytorch/pytorch/pull/186756,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-16T08:19:28Z +pr,186988,Move header-only c10/util Layer 1 utils to torch/headeronly,https://github.com/pytorch/pytorch/pull/186988,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-16T06:07:48Z +pr,186987,Move header-only c10/util large/vendored headers to torch/headeronly,https://github.com/pytorch/pytorch/pull/186987,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-16T06:07:36Z +pr,186986,Move header-only c10/util string/filesystem utils to torch/headeronly,https://github.com/pytorch/pytorch/pull/186986,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-16T06:06:59Z +pr,186984,Move header-only c10/util container/storage utils to torch/headeronly,https://github.com/pytorch/pytorch/pull/186984,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-16T06:06:56Z +pr,180423,Fix checkpoint DeviceContext leak onto TorchFunction mode stack,https://github.com/pytorch/pytorch/pull/180423,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-06-16T03:48:36Z +pr,183579,[OpenReg][Quantization] Register dequantize_tensor_per_tensor_affine stub for PrivateUse1,https://github.com/pytorch/pytorch/pull/183579,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-16T02:38:34Z +pr,183270,[inductor][overlap] Post-scheduling foreach_mm for optimizer mm batching,https://github.com/pytorch/pytorch/pull/183270,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-15T22:49:47Z +pr,182631,[aten] Add aten._foreach_mm operator,https://github.com/pytorch/pytorch/pull/182631,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-15T22:47:50Z +pr,180597,[nocommit] symmem custreamwaitvalue32,https://github.com/pytorch/pytorch/pull/180597,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem,"identify whether block is CI, merge conflict, or review gate",2026-06-15T22:18:01Z +pr,180407,[nocommit] symmem nccl ib,https://github.com/pytorch/pytorch/pull/180407,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/h100-symm-mem,"identify whether block is CI, merge conflict, or review gate",2026-06-15T22:18:00Z +pr,180411,stateless dtensor,https://github.com/pytorch/pytorch/pull/180411,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-15T01:36:39Z +pr,180409,stateless ghstack,https://github.com/pytorch/pytorch/pull/180409,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-14T21:56:10Z +pr,180410,debug,https://github.com/pytorch/pytorch/pull/180410,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-14T21:56:09Z +pr,177438,Infer device_type in fork_rng from devices argument,https://github.com/pytorch/pytorch/pull/177438,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-14T18:55:52Z +pr,180412,stateless spmd,https://github.com/pytorch/pytorch/pull/180412,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-14T06:06:23Z +pr,187249,perf(dynamo): add guard partial memo,https://github.com/pytorch/pytorch/pull/187249,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-13T16:44:55Z +pr,177401,Allow non-CL images to go through NEON path in `F.interpolate()` for bilinear and bicubic modes when antialias=True.,https://github.com/pytorch/pytorch/pull/177401,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-13T15:10:10Z +pr,187056,[TESTING] Separated ROCm triton commit pin,https://github.com/pytorch/pytorch/pull/187056,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/inductor-rocm, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-13T09:03:51Z +pr,180207,[TEST] Refactor test_structured_trace.py to be device-generic,https://github.com/pytorch/pytorch/pull/180207,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-13T07:47:52Z +pr,187116,Add stable tie handling to topk,https://github.com/pytorch/pytorch/pull/187116,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-13T02:08:04Z +pr,177471,tag on ops,https://github.com/pytorch/pytorch/pull/177471,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-12T22:03:54Z +pr,187050,[TESTING] [Inductor] Add regression tests for large-batch BMM grid Y/Z split,https://github.com/pytorch/pytorch/pull/187050,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-rocm, ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-12T18:01:04Z +pr,186343,Enable FA build for SDPA on Windows+CUDA,https://github.com/pytorch/pytorch/pull/186343,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel,"identify whether block is CI, merge conflict, or review gate",2026-06-12T11:45:28Z +pr,179675,[WIP]port dynamo cases to Intel GPU,https://github.com/pytorch/pytorch/pull/179675,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-06-12T08:05:06Z +pr,180192,[DO NOT REVIEW] fix torchtitan CI error,https://github.com/pytorch/pytorch/pull/180192,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-12T03:48:16Z +pr,187004,perf(dynamo): add guard lookup fast paths,https://github.com/pytorch/pytorch/pull/187004,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-11T14:53:09Z +pr,186753,[fx] Promote raw-SymInt arg warning to RuntimeError,https://github.com/pytorch/pytorch/pull/186753,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-10T19:35:55Z +pr,178485,[DO NOT MERGE][Inductor] Migrate oneDNN Inner Product to oneDNN MatMul for MKLDNN linear ops,https://github.com/pytorch/pytorch/pull/178485,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-10T12:04:13Z +pr,186590,[Draft][xpu] Add NVFP4 support on Intel GPU,https://github.com/pytorch/pytorch/pull/186590,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-06-10T08:10:27Z +pr,186209,Unify triton-xpu CD nightly build into 'triton' package with version suffix,https://github.com/pytorch/pytorch/pull/186209,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-10T06:01:31Z +pr,186835,[RFC]: Auto-select and run Operator Performance Benchmarks from PR Diffs,https://github.com/pytorch/pytorch/pull/186835,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-10T02:01:33Z +pr,178277,Skip fake autocast tests for `linalg.pinv` and `pinverse` on XPU,https://github.com/pytorch/pytorch/pull/178277,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-06-09T23:09:15Z +pr,186798,[TESTING ONLY] [ROCm] testing heuristic config approach,https://github.com/pytorch/pytorch/pull/186798,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-perf-test-nightly-rocm-mi350, ciflow/inductor-rocm-mi355, ciflow/rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-06-09T23:08:29Z +pr,186541,[MPS] [WIP] Port all MPSGraph matmuls to Metal,https://github.com/pytorch/pytorch/pull/186541,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-06-09T19:15:22Z +pr,186378,[activation_checkpointing] Support AutoNamingMode + policy name keys under torch.compile,https://github.com/pytorch/pytorch/pull/186378,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-05T19:51:42Z +pr,186382,[activation_checkpointing] AutoNamingMode + name/module-FQN keys in checkpoint policy= (eager),https://github.com/pytorch/pytorch/pull/186382,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-05T19:51:35Z +pr,186277,[activation_checkpointing] Add policy= kwarg and name() to torch.utils.checkpoint,https://github.com/pytorch/pytorch/pull/186277,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-05T19:09:33Z +pr,185555,Add torch._rust Rust extension and build plumbing (#185555),https://github.com/pytorch/pytorch/pull/185555,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/inductor","identify whether block is CI, merge conflict, or review gate",2026-06-05T06:35:13Z +pr,179769,[cuda graphs] Integrate kernel annotations into inductor cudagraph trees,https://github.com/pytorch/pytorch/pull/179769,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-04T17:45:42Z +pr,186002,"Temporary Commit at 6/2/2026, 2:55:48 PM",https://github.com/pytorch/pytorch/pull/186002,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-03T03:45:31Z +pr,185935,[DO NOT MERGE] ci: add manywheel build telemetry (May 6f7bb01c),https://github.com/pytorch/pytorch/pull/185935,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-02T18:31:35Z +pr,185822,[DO NOT MERGE] ci: add manywheel build telemetry (Feb baseline 0cd9ecaf),https://github.com/pytorch/pytorch/pull/185822,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/binaries_wheel, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-02T14:54:15Z +pr,185936,[DO NOT MERGE] ci: add manywheel build telemetry (Apr midpoint 977c5623),https://github.com/pytorch/pytorch/pull/185936,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-02T14:31:34Z +pr,185706,Fix typos across comments and docstrings in torch,https://github.com/pytorch/pytorch/pull/185706,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-06-02T14:01:32Z +pr,181578,[Draft] [04/27/26] Dashboard result update for ROCm,https://github.com/pytorch/pytorch/pull/181578,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/inductor-periodic, ciflow/inductor-rocm-mi300, ciflow/inductor-rocm-mi355","identify whether block is CI, merge conflict, or review gate",2026-06-02T11:45:26Z +pr,183626,[Inductor] Seed combo-kernel autotune from standalone subkernels,https://github.com/pytorch/pytorch/pull/183626,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-06-02T05:18:51Z +pr,185835,[c10d] Add delegate backend to FakeProcessGroup for numerics testing,https://github.com/pytorch/pytorch/pull/185835,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-01T21:41:43Z +pr,185286,Add version counter check for gradient tensors in the autograd engine,https://github.com/pytorch/pytorch/pull/185286,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-06-01T14:40:50Z +pr,185787,[WIP] Migrate XPU test runners to linux.xpu.test and disable legacy jobs,https://github.com/pytorch/pytorch/pull/185787,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/xpu,"identify whether block is CI, merge conflict, or review gate",2026-06-01T07:12:16Z +pr,185712,"Fix typos in comments, docstrings, and strings across the codebase",https://github.com/pytorch/pytorch/pull/185712,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-30T20:18:47Z +pr,185713,"Fix typos in comments, docstrings, and error messages",https://github.com/pytorch/pytorch/pull/185713,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/h100-symm-mem, ciflow/inductor, ciflow/mps","identify whether block is CI, merge conflict, or review gate",2026-05-30T19:27:43Z +pr,185705,"Fix typos in comments, docstrings, and error messages",https://github.com/pytorch/pytorch/pull/185705,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-30T16:46:33Z +pr,185565,upstream torch.comms,https://github.com/pytorch/pytorch/pull/185565,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/docker, ciflow/h100-distributed, ciflow/h100-symm-mem, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-30T09:28:11Z +pr,185339,[not-for-land][aten] Add cuBLAS/cublasLt grouped GEMM backends for _foreach_mm,https://github.com/pytorch/pytorch/pull/185339,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-29T22:43:48Z +pr,185548,[test] Migrate in-test OpInfo skip to OpInfo skip DB,https://github.com/pytorch/pytorch/pull/185548,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-29T06:28:59Z +pr,177445,Add SymInt and SymFloat to Number,https://github.com/pytorch/pytorch/pull/177445,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-29T02:21:36Z +pr,185364,[not-for-land][native_dsl] Add _grouped_mm_from_ptrs pointer-packing path,https://github.com/pytorch/pytorch/pull/185364,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-28T08:58:03Z +pr,185172,Add support to custom devices to operator_benchmark.,https://github.com/pytorch/pytorch/pull/185172,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/op-benchmark,"identify whether block is CI, merge conflict, or review gate",2026-05-28T08:57:01Z +pr,185363,[not-for-land][native_dsl] Python _foreach_mm override via stack + _grouped_mm,https://github.com/pytorch/pytorch/pull/185363,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-28T07:42:14Z +pr,185457,[Inductor][xpu] Enable grouped_mm Triton template on XPU with TMA disabled,https://github.com/pytorch/pytorch/pull/185457,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-28T07:41:19Z +pr,185376,[not-for-land][native_dsl] Add _foreach_mm_from_ptrs pointer-bridge and py_ptrs override,https://github.com/pytorch/pytorch/pull/185376,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-27T23:06:36Z +pr,184302,Add Stream-K MM backend support to Inductor,https://github.com/pytorch/pytorch/pull/184302,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/inductor-rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-27T21:13:51Z +pr,185340,[not-for-land][native_dsl] Add Python _foreach_mm override via _grouped_mm,https://github.com/pytorch/pytorch/pull/185340,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-27T15:01:38Z +pr,185284,Add checkpoint_name() API for explicit tensor naming in SAC,https://github.com/pytorch/pytorch/pull/185284,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-27T01:25:54Z +pr,185287,Add memory_budget() context manager for per-region activation memory budgets,https://github.com/pytorch/pytorch/pull/185287,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-27T01:25:20Z +pr,185285,Allow debug mode with context_fn in checkpoint(),https://github.com/pytorch/pytorch/pull/185285,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-05-27T00:21:48Z +pr,184991,[inductor] Add memory-aware graph-level CSE pass,https://github.com/pytorch/pytorch/pull/184991,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-26T22:23:37Z +pr,185253,[c10d] Make FakeProcessGroup reduce collectives numerically consistent,https://github.com/pytorch/pytorch/pull/185253,draft/noise,51,PR is draft; failed checks or blocked/unstable merge state,"identify whether block is CI, merge conflict, or review gate",2026-05-26T19:37:58Z +pr,185219,[cherry-pick] no topological sort in manual overlaps,https://github.com/pytorch/pytorch/pull/185219,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-26T12:36:13Z +pr,185218,[not-for-land][squash] Regional inductor cache fix,https://github.com/pytorch/pytorch/pull/185218,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-26T12:36:12Z +pr,159718,[Don't Review] Test XPU CI,https://github.com/pytorch/pytorch/pull/159718,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-05-26T10:08:57Z +pr,184909,[Inductor][Draft] A Helion backend for Inductor,https://github.com/pytorch/pytorch/pull/184909,draft/noise,51,"PR is draft; failed checks or blocked/unstable merge state; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-25T23:36:49Z +issue,171675,PyTorch quantization `QConfig` triggers INTERNAL ASSERT FAILED when printing config with invalid observer `dtype`,https://github.com/pytorch/pytorch/issues/171675,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2026-01-06T04:13:26Z +issue,171516,How to verify that default_decompositions successfully reduce operators to the Core ATen IR set?,https://github.com/pytorch/pytorch/issues/171516,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2026-01-05T16:23:29Z +issue,171377,PyTorch JIT `get_debug_state()` triggers INTERNAL ASSERT FAILED after `torch.jit.trace`,https://github.com/pytorch/pytorch/issues/171377,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-01-05T14:58:15Z +issue,154613,`torch.jit.script` doesn't accept `axis` as alias for `dim`,https://github.com/pytorch/pytorch/issues/154613,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2026-01-01T19:19:47Z +issue,170548,unable to export a model with bucketize,https://github.com/pytorch/pytorch/issues/170548,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-12-23T01:32:59Z +issue,170783,New higher order ops to handle loops,https://github.com/pytorch/pytorch/issues/170783,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-12-21T04:07:21Z +issue,170680,Error when composing jit.script and jacrev,https://github.com/pytorch/pytorch/issues/170680,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-12-20T23:38:52Z +issue,170673,`torch.export` Fails with RuntimeError when Accessing Parameter via `.data` in `forward`,https://github.com/pytorch/pytorch/issues/170673,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-12-20T23:36:42Z +issue,170774,CUPTI metrics not recorded in profiler,https://github.com/pytorch/pytorch/issues/170774,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-12-20T23:06:06Z +issue,170867,Operator benchmark: option to measure GPU execution time only (less CPU noise),https://github.com/pytorch/pytorch/issues/170867,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-12-20T22:52:15Z +issue,170319,torch.profiler shows GPU ops (Triton / FlashAttention) to have `cpu_op` category for some reason,https://github.com/pytorch/pytorch/issues/170319,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-12-17T04:29:26Z +issue,170169,questions about default_weight_fake_quant config with pytorch quantizations,https://github.com/pytorch/pytorch/issues/170169,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-12-11T16:51:50Z +issue,170137,PyTorch memory profiler conflicts with torch.util.swap_tensors due to strict reference counter check,https://github.com/pytorch/pytorch/issues/170137,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: python frontend, oncall: profiler",categorize and decide next maintainer owner/action,2025-12-11T04:40:15Z +issue,161179,Consolidate magma-cuda build for Linux and Windows,https://github.com/pytorch/pytorch/issues/161179,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",categorize and decide next maintainer owner/action,2025-12-08T20:38:41Z +issue,169666,Inconsistent`flop_counter.get_total_flops()` results for a custom decoder model,https://github.com/pytorch/pytorch/issues/169666,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-12-05T15:10:05Z +issue,165829,Floating Point Exception (Core Dumped) with Dynamic Quantization on nn.Linear with Output Dimension 0,https://github.com/pytorch/pytorch/issues/165829,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-12-05T00:45:18Z +issue,169357,"prepareProfiler() skips collecting info of ""privateuseone""",https://github.com/pytorch/pytorch/issues/169357,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: privateuse1, oncall: profiler",categorize and decide next maintainer owner/action,2025-12-03T12:11:39Z +issue,162481,Incosistent tracking of device activities when calling profiler.step() in torch profiler,https://github.com/pytorch/pytorch/issues/162481,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-12-01T18:41:45Z +issue,151536,[inductor][cpu] pytorch_CycleGAN_and_pix2pix AMP/AMP_FP16 multiple thread performance regression in 2025-04-07 nightly release,https://github.com/pytorch/pytorch/issues/151536,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-11-26T07:19:42Z +issue,169069,Triton Pin update 3.6 related - All periodic linux-jammy-cuda12.4-py3.10-gcc11 are in broken state. Some linux-jammy-cuda12.8-py3.10-gcc11-debug are broken as well,https://github.com/pytorch/pytorch/issues/169069,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",categorize and decide next maintainer owner/action,2025-11-25T18:48:55Z +issue,169009,Dynamo export: Qwen3 Omni model running into data-dependent expression errors,https://github.com/pytorch/pytorch/issues/169009,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-11-25T00:39:27Z +issue,168077,scripted models wrongly cast integers to floats,https://github.com/pytorch/pytorch/issues/168077,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-11-18T21:39:56Z +issue,65607,torch.dtype is int in JIT instead of torch.dtype,https://github.com/pytorch/pytorch/issues/65607,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-11-18T11:42:23Z +issue,167730,Feedback about ExportDB,https://github.com/pytorch/pytorch/issues/167730,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-11-15T01:35:00Z +issue,147818,Bug in `torch.ao.nn.quantized.Sigmoid` Parameter Restoration after `state_dict ` Loading,https://github.com/pytorch/pytorch/issues/147818,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-11-14T18:46:17Z +issue,167199,[CPU] torch.compile consumes too much extra memory,https://github.com/pytorch/pytorch/issues/167199,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-11-14T05:34:44Z +issue,167412,How can I train in C++ using a Pytorch torchscript model,https://github.com/pytorch/pytorch/issues/167412,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-11-10T19:11:24Z +issue,126703,PyTorch's Execution Trace Observer Producing Invalid JSON,https://github.com/pytorch/pytorch/issues/126703,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-11-09T03:36:41Z +issue,166939,Reduce aten::empty Calls for Models with Constant Input Shapes in AOTInductor,https://github.com/pytorch/pytorch/issues/166939,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-11-08T00:24:54Z +issue,166311,Unable to export profiler trace,https://github.com/pytorch/pytorch/issues/166311,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-11-07T09:19:37Z +issue,47163,Grad strides do not match bucket view strides.,https://github.com/pytorch/pytorch/issues/47163,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: memory format, oncall: distributed",categorize and decide next maintainer owner/action,2025-11-06T15:17:58Z +issue,164145,Improvements to profiler for bitwise equivalence use case,https://github.com/pytorch/pytorch/issues/164145,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-10-26T03:18:33Z +issue,165395,[export] run_decompositions fails with index error,https://github.com/pytorch/pytorch/issues/165395,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-10-21T21:37:04Z +issue,165929,Numerical mismatch b/w eager & aot_eager_decomp_partition backend in SegFormer layer (2.9.0),https://github.com/pytorch/pytorch/issues/165929,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-10-21T14:27:18Z +issue,165900,Converting weights `.pt` content between `dict` and `RecursiveScriptModule`,https://github.com/pytorch/pytorch/issues/165900,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-10-20T14:27:26Z +issue,128114,Import Error: cannot import name 'XNNPACKQuantizer' from 'torch.ao.quantization.quantizer',https://github.com/pytorch/pytorch/issues/128114,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-10-20T11:35:30Z +issue,163791,torch.export fails on Phi-3.5 model,https://github.com/pytorch/pytorch/issues/163791,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-10-18T17:54:39Z +issue,165230,_int_mm is slow with row-major rhs matrix,https://github.com/pytorch/pytorch/issues/165230,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-10-17T04:28:23Z +issue,165400,aten::bitwise_xor,https://github.com/pytorch/pytorch/issues/165400,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-10-14T13:33:35Z +issue,156814,torchscript: RuntimeError Unsupported dtype complex,https://github.com/pytorch/pytorch/issues/156814,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: complex, oncall: jit",categorize and decide next maintainer owner/action,2025-10-14T07:40:12Z +issue,164083,Broaden support of unflatten to split graphs,https://github.com/pytorch/pytorch/issues/164083,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2025-10-07T15:50:53Z +issue,164534,DISABLED RecordDebugHandles.ScopedCallbacks (__main__.test_jit),https://github.com/pytorch/pytorch/issues/164534,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: unknown, oncall: jit",categorize and decide next maintainer owner/action,2025-10-03T14:31:07Z +issue,164208,CUDA Illegal Memory Access in Multi-threaded LibTorch Applications,https://github.com/pytorch/pytorch/issues/164208,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-09-30T18:31:23Z +issue,69178,DISABLED test_remote_timeout_to_here_in_jit (__main__.FaultyJitFaultyAgentRpcTest),https://github.com/pytorch/pytorch/issues/69178,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: rpc, oncall: distributed, oncall: jit",categorize and decide next maintainer owner/action,2025-09-29T23:19:15Z +issue,70865,Support hooks-based checkpointing API with DDP,https://github.com/pytorch/pytorch/issues/70865,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2025-09-26T23:46:38Z +issue,163945,Failed to run a backward op in TorchScript interpreter,https://github.com/pytorch/pytorch/issues/163945,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-09-26T11:32:33Z +issue,149737,`INTERNAL ASSERT FAILED` in `interpolate` and `torch.import_ir_module`,https://github.com/pytorch/pytorch/issues/149737,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: error checking, oncall: quantization",categorize and decide next maintainer owner/action,2025-09-21T09:47:56Z +issue,162948,Metadata object loses arbitrary attributes added dynamically,https://github.com/pytorch/pytorch/issues/162948,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed checkpointing,categorize and decide next maintainer owner/action,2025-09-18T01:13:04Z +issue,162365,[Windows][CPU] 3 UT of test/inductor/test_torchinductor.py failed with 0903 nightly whl on windows,https://github.com/pytorch/pytorch/issues/162365,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-09-11T15:30:21Z +issue,162451,Some Windows AOTI UTs regression,https://github.com/pytorch/pytorch/issues/162451,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: dynamo, module: windows, oncall: export",categorize and decide next maintainer owner/action,2025-09-11T00:12:36Z +issue,162135,[inductor][cpu]dla102 AMP default wrapper hang issue,https://github.com/pytorch/pytorch/issues/162135,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-09-10T22:50:40Z +issue,161815,Windows AMI release issues and improvements,https://github.com/pytorch/pytorch/issues/161815,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: releng",categorize and decide next maintainer owner/action,2025-09-09T20:26:36Z +issue,130407,[RFC][c10d] a new Pytorch API (split_group) to create a process group through ncclCommSplit,https://github.com/pytorch/pytorch/issues/130407,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2025-09-05T23:03:35Z +issue,161935,torch.jit.trace_module.register_parameter :cannot assign 'torch.cuda.FloatTensor' object to parameter 'new_param' (torch.nn.Parameter or None required),https://github.com/pytorch/pytorch/issues/161935,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-09-02T13:04:18Z +issue,161854,torch.ScriptObjectProperty.name.deleter occurs Segmentation fault (core dumped),https://github.com/pytorch/pytorch/issues/161854,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-09-02T06:08:31Z +issue,154094,"[inductor][cpu] aoti shared memory run on multiple cores, performance will drop",https://github.com/pytorch/pytorch/issues/154094,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: cpu inductor, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-09-01T03:00:33Z +issue,63476,Error loading Pytorch lite model in Android,https://github.com/pytorch/pytorch/issues/63476,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2025-08-29T14:27:38Z +issue,157728,"fill_ overflows on uint64 in range [2**63, 2**64) when profiler is engaged",https://github.com/pytorch/pytorch/issues/157728,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-08-28T16:11:29Z +issue,151969,Unrecognized in PyTorch traces w/ Cuda 12.6,https://github.com/pytorch/pytorch/issues/151969,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-08-18T19:05:28Z +issue,49460,Libtorch: Segmentation fault when running torch::jit::load,https://github.com/pytorch/pytorch/issues/49460,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-08-17T10:50:56Z +issue,160631,torch.quantize_per_tensor not implemented for SparseCPU tensors,https://github.com/pytorch/pytorch/issues/160631,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: sparse, oncall: quantization",categorize and decide next maintainer owner/action,2025-08-15T23:23:52Z +issue,160391,`torch.compile` backward pass fails with `AssertionError` in Inductor C++ codegen when model returns tuple outputs,https://github.com/pytorch/pytorch/issues/160391,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-08-15T18:28:19Z +issue,160392,bug in libtorch optimize_for_inference,https://github.com/pytorch/pytorch/issues/160392,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-08-13T20:33:33Z +issue,160005,[BE][download.pytorch.org] Index fix user navigation,https://github.com/pytorch/pytorch/issues/160005,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: releng,categorize and decide next maintainer owner/action,2025-08-12T20:28:33Z +issue,144770,[ARM] multiple test failures in TestQuantizedConv on Aarch64,https://github.com/pytorch/pytorch/issues/144770,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: tests, oncall: quantization",categorize and decide next maintainer owner/action,2025-08-12T09:43:42Z +issue,159316,Support for non-scalar valued Loss Tensor in `_export_forward_backward()`,https://github.com/pytorch/pytorch/issues/159316,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-08-09T00:57:34Z +issue,159505,Cannot pass through None for example_inputs in prepare_fx,https://github.com/pytorch/pytorch/issues/159505,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-08-07T17:25:26Z +issue,55282,Poor error message when trying to jit a function instead of a module (RuntimeError: Cannot insert a Tensor that requires grad as a constant.),https://github.com/pytorch/pytorch/issues/55282,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-08-07T14:51:52Z +issue,159169,[DTensor][FSDP2][DDP] benchmark dtensor cpu overhead for adam optimizer,https://github.com/pytorch/pytorch/issues/159169,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-08-07T03:39:02Z +issue,159256,can kineto profile only one thread with corresponding gpu tasks among multithreads,https://github.com/pytorch/pytorch/issues/159256,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-07-31T15:28:22Z +issue,159269,Feature Request: Allow torch::inductor::ModelPackageLoader to load from an in-memory buffer,https://github.com/pytorch/pytorch/issues/159269,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-07-29T14:54:04Z +issue,159265,Unpin dependency version when possible,https://github.com/pytorch/pytorch/issues/159265,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, oncall: package/deploy",categorize and decide next maintainer owner/action,2025-07-28T22:01:03Z +issue,158331,DISABLED test_conv2d_unary_cpu_cpp_wrapper (__main__.TestCppWrapper),https://github.com/pytorch/pytorch/issues/158331,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, oncall: cpu inductor",categorize and decide next maintainer owner/action,2025-07-15T06:50:49Z +issue,158227,Degraded JIT execution performance on 5090 Windows,https://github.com/pytorch/pytorch/issues/158227,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-07-14T18:08:15Z +issue,156915,Unable to compile with Vulkan support,https://github.com/pytorch/pytorch/issues/156915,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: build, module: vulkan, oncall: mobile",categorize and decide next maintainer owner/action,2025-07-14T07:21:35Z +issue,148902,Remove Direct Arm Compute Libray (ACL) Integration for Quantized Matmuls: `qlinear`/`qlinear_dynamic`,https://github.com/pytorch/pytorch/issues/148902,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, oncall: quantization",categorize and decide next maintainer owner/action,2025-07-09T17:03:58Z +issue,155862,Export Huggingface models with StaticCache,https://github.com/pytorch/pytorch/issues/155862,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-07-09T16:48:49Z +issue,156882,Export always give a value range with max length - 1,https://github.com/pytorch/pytorch/issues/156882,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-07-09T00:55:21Z +issue,157625,Is there some official method to extract the featuremap of each node in pt2 graph like the function torchvision.models.feature_extraction.create_feature_extractor(),https://github.com/pytorch/pytorch/issues/157625,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-07-08T17:33:56Z +issue,153429,[Export] Non-strict mode can't handle conditionals on tensor subclass types,https://github.com/pytorch/pytorch/issues/153429,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-07-07T17:41:39Z +issue,157603,Incorrect inference of the groups parameter type for channel_stuffle (int misclassified as Tensor),https://github.com/pytorch/pytorch/issues/157603,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-07-07T08:07:53Z +issue,156544,`TorchScript` does not allow accessing methods of nested tensors,https://github.com/pytorch/pytorch/issues/156544,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, oncall: jit",categorize and decide next maintainer owner/action,2025-06-30T10:56:11Z +issue,157103,Issue with super() resolution in TorchScript when inheriting from custom class and nn.Module,https://github.com/pytorch/pytorch/issues/157103,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-30T01:57:19Z +issue,156955,`torch.jit.load()` crashes with C++ `c10::Error` if file system is full (no graceful handling),https://github.com/pytorch/pytorch/issues/156955,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, oncall: jit",categorize and decide next maintainer owner/action,2025-06-26T14:49:29Z +issue,156558,Using collections. namedtuple in the forward method of the model resulted in compilation failure,https://github.com/pytorch/pytorch/issues/156558,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export, oncall: jit, oncall: pt2",categorize and decide next maintainer owner/action,2025-06-26T01:29:10Z +issue,156557,[JIT] TorchScript fails to compile model using random.choice() with confusing error message,https://github.com/pytorch/pytorch/issues/156557,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-22T19:22:45Z +issue,112844,torch.export does not support torchaudio.transforms.Spectrogram,https://github.com/pytorch/pytorch/issues/112844,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export",categorize and decide next maintainer owner/action,2025-06-20T23:40:06Z +issue,155550,"[torch.export] Cannot export TorchVision raft_small, raft_large",https://github.com/pytorch/pytorch/issues/155550,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-06-19T10:50:53Z +issue,155321,mishandling torch.package.PackageExporter raises Aborted when given a tensor instead of an importer,https://github.com/pytorch/pytorch/issues/155321,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2025-06-09T14:59:56Z +issue,137252,RuntimeError: Expression of type - cannot be used in a type expression: __torch__.transformers_modules.code-5p-110m-embedding.modeling_codet5p_embedding.___torch_mangle_1368.CodeT5pEmbeddingModel ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ <--- HERE,https://github.com/pytorch/pytorch/issues/137252,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-05T08:17:14Z +issue,111480,torch.jit.script persistently changes default from utf-8 to ascii,https://github.com/pytorch/pytorch/issues/111480,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-04T03:03:00Z +issue,155003,Internal Assertion Failure with Invalid Arguments in max_unpool2d under TorchScript,https://github.com/pytorch/pytorch/issues/155003,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-03T18:01:40Z +issue,155004,Internal Assertion Failure with Invalid Arguments in max_pool2d under TorchScript,https://github.com/pytorch/pytorch/issues/155004,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-03T18:01:30Z +issue,155007,Internal Assertion Failure with Invalid Arguments in max_pool1d under TorchScript,https://github.com/pytorch/pytorch/issues/155007,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-03T18:01:21Z +issue,155009,Internal Assertion Failure with Invalid Arguments in max_unpool1d under TorchScript,https://github.com/pytorch/pytorch/issues/155009,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-06-03T18:01:12Z +issue,154853,"Add node.meta[""stack_trace""] to make_fx",https://github.com/pytorch/pytorch/issues/154853,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-06-02T17:01:30Z +issue,130309,"【Pytorch mobile android】torch jit forward fail,Calling torch.geqrf on a CPU tensor requires compiling PyTorch with LAPACK. Please use PyTorch built with LAPACK support.",https://github.com/pytorch/pytorch/issues/130309,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-31T20:11:48Z +issue,154607,Internal Assertion Failure with torch.cuda.Stream in @script_method on CPU-only Systems,https://github.com/pytorch/pytorch/issues/154607,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-29T13:00:01Z +issue,154478,`__jit_ignored_attributes__` is not respected in `torch.jit.trace_module`,https://github.com/pytorch/pytorch/issues/154478,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-28T13:39:42Z +issue,154502,`typing.get_type_hints` fails on TorchScript model after loading with `torch.jit.load`,https://github.com/pytorch/pytorch/issues/154502,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-28T13:34:47Z +issue,154407,[JIT] torch.jit.script raises an exception with view(dtype),https://github.com/pytorch/pytorch/issues/154407,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-28T08:09:26Z +issue,154423,`Segmentation fault` in `torch.jit.ignore`,https://github.com/pytorch/pytorch/issues/154423,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-27T18:57:18Z +issue,154323,"when pass a scale generated from a registered_buff of a nn.Module to fake_quantize_per_channel_affine, onnx runntime will raise Exception： “For per axis quantization, scale must be 1D tensor with”",https://github.com/pytorch/pytorch/issues/154323,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-05-27T18:24:47Z +issue,154324,torch.jit.script gives false results for autograd if complex data types are involved,https://github.com/pytorch/pytorch/issues/154324,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: complex, oncall: jit",categorize and decide next maintainer owner/action,2025-05-25T13:05:16Z +issue,154101,"INTERNAL ASSERT FAILED at ""/pytorch/torch/csrc/jit/ir/ir.cpp"":1307",https://github.com/pytorch/pytorch/issues/154101,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-22T12:46:27Z +issue,153901,torch.profiler support for tf32,https://github.com/pytorch/pytorch/issues/153901,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-05-20T15:41:54Z +issue,153931,TorchScript Fails to Handle type() for Any Input Type (Tensor/Non-Tensor),https://github.com/pytorch/pytorch/issues/153931,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-20T14:16:16Z +issue,150526,Consider context when tuning kernels for max-autotune to more accurately reflect the performance of real workloads,https://github.com/pytorch/pytorch/issues/150526,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-05-20T05:50:02Z +issue,153722,JIT Script doesn't handle LLONG_MIN correctly,https://github.com/pytorch/pytorch/issues/153722,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-16T14:34:24Z +issue,152921,[inductor][cpu] pytorch_CycleGAN_and_pix2pix AMP multiple thread performance regression in 2025-04-27 nightly release,https://github.com/pytorch/pytorch/issues/152921,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-05-15T08:48:47Z +issue,153315,[JIT] Compilation-induced discrepancy in F.instance_norm when passing input as running stats,https://github.com/pytorch/pytorch/issues/153315,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), oncall: jit",categorize and decide next maintainer owner/action,2025-05-13T17:51:18Z +issue,153358,torch.dequantize result inconsistent on CPU and GPU,https://github.com/pytorch/pytorch/issues/153358,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-05-13T07:23:37Z +issue,152813,Mismatch in dynamic quantization performance for torchao and torch.quantization,https://github.com/pytorch/pytorch/issues/152813,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-05-12T05:48:32Z +issue,153224,[JIT] Inconsistent results between eager and script modes for instance_norm with custom running stats,https://github.com/pytorch/pytorch/issues/153224,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-05-10T11:29:31Z +issue,153179,export: No error raised on data-dependent assertion.,https://github.com/pytorch/pytorch/issues/153179,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: error checking, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-05-09T20:24:48Z +issue,152465,torch.export with dynamic shapes on Static Cache HF LLama model fails,https://github.com/pytorch/pytorch/issues/152465,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-05-05T23:57:45Z +issue,152617,Pytorch Profiler crashes while using it with Pytorch Lightning module,https://github.com/pytorch/pytorch/issues/152617,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-05-02T13:28:56Z +issue,152612,Enable AOTI for Metal inductor,https://github.com/pytorch/pytorch/issues/152612,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: mps, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-05-01T22:58:02Z +issue,149500,[RFC] : Dynamically Quantized 8-bit Matrix Multiplication support,https://github.com/pytorch/pytorch/issues/149500,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-05-01T19:47:48Z +issue,152283,Forward compatibility in torch.export,https://github.com/pytorch/pytorch/issues/152283,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-04-28T23:03:31Z +issue,150607,`torch.jit.script` does not respect `torch.set_default_dtype`,https://github.com/pytorch/pytorch/issues/150607,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-28T02:05:26Z +issue,151746,[AotInductor][Export][Triton] how to export custom triton kernels when use torch.export.export,https://github.com/pytorch/pytorch/issues/151746,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: user triton, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-04-25T23:11:04Z +issue,94974,Slow inference of torchscript model in C++,https://github.com/pytorch/pytorch/issues/94974,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-24T21:19:47Z +issue,148557,AOTI is OOM-ing when eager doesn't,https://github.com/pytorch/pytorch/issues/148557,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-04-22T00:26:55Z +issue,151468,"Bug At `d\\dependencies\\libtorch\\include\\ATen\\core\\jit_type_base.h"":289`",https://github.com/pytorch/pytorch/issues/151468,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-17T13:50:36Z +issue,151428,TorchScript Model Saved on x86 Returns NaNs When Loaded on s390x,https://github.com/pytorch/pytorch/issues/151428,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-16T14:55:07Z +issue,110332,`torch.jit.load()` might unresponsive in IBM s390x when loading some certain torchscript saved by x86 machine.,https://github.com/pytorch/pytorch/issues/110332,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: power, oncall: jit",categorize and decide next maintainer owner/action,2025-04-15T23:28:04Z +issue,151233,Code Injection via `torch.jit.annotations.parse_type_line()`,https://github.com/pytorch/pytorch/issues/151233,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-15T01:36:30Z +issue,151016,[Dynamo][Export] Untraceable issues when exporting the Stable Diffusion 3.5 model,https://github.com/pytorch/pytorch/issues/151016,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-04-14T18:43:55Z +issue,151165,Bug on running TorchScript on H100,https://github.com/pytorch/pytorch/issues/151165,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-13T23:01:42Z +issue,150577,[TorchScript] Enum scripting failures in python 3.11+,https://github.com/pytorch/pytorch/issues/150577,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-03T23:40:13Z +issue,92855,Profiler documentation doesn't mention some exports are mutually exclusive,https://github.com/pytorch/pytorch/issues/92855,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: profiler",categorize and decide next maintainer owner/action,2025-04-03T17:58:38Z +issue,146914,TestSelectAlgorithmCPU.test_int8_woq_mm fails on Aarch64,https://github.com/pytorch/pytorch/issues/146914,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: tests, oncall: cpu inductor, oncall: jit",categorize and decide next maintainer owner/action,2025-04-01T23:24:25Z +issue,150342,Torch trace doesn't respect @torch.jit.ignore on torch.nn.Module forward method,https://github.com/pytorch/pytorch/issues/150342,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-04-01T00:03:16Z +issue,137691,[inductor][cpu]float32 and AMP performance regression in 2024-09-30 nightly release,https://github.com/pytorch/pytorch/issues/137691,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-26T18:28:48Z +issue,149186,torch.fx.symbolic_trace failed on deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B,https://github.com/pytorch/pytorch/issues/149186,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-26T15:34:31Z +issue,134679,[CPU] jx_nest_base float32 both inductor and eager performance regression in 2024-08-26 nightly release,https://github.com/pytorch/pytorch/issues/134679,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-26T00:21:47Z +issue,149800,`Segmentation fault` in `torch.jit.jit_module_from_flatbuffer`,https://github.com/pytorch/pytorch/issues/149800,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-24T03:58:03Z +issue,149736,`INTERNAL ASSERT FAILED` in `torch.jit.script`,https://github.com/pytorch/pytorch/issues/149736,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-21T15:54:16Z +issue,149619,Unexpected return type from `torch.split` under `@torch.jit.script` decorator,https://github.com/pytorch/pytorch/issues/149619,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-20T19:29:44Z +issue,149501,Inductor produce significantly different inference results with the originl original model,https://github.com/pytorch/pytorch/issues/149501,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-20T08:53:28Z +issue,148946,export deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B failed,https://github.com/pytorch/pytorch/issues/148946,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-19T22:06:41Z +issue,149391,Cannot torch.jit.script nn.MultiheadAttention when bias is set to False,https://github.com/pytorch/pytorch/issues/149391,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-18T21:26:30Z +issue,82550,`torch.cat` can break `torch.jit.ScriptModule` when in inference mode,https://github.com/pytorch/pytorch/issues/82550,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-18T13:15:32Z +issue,149251,false INTERNAL ASSERT FAILED in `torch.jit.trace`,https://github.com/pytorch/pytorch/issues/149251,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-17T15:27:11Z +issue,107568,RuntimeError: Unsupported value kind: Tensor while torch.jit.script nn.Module,https://github.com/pytorch/pytorch/issues/107568,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-17T10:07:28Z +issue,149223,Inconsistent results in using torch.jit.script API from API documentation.,https://github.com/pytorch/pytorch/issues/149223,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-14T23:20:43Z +issue,136618,[inductor][cpu]pyhpc_isoneutral_mixing singel thread AMP cpp wrapper performance regression in 2024-09-22 nightly release,https://github.com/pytorch/pytorch/issues/136618,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-11T01:34:13Z +issue,148382,[Inductor-CPU] Debug util request: fine-grained mechanism to disable out-of-template epilogues,https://github.com/pytorch/pytorch/issues/148382,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-10T19:54:13Z +issue,148071,[inductor][cpu] maml performance regression in 2025-02-24 nightly release,https://github.com/pytorch/pytorch/issues/148071,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-10T19:54:13Z +issue,147919,[inductor][cpu]AOT inductor AMP static shape default wrapper occupied almost 3x disk than before,https://github.com/pytorch/pytorch/issues/147919,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-10T19:54:12Z +issue,81552,Support Swift Package Manager (SPM) for iOS,https://github.com/pytorch/pytorch/issues/81552,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2025-03-06T18:58:48Z +issue,148543,"When Ues torch::jit::Module in UE env, there are some structured exception caused by calling module.forward()",https://github.com/pytorch/pytorch/issues/148543,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-06T01:49:47Z +issue,148447,"Union type raise error when running python with argument ""-O"" for torch script.",https://github.com/pytorch/pytorch/issues/148447,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-06T01:14:12Z +issue,148171,Inference llama after Export PTQ,https://github.com/pytorch/pytorch/issues/148171,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-03-03T18:23:06Z +issue,36061,[JIT] support list of nn.Module in torchscript,https://github.com/pytorch/pytorch/issues/36061,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-03-03T08:27:33Z +issue,146824,Inductor-CPU might load (and store) fewer elements than the vector-width,https://github.com/pytorch/pytorch/issues/146824,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-03-01T23:26:12Z +issue,93134,Failed to Open libnvrtc-builtins.so.11.7,https://github.com/pytorch/pytorch/issues/93134,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: jit",categorize and decide next maintainer owner/action,2025-02-27T16:31:10Z +issue,147844,Python warnings are printed multiple times,https://github.com/pytorch/pytorch/issues/147844,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-25T19:37:54Z +issue,147739,torchvision export model error:: torchvision.models.detection.retinanet_resnet50_fpn_v2,https://github.com/pytorch/pytorch/issues/147739,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-25T15:28:44Z +issue,147722,Segmentation fault in `torch.ops.profiler._call_end_callbacks_on_jit_fut`,https://github.com/pytorch/pytorch/issues/147722,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-02-25T15:25:18Z +issue,147604,RuntimeError when running profiler in a loop,https://github.com/pytorch/pytorch/issues/147604,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-02-25T15:24:55Z +issue,146974,[inductor][cpu]detectron2_fcos_r_50_fpn multiple thread accuracy failure in 2025-02-10 nightly release,https://github.com/pytorch/pytorch/issues/146974,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-24T16:06:15Z +issue,146483,[ARM] Unit test failure on Neoverse-N1 - CPUReproTests.test_lstm_packed,https://github.com/pytorch/pytorch/issues/146483,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: tests, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-24T16:05:51Z +issue,147444,NaN values originating from the normalize_weight_jit function during the backward pass,https://github.com/pytorch/pytorch/issues/147444,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-24T15:56:27Z +issue,145394,[EXPORT AOTI] `aoti_compile_and_package` custom_ops dependecies,https://github.com/pytorch/pytorch/issues/145394,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-20T20:29:47Z +issue,146127,When I try to run rt-detr model in C++ libtorch i face the given error,https://github.com/pytorch/pytorch/issues/146127,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: jit",categorize and decide next maintainer owner/action,2025-02-19T09:17:41Z +issue,147187,[torch.export] How to export a model with kv cache,https://github.com/pytorch/pytorch/issues/147187,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-18T19:20:39Z +issue,144331,[Export] UserWarning: Attempted to insert a get_attr Node with no underlying reference in the owning GraphModule,https://github.com/pytorch/pytorch/issues/144331,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-15T12:53:16Z +issue,146315,[export] dynamic_shapes with `Dim` fails when `DYNAMIC` succeeds,https://github.com/pytorch/pytorch/issues/146315,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-15T03:01:55Z +issue,75960,"jit/_trace.py"", line 71, in _unique_state_dict filtered_dict[k] = v.detach() AttributeError: 'torch.dtype' object has no attribute 'detach'",https://github.com/pytorch/pytorch/issues/75960,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-14T09:36:04Z +issue,147112,Add quantized version of BatchNorm1d module,https://github.com/pytorch/pytorch/issues/147112,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-02-13T18:20:01Z +issue,147055,INTERNAL ASSERT FAILED or SEGFAULT when JITting a function that can return different types,https://github.com/pytorch/pytorch/issues/147055,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-13T07:01:40Z +issue,146915,Unit test failure on Aarch64 - TestSelectAlgorithmCPU.test_linear_with_in_out_buffer,https://github.com/pytorch/pytorch/issues/146915,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: tests, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-11T18:44:19Z +issue,146698,Possible memory leak when running test_model_exports_to_core_aten.py test,https://github.com/pytorch/pytorch/issues/146698,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-11T10:40:40Z +issue,146773,"INTERNAL ASSERT FAILED at ""/pytorch/aten/src/ATen/native/quantized/cpu/qsigmoid.cpp"":65, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/146773,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: quantization",categorize and decide next maintainer owner/action,2025-02-10T21:46:01Z +issue,143289,RFC: Dynamically Quantized 4 bit matmul API and usage,https://github.com/pytorch/pytorch/issues/143289,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-02-10T15:40:10Z +issue,146760,"[torch.jit] INTERNAL ASSERT FAILED at ""../aten/src/ATen/core/ivalue_inl.h"":1967, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/146760,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-09T20:30:55Z +issue,146758,"INTERNAL ASSERT FAILED at ""/pytorch/torch/csrc/autograd/functions/utils.h"":74, please report a bug to PyTorch",https://github.com/pytorch/pytorch/issues/146758,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-09T20:30:46Z +issue,146759,"INTERNAL ASSERT FAILED at ""../torch/csrc/jit/ir/alias_analysis.cpp"":617, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/146759,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-09T20:30:36Z +issue,146776,"INTERNAL ASSERT FAILED at ""/pytorch/torch/csrc/jit/testing/file_check.cpp"":607, please report a bug to PyTorch",https://github.com/pytorch/pytorch/issues/146776,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: testing, oncall: jit",categorize and decide next maintainer owner/action,2025-02-09T20:30:24Z +issue,146778,"[torch.jit] INTERNAL ASSERT FAILED at ""/pytorch/torch/csrc/jit/mobile/register_ops_common_utils.cpp"":34, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/146778,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-09T20:29:58Z +issue,146775,[torch.jit] Crash would be raised when using torch.jit.script,https://github.com/pytorch/pytorch/issues/146775,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-09T20:12:40Z +issue,146757,"[torch.jit.script] INTERNAL ASSERT FAILED at ""./torch/csrc/jit/ir/ir.h"":505, please report a bug to PyTorch",https://github.com/pytorch/pytorch/issues/146757,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-08T14:59:57Z +issue,114112,[JIT] - torch.script - 'Optional[Tensor]' object has no attribute or method 'size',https://github.com/pytorch/pytorch/issues/114112,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-02-07T10:50:58Z +issue,146576,Torch Dynamo Export Failed on RetinaNet from Torchvison,https://github.com/pytorch/pytorch/issues/146576,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-06T17:28:58Z +issue,146368,DeepSeek: block quantization,https://github.com/pytorch/pytorch/issues/146368,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-02-05T08:34:55Z +issue,145456,torch._dynamo.exc.Unsupported: Graph break due to unsupported builtin torch._C._dynamo.eval_frame.set_eval_frame.,https://github.com/pytorch/pytorch/issues/145456,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-05T01:20:34Z +issue,145711,[Inductor-CPU] Templated codegened kernel name may not correctly reflect epilogue fusions,https://github.com/pytorch/pytorch/issues/145711,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-04T22:33:11Z +issue,145709,[Inductor-CPU] Templated codegened kernel names are ambiguous in PyTorch profiler results,https://github.com/pytorch/pytorch/issues/145709,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-02-04T22:33:10Z +issue,145743,TorchScript run_method fails from 2.5.0 onward on Ubuntu,https://github.com/pytorch/pytorch/issues/145743,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: deadlock, oncall: jit",categorize and decide next maintainer owner/action,2025-02-03T10:31:47Z +issue,145995,torch.jit.trace does not move tensors to GPU.,https://github.com/pytorch/pytorch/issues/145995,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-01-30T01:12:10Z +issue,129889,custom gradient for int8,https://github.com/pytorch/pytorch/issues/129889,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2025-01-22T07:55:12Z +issue,68559,LibTorch -> TorchScript -> PyTorch (Python) fails with `AttributeError: 'RecursiveScriptModule' object has no attribute 'forward'`,https://github.com/pytorch/pytorch/issues/68559,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-01-20T22:21:17Z +issue,144778,`torch.profiler.record_function` doesn't register kernels from `backward` function,https://github.com/pytorch/pytorch/issues/144778,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-01-16T03:56:16Z +issue,144613,The label marked by torch.profiler.profile.record_function() appears twice in the output,https://github.com/pytorch/pytorch/issues/144613,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2025-01-13T17:27:47Z +issue,127571,[export] Cannot mutate tensors with frozen storage,https://github.com/pytorch/pytorch/issues/127571,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2025-01-08T22:04:05Z +issue,112372,[Tracer] RuntimeError: _Map_base::at when tracing using autograd.Function,https://github.com/pytorch/pytorch/issues/112372,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-01-08T02:02:58Z +issue,144148,"Unable to parse and visualize ""trace view"" with tensorboard",https://github.com/pytorch/pytorch/issues/144148,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: profiler",categorize and decide next maintainer owner/action,2025-01-06T15:40:24Z +issue,76774,Support Positional-only Arguments in JIT,https://github.com/pytorch/pytorch/issues/76774,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2025-01-03T20:06:13Z +issue,144039,[Mac/M1] torch.compile() -- expm1 returns an inaccurate result compared to the interpreted version,https://github.com/pytorch/pytorch/issues/144039,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2025-01-03T04:20:48Z +issue,141854,RuntimeError: quantized::conv2d_relu (ONEDNN): data type of input should be QUint8.,https://github.com/pytorch/pytorch/issues/141854,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: quantization",categorize and decide next maintainer owner/action,2024-12-30T21:53:10Z +issue,143886,[RFC] Identifying dynamic int8 symmetric vs asymmetric quantization of activation/input in Inductor-CPU,https://github.com/pytorch/pytorch/issues/143886,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-12-27T17:21:50Z +issue,143854,Can't script a tensorrt model,https://github.com/pytorch/pytorch/issues/143854,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-12-27T02:59:11Z +issue,135689,[inductor][cpu]GPTNeoForSequenceClassification AMP single/multiple thread static/dynamic shape default/cpp accuracy failure,https://github.com/pytorch/pytorch/issues/135689,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-12-27T02:35:03Z +issue,143831,"After pth is converted into ptl, the prediction result is very different from pth",https://github.com/pytorch/pytorch/issues/143831,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-12-26T15:42:23Z +issue,143857,`@torch.jit.script` causes `pytest-cov` to miss function body,https://github.com/pytorch/pytorch/issues/143857,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-12-26T14:29:55Z +issue,143565,Failed to load TorchScript of SSDLite in android,https://github.com/pytorch/pytorch/issues/143565,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-12-20T05:15:26Z +issue,132009,2:4 semi structured sparsity matmul not supported on hopper,https://github.com/pytorch/pytorch/issues/132009,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: quantization",categorize and decide next maintainer owner/action,2024-12-19T09:48:39Z +issue,143379,Running decompositions over torch.export program adds one more output to the graph result,https://github.com/pytorch/pytorch/issues/143379,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2024-12-17T21:38:08Z +issue,140689,[inductor][cpu]opacus_cifar10 and functorch_dp_cifar10 AMP performance regression in 2024-11-11 nightly release,https://github.com/pytorch/pytorch/issues/140689,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-12-17T01:19:15Z +issue,143258,Proper support for optionals in TorchScript,https://github.com/pytorch/pytorch/issues/143258,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-12-16T16:19:16Z +issue,81912,Unknown builtin op: aten::broadcast_shapes,https://github.com/pytorch/pytorch/issues/81912,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-12-16T08:04:07Z +issue,127521,Modeling ViT does not support quantized models,https://github.com/pytorch/pytorch/issues/127521,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-12-08T18:23:02Z +issue,141893,Forward from pytorch/executorch: export error with Macbook pro M4,https://github.com/pytorch/pytorch/issues/141893,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-12-07T01:13:45Z +issue,124190,Re-enable skipped 3.12 dynamo-wrapped profiler tests,https://github.com/pytorch/pytorch/issues/124190,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: python version, oncall: profiler",categorize and decide next maintainer owner/action,2024-12-06T21:30:51Z +issue,142170,[3.13] Fix failing torch.package dynamo-wrapped test,https://github.com/pytorch/pytorch/issues/142170,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: python frontend, oncall: package/deploy",categorize and decide next maintainer owner/action,2024-12-05T22:00:33Z +issue,141996,[export] Move pytree serialization into export serializer,https://github.com/pytorch/pytorch/issues/141996,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2024-12-03T21:19:51Z +issue,141855,Can not convert Llama-3.2-11B-Vision-Instruct by torch.jit.script / torch.jit.trace,https://github.com/pytorch/pytorch/issues/141855,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-12-02T16:21:36Z +issue,115511,[Export] aten::_transformer_encoder_layer_fwd is not core aten and has no decomposition.,https://github.com/pytorch/pytorch/issues/115511,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nn, oncall: export",categorize and decide next maintainer owner/action,2024-11-27T19:55:38Z +issue,140805,[AOTI] Test case `test_aot_inductor.py::test_multi_device_cpu` is Running GPU AOT Inductor in CPU container runner.,https://github.com/pytorch/pytorch/issues/140805,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2024-11-25T07:12:45Z +issue,92041,jit testing fails on 3.11 debug build,https://github.com/pytorch/pytorch/issues/92041,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-23T20:23:55Z +issue,141116,How to fuse batchnorm to conv2d in the graph exported by torch.export,https://github.com/pytorch/pytorch/issues/141116,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2024-11-20T19:06:46Z +issue,111142,'torch._C.Node' object has no attribute 'cs',https://github.com/pytorch/pytorch/issues/111142,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-20T15:02:09Z +issue,137202,"Met error ""we don't have an op for aten::full but it isn't a special case."" when running microsoft/Florence-2-base",https://github.com/pytorch/pytorch/issues/137202,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-20T02:24:56Z +issue,81855,Enable jit error when using FSDP,https://github.com/pytorch/pytorch/issues/81855,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-19T21:22:44Z +issue,121581,The results of PyTorch and TorchScript models are inconsistent.,https://github.com/pytorch/pytorch/issues/121581,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-19T14:27:21Z +issue,140468,[TorchScript] Failure if you script a wrapper module and then an interface-implementing submodule.,https://github.com/pytorch/pytorch/issues/140468,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-18T18:06:54Z +issue,132240,torch.nn.functional.interpolate completely broken with torch.jit.script and torch.fx,https://github.com/pytorch/pytorch/issues/132240,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-16T13:03:17Z +issue,140684,Segmentation fault (core dumped) in `torch.autograd.profiler.profile`,https://github.com/pytorch/pytorch/issues/140684,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-11-15T17:28:10Z +issue,140702,`Tensor.prod` gives a jit warning,https://github.com/pytorch/pytorch/issues/140702,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-14T11:35:18Z +issue,140539,torch.jit.trace doesn't trace all operations completely (no data-dependent control flow),https://github.com/pytorch/pytorch/issues/140539,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-14T03:07:29Z +issue,68893,NCCL Network is unreachable / Connection refused when initializing DDP,https://github.com/pytorch/pytorch/issues/68893,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",categorize and decide next maintainer owner/action,2024-11-13T09:18:50Z +issue,76026,Jit torchscript for prediction is missing 'forward' when using forward hooks,https://github.com/pytorch/pytorch/issues/76026,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-11T13:32:33Z +issue,135569,[inductor][cpu]AOT inductor performance regression and accuracy failure in 2024-09-04 nightly release,https://github.com/pytorch/pytorch/issues/135569,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-11-11T07:53:24Z +issue,66046,PicklingError when saving a ddp module with torch.save(),https://github.com/pytorch/pytorch/issues/66046,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2024-11-09T13:29:58Z +issue,140101,Inconsistent gradients in torch script function,https://github.com/pytorch/pytorch/issues/140101,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-08T17:11:11Z +issue,120166,QAT using pytorch-quantization cause accuracy loss after exporting to onnx,https://github.com/pytorch/pytorch/issues/120166,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: quantization",categorize and decide next maintainer owner/action,2024-11-08T10:17:10Z +issue,139470,serialization of PT2E model impacts torch.fx.passes.utils.source_matcher_utils.get_source_partitions,https://github.com/pytorch/pytorch/issues/139470,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-11-05T19:29:38Z +issue,45398,build_android.sh,https://github.com/pytorch/pytorch/issues/45398,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-11-05T09:20:03Z +issue,96726,Memory not release after jit.trace/freeze,https://github.com/pytorch/pytorch/issues/96726,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-05T08:59:43Z +issue,139556,[JIT] ModuleList getitem error when saving and loading with torchscript,https://github.com/pytorch/pytorch/issues/139556,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-11-04T14:02:30Z +issue,139464,Maybe there is a precision issue with torch.quantize_per_channel?,https://github.com/pytorch/pytorch/issues/139464,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-11-01T19:49:27Z +issue,139297,`torch.package` warning -- `TypedStorage` is deprecated,https://github.com/pytorch/pytorch/issues/139297,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2024-10-31T21:44:35Z +issue,139063,Quantization API examples in the docs are outdated,https://github.com/pytorch/pytorch/issues/139063,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: quantization",categorize and decide next maintainer owner/action,2024-10-28T21:20:10Z +issue,138973,"When using torch.jit.trace for inference, the element of the shape attribute of tensor and the return value of numel() are tensor",https://github.com/pytorch/pytorch/issues/138973,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-10-28T14:27:24Z +issue,70544,Tensor loses `bool` method during scripting,https://github.com/pytorch/pytorch/issues/70544,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-10-27T10:28:18Z +issue,138820,RuntimeError: HalfTensor is not supported,https://github.com/pytorch/pytorch/issues/138820,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-10-25T15:07:34Z +issue,138341,`torch.package` bug,https://github.com/pytorch/pytorch/issues/138341,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2024-10-18T20:12:17Z +issue,138221,auto_functionalized_v2 doesn't support export,https://github.com/pytorch/pytorch/issues/138221,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2024-10-18T19:58:11Z +issue,138193,[profiler] are CUDA events mutually exclusive with _ExperimentalConfig.profiler_metrics?,https://github.com/pytorch/pytorch/issues/138193,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-10-18T02:22:43Z +issue,70450,torch.jit.optimized_execution is not mentioned anywhere in the docs,https://github.com/pytorch/pytorch/issues/70450,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-10-15T08:39:52Z +issue,137561,Segmentation fault would be triggered when using torch.jit.trace and torch.jit._fork,https://github.com/pytorch/pytorch/issues/137561,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-10-09T13:30:42Z +issue,91000,node.stack_trace does not handle escaping correctly,https://github.com/pytorch/pytorch/issues/91000,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: fx,categorize and decide next maintainer owner/action,2024-10-04T15:29:05Z +issue,131452,"JIT load traced model with ""Expression of type"" error",https://github.com/pytorch/pytorch/issues/131452,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-10-03T05:05:29Z +issue,69250,TorchScript incomptaible with non-persistent buffers,https://github.com/pytorch/pytorch/issues/69250,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-10-01T14:27:11Z +issue,136829,false INTERNAL ASSERT FAILED in `torch.jit.set_fusion_strategy`,https://github.com/pytorch/pytorch/issues/136829,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-27T14:04:29Z +issue,136828,false INTERNAL ASSERT FAILED in `torch.empty`/`torch.ones`,https://github.com/pytorch/pytorch/issues/136828,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-09-27T14:04:19Z +issue,136654,[TorchScript] typing_extensions.deprecated doesn't work,https://github.com/pytorch/pytorch/issues/136654,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-25T18:27:20Z +issue,134971,"[torch.jit] INTERNAL ASSERT FAILED at ""../torch/csrc/jit/passes/symbolic_shape_analysis.cpp"":741, please report a bug to PyTorch. input_shapes size: 1 n inputs size: 2",https://github.com/pytorch/pytorch/issues/134971,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-23T23:01:39Z +issue,135690,[inductor][cpu] hf_BigBird float32 multiple threads performance regression in 2024-09-08 nightly release,https://github.com/pytorch/pytorch/issues/135690,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-09-23T22:59:06Z +issue,69078,RuntimeError: Global alloc not supported yet in TorchScript,https://github.com/pytorch/pytorch/issues/69078,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-20T17:56:18Z +issue,134924,`TorchScriptTesting._PickleTester` caused a segmentation fault with empty input lists in a jit traced function,https://github.com/pytorch/pytorch/issues/134924,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-20T16:16:39Z +issue,134923,A crash would be raised when using `._StackString` with empty lists,https://github.com/pytorch/pytorch/issues/134923,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-20T16:16:31Z +issue,134922,[torch.jit.script] An INTERNAL ASSERT error will be raised when permuting the inputs of a jit scripted graph with mismatched indices.,https://github.com/pytorch/pytorch/issues/134922,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-20T16:16:13Z +issue,134903,[torch.jit] Crash would be raised when using TorchScript modules,https://github.com/pytorch/pytorch/issues/134903,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-20T16:15:47Z +issue,134898,Segmentation fault would be triggered when using `torch.jit.script` and `torch._C._jit_shape_compute_graph_for_node`.,https://github.com/pytorch/pytorch/issues/134898,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-20T16:15:34Z +issue,134937,"[torch.jit.script] INTERNAL ASSERT FAILED at ""../aten/src/ATen/core/jit_type_base.h"":556, please report a bug to PyTorch",https://github.com/pytorch/pytorch/issues/134937,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-20T16:10:16Z +issue,130229,"RuntimeError: 0 INTERNAL ASSERT FAILED at ""../torch/csrc/jit/ir/alias_analysis.cpp"":615, please report a bug to PyTorch. We don't have an op for aten::full but it isn't a special case. Argument types: int[], bool, int, NoneType, Device, bool,",https://github.com/pytorch/pytorch/issues/130229,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-18T22:30:56Z +issue,136256,TorchScript raises an OSError when using frozen dataclass,https://github.com/pytorch/pytorch/issues/136256,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-18T16:48:23Z +issue,68866,wait() does not block the default stream for NCCL's asynchronous P2P operations,https://github.com/pytorch/pytorch/issues/68866,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",categorize and decide next maintainer owner/action,2024-09-18T04:25:18Z +issue,58298,"[JIT] Can't handle creating new objects in compiled methods: ""Class does not have an __init__ function defined""",https://github.com/pytorch/pytorch/issues/58298,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-17T02:05:57Z +issue,136094,Quantized model is way slower than regular model.,https://github.com/pytorch/pytorch/issues/136094,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-09-16T15:52:35Z +issue,136127,A division-by-zero operation in a scripted function can cause the core dumped,https://github.com/pytorch/pytorch/issues/136127,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-16T15:46:57Z +issue,122813,cpu performance for int4mm kernels,https://github.com/pytorch/pytorch/issues/122813,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-09-15T16:34:49Z +issue,133444,Collect tensor shapes only when using record_shapes in profiler,https://github.com/pytorch/pytorch/issues/133444,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-09-13T16:28:54Z +issue,135580,Inconsistent results between libtorch and pytorch for model with shared weights,https://github.com/pytorch/pytorch/issues/135580,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-11T19:26:24Z +issue,62474,Distributed: RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [2048]] is at version 4; expected version 3 instead,https://github.com/pytorch/pytorch/issues/62474,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2024-09-11T08:49:51Z +issue,134220,[Operator Benchmark] [CPU]OP: weight_norm_sparsifier_step functionality failed with parts of long config shapes,https://github.com/pytorch/pytorch/issues/134220,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-09-10T03:21:28Z +issue,93943,"`torch.jit.trace` memory usage increase although forward is constant, and gets much slower than forward with model depth increase",https://github.com/pytorch/pytorch/issues/93943,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-09T23:14:40Z +issue,82099,UserWarning: operator() sees varying value in profiling,https://github.com/pytorch/pytorch/issues/82099,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-09T23:14:39Z +issue,134303,[FakeTensor] Error: SymIntArrayRef expected to contain only concrete integers in UpSample2D block,https://github.com/pytorch/pytorch/issues/134303,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: export",categorize and decide next maintainer owner/action,2024-09-04T14:34:44Z +issue,134918,"DYNAMIC_TRT_MODEL_CONVERSION 0 INTERNAL ASSERT FAILED or, comfyui made me come here",https://github.com/pytorch/pytorch/issues/134918,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-09-03T19:15:39Z +issue,135003,"[BUG]: torch.profiler is blocked in ""_disable_profiler()"" when profiling cuda graph on single GPU",https://github.com/pytorch/pytorch/issues/135003,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-09-03T19:06:46Z +issue,60252,UserWarning: Failed to initialize NumPy: No module named 'numpy',https://github.com/pytorch/pytorch/issues/60252,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2024-09-03T15:05:55Z +issue,134849,torch::jit::load invalid_path error,https://github.com/pytorch/pytorch/issues/134849,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-30T15:18:54Z +issue,40870,[JIT] cannot statically infer the expected size of a list in this context for positional arguments to Tensor.view(),https://github.com/pytorch/pytorch/issues/40870,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-30T03:45:52Z +issue,134562,"ScriptModel Save ERROR, when I convert part of the model to TRT",https://github.com/pytorch/pytorch/issues/134562,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-27T14:20:53Z +issue,134559,[JIT][foreach_add] Arguments for call are not valid,https://github.com/pytorch/pytorch/issues/134559,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-27T14:20:46Z +issue,121191,Export-based GraphModules fail with deepcopy,https://github.com/pytorch/pytorch/issues/121191,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-08-27T11:18:28Z +issue,133552,fx qat symmetric config zero_point not zero,https://github.com/pytorch/pytorch/issues/133552,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-08-23T03:02:11Z +issue,133255,torch.export ExportedProgram from GPU cannot be loaded on a CPU only machine,https://github.com/pytorch/pytorch/issues/133255,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-08-23T00:09:46Z +issue,123168,TorchScript: Union cannot be used as a tuple,https://github.com/pytorch/pytorch/issues/123168,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-21T15:24:45Z +issue,133678,[torch.jit.trace] error encountered when tracing model with weight clamping during forward,https://github.com/pytorch/pytorch/issues/133678,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-21T07:13:42Z +issue,133795,"RuntimeError: createStatus == pytorch_qnnp_status_success INTERNAL ASSERT FAILED at ""../aten/src/ATen/native/quantized/cpu/BinaryOps.cpp"":204, please report a bug to PyTorch. failed to create QNNPACK Add operator",https://github.com/pytorch/pytorch/issues/133795,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: quantization",categorize and decide next maintainer owner/action,2024-08-20T14:30:36Z +issue,133847,"[Bug Report] when using % in torch.jit.script function, the output conforms to the form of C++ % operation, rather than Python % operation.",https://github.com/pytorch/pytorch/issues/133847,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-19T15:10:06Z +issue,133840,Pytorch Mobile - Different output on Android and PC with the same model,https://github.com/pytorch/pytorch/issues/133840,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2024-08-19T15:01:15Z +issue,132930,Treating SymNode compute more fungibly for the purpose of export flattener (or not?),https://github.com/pytorch/pytorch/issues/132930,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export",categorize and decide next maintainer owner/action,2024-08-16T22:05:13Z +issue,133675,"[torch.jit] RuntimeError: false INTERNAL ASSERT FAILED at ""../torch/csrc/jit/ir/node_hashing.cpp"":148, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/133675,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-16T20:52:31Z +issue,133557,torch quantize error,https://github.com/pytorch/pytorch/issues/133557,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-08-16T20:43:48Z +issue,133537,`torch.jit.script` fails on a simple function,https://github.com/pytorch/pytorch/issues/133537,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-16T13:26:14Z +issue,122934,Cannot export DETR model with torch.export,https://github.com/pytorch/pytorch/issues/122934,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-08-15T18:18:01Z +issue,132738,Profiler getProfilerConfig interface can not support Ascend npu platform,https://github.com/pytorch/pytorch/issues/132738,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-08-14T18:25:42Z +issue,116662,test/test_vulkan.py fails with vulkan_prepack::conv2d_clamp_prepack not found,https://github.com/pytorch/pytorch/issues/116662,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: vulkan, oncall: mobile",categorize and decide next maintainer owner/action,2024-08-14T15:21:29Z +issue,51328,Support torch::jit::script::Module converting to torch::nn::Module.,https://github.com/pytorch/pytorch/issues/51328,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-13T11:41:12Z +issue,132559,How to fix tensor.numpy() not supported for torch.export with strict=False,https://github.com/pytorch/pytorch/issues/132559,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: numpy, oncall: export",categorize and decide next maintainer owner/action,2024-08-06T18:42:05Z +issue,132459,"[export] Implement a common ""convert dtype to int"" function",https://github.com/pytorch/pytorch/issues/132459,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-08-06T16:39:55Z +issue,132385,crnn model from pth to torchscript problem,https://github.com/pytorch/pytorch/issues/132385,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-08-01T17:37:06Z +issue,132315,Support Enum for dictionaries in TorchScript,https://github.com/pytorch/pytorch/issues/132315,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-07-31T19:33:35Z +issue,128292,put profiling results from different ranks into one webpage,https://github.com/pytorch/pytorch/issues/128292,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-07-30T17:19:28Z +issue,132029,torch.QUInt4x2Storage: RuntimeError cuda_dispatch_ptr INTERNAL ASSERT FAILED on torch 2.5.0.dev20240708+cu121,https://github.com/pytorch/pytorch/issues/132029,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-07-29T16:26:38Z +issue,88617,"The libtorch tests Simplify.{SimplifySymbolicMinMax,SimplifyNestedMax,SimplifyNestedMin} fail on Apple Silicon",https://github.com/pytorch/pytorch/issues/88617,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: macos, oncall: jit",categorize and decide next maintainer owner/action,2024-07-25T11:28:36Z +issue,18476,[CPP] Allow binding config structs into the Python front end,https://github.com/pytorch/pytorch/issues/18476,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: jit",categorize and decide next maintainer owner/action,2024-07-24T20:50:02Z +issue,130726,TorchDispatchMode fails on jit.trace,https://github.com/pytorch/pytorch/issues/130726,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-07-23T23:00:32Z +issue,131022,Code bug in call_prepare_scriptable_func_impl function.,https://github.com/pytorch/pytorch/issues/131022,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-07-18T19:55:06Z +issue,131055,PT2E XNNPACKQuantizer has unused operator_type_config and set_operator_type,https://github.com/pytorch/pytorch/issues/131055,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-07-18T19:46:02Z +issue,130820,[Export] Mutating module attribute error during export on YOLOX,https://github.com/pytorch/pytorch/issues/130820,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-07-17T06:48:47Z +issue,130735,"Problem with combination of jacref, vjp, tanh in torchscript",https://github.com/pytorch/pytorch/issues/130735,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-07-15T23:46:45Z +issue,67834,Model tracing not working,https://github.com/pytorch/pytorch/issues/67834,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-07-15T09:24:30Z +issue,130555,torch.export.export re-export preserving metadata for output node,https://github.com/pytorch/pytorch/issues/130555,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-07-11T17:54:20Z +issue,104405,Detailed error: Tensor-likes are not close! When use torch.jit.trace,https://github.com/pytorch/pytorch/issues/104405,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-07-07T01:18:44Z +issue,76486,[Tracer] RuntimeError: _Map_base::at when tracing fake quantization,https://github.com/pytorch/pytorch/issues/76486,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-07-01T06:03:35Z +issue,48387,"Error occurred when trying to call ""torch::jit::load""method for loading "".pt"" file on IOS",https://github.com/pytorch/pytorch/issues/48387,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2024-06-28T23:44:50Z +issue,129633,pytorch_android-2.1.0.aar/jni/arm64-v8a/libpytorch_jni.so: invalid local symbol '__bss_end__' in global part of symbol table,https://github.com/pytorch/pytorch/issues/129633,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-06-27T14:48:24Z +issue,129347,Error when loading a module file by calling torch::jit::load(..),https://github.com/pytorch/pytorch/issues/129347,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-24T15:47:21Z +issue,125272,Profiler Kineto Config: Warning: Failed to recover relationship between all profiler and kineto events: 51361 vs. 0 reassociated. (function reassociate),https://github.com/pytorch/pytorch/issues/125272,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-06-24T15:22:35Z +issue,129282,Support for torch.Generator with JIT,https://github.com/pytorch/pytorch/issues/129282,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-23T16:50:57Z +issue,89064,UserWarning: The TorchScript type system doesn't support instance-level annotations on empty non-base types in `__init__`.,https://github.com/pytorch/pytorch/issues/89064,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-22T15:56:43Z +issue,123499,Support Exceptions in Pytorch Export,https://github.com/pytorch/pytorch/issues/123499,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export",categorize and decide next maintainer owner/action,2024-06-17T23:48:10Z +issue,128491,Improve type inference in torchscript,https://github.com/pytorch/pytorch/issues/128491,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-13T15:02:40Z +issue,128585,TorchScript JIT UnicodeDecodeError: ‘utf-8’ codec can’t decode byte 0xe0 in position 0: unexpected end of data,https://github.com/pytorch/pytorch/issues/128585,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-13T15:02:01Z +issue,128496,torch.jit.script not work for ParameterDict.items,https://github.com/pytorch/pytorch/issues/128496,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-13T15:02:00Z +issue,128385,[export] Llama2 export with dynamic shapes fails using Torch 2.4 nightly,https://github.com/pytorch/pytorch/issues/128385,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-06-12T18:45:55Z +issue,42207,torchscript does not support unicode strings,https://github.com/pytorch/pytorch/issues/42207,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-12T11:26:30Z +issue,126266,Quantization occurs with RuntimeError: `zero_point` must be between `quant_min` and `quant_max`.,https://github.com/pytorch/pytorch/issues/126266,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-06-12T06:47:18Z +issue,127283,[BUG][JIT] `torch.jit.script` is not compatible with `DeprecationWarning` and `FutureWarning`,https://github.com/pytorch/pytorch/issues/127283,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-12T01:14:55Z +issue,116795,multiple output support of jit mobile model?,https://github.com/pytorch/pytorch/issues/116795,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-06-10T19:33:04Z +issue,118069,Quantising a multimodal large language PyTorch model to run on Android,https://github.com/pytorch/pytorch/issues/118069,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-06-10T19:31:00Z +issue,120976,RuntimeError: xnnpack::convolution not available！,https://github.com/pytorch/pytorch/issues/120976,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-06-10T19:28:45Z +issue,128261,JIT script can cause unneeded `requires_grad` inside `no_grad` blocks,https://github.com/pytorch/pytorch/issues/128261,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-10T16:29:38Z +issue,128125,optional memory leak with torch.jit.script on models with fft,https://github.com/pytorch/pytorch/issues/128125,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-10T15:41:39Z +issue,126127,torch.export.export() fails with dynamic shapes when more than one shape is dynamic,https://github.com/pytorch/pytorch/issues/126127,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export",categorize and decide next maintainer owner/action,2024-06-05T21:12:01Z +issue,67983,torch.jit.export does not working,https://github.com/pytorch/pytorch/issues/67983,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-05T20:20:54Z +issue,127357,torch.export.export() throws out an error when dealing weighttying model.,https://github.com/pytorch/pytorch/issues/127357,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-06-05T15:25:57Z +issue,128019,"RuntimeError: r INTERNAL ASSERT FAILED at ""../aten/src/ATen/core/jit_type_base.h"":544 when using torch.jit.trace and torch._C._jit_interpret_graph",https://github.com/pytorch/pytorch/issues/128019,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-05T14:49:23Z +issue,46948,Record shaping assertions and use them for tracing / scripting optimization and codegen,https://github.com/pytorch/pytorch/issues/46948,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: codegen, oncall: jit",categorize and decide next maintainer owner/action,2024-06-04T19:35:57Z +issue,127679,RuntimeError: `jit.freeze` fails to find externally assigned attributes,https://github.com/pytorch/pytorch/issues/127679,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-06-03T17:52:32Z +issue,127643,[export] Errors out when unflattening TorchTitan,https://github.com/pytorch/pytorch/issues/127643,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-06-03T17:46:32Z +issue,124697,[INDUCTOR] [CPU] [GPT-FAST-MOE] large perf regression with coordinate_descent_tuning disabled,https://github.com/pytorch/pytorch/issues/124697,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-06-03T02:20:40Z +issue,127062,T5 -small Dynamic quantization in graviton3,https://github.com/pytorch/pytorch/issues/127062,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-30T09:36:26Z +issue,127076,[PT2E Quantization] `prepare_pt2e` produces inconsistent data types for primitive int,https://github.com/pytorch/pytorch/issues/127076,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-30T01:18:40Z +issue,115985,Faster Pytorch dequantize() + matmul for quantized models,https://github.com/pytorch/pytorch/issues/115985,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-29T19:02:28Z +issue,127215,No way to config profiling scope in torch.autograd.profile,https://github.com/pytorch/pytorch/issues/127215,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-05-28T12:15:52Z +issue,126770,_convert_input_to_fake doesn't work when there are no inputs,https://github.com/pytorch/pytorch/issues/126770,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-05-27T16:26:00Z +issue,126974,TypeError: slice indices must be integers or None or have an __index__ method,https://github.com/pytorch/pytorch/issues/126974,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-24T06:14:06Z +issue,127013,[CI] Download android and ios dependencies from pip instead of conda,https://github.com/pytorch/pytorch/issues/127013,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-05-23T22:35:45Z +issue,126843,Torchscript export of `Salesforce/blip2-opt-2.7b` model fails with Conv shape error,https://github.com/pytorch/pytorch/issues/126843,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-22T19:33:49Z +issue,118165,[Quantization-aware training] Can not quantize nn.MultiheadAttention module,https://github.com/pytorch/pytorch/issues/118165,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-22T03:36:40Z +issue,126291,Pytorch -> ExecuTorch conversion fails with BatchNorm2d layer,https://github.com/pytorch/pytorch/issues/126291,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-05-16T20:03:38Z +issue,125293,Need a C++ way to convert PyTorch model to TorchScript model,https://github.com/pytorch/pytorch/issues/125293,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-16T11:07:20Z +issue,126277,[JIT] Unexpected change in requires_grad attribute after calling JIT traced function with the profiling executor disabled,https://github.com/pytorch/pytorch/issues/126277,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-16T00:53:33Z +issue,126276,torch script not support ModuleList/Sequential ?,https://github.com/pytorch/pytorch/issues/126276,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-16T00:53:19Z +issue,126167,`aten::transpose.int` is acting errouneously,https://github.com/pytorch/pytorch/issues/126167,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-15T13:59:44Z +issue,125266,[export] linalg ops after core aten decomposition,https://github.com/pytorch/pytorch/issues/125266,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: decompositions, oncall: export",categorize and decide next maintainer owner/action,2024-05-14T20:25:16Z +issue,126140,[PT2E Quantization] _fuse_conv_bn_ | RuntimeError: !argument.default_value() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/126140,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-14T18:19:10Z +issue,125749,'MultiHeadAttention.attention' is being compiled since it was called from 'MultiHeadAttention.forward',https://github.com/pytorch/pytorch/issues/125749,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-08T09:58:40Z +issue,59612,Why Pytroch.distributed does not expose NCCL cuda stream?,https://github.com/pytorch/pytorch/issues/59612,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2024-05-08T07:06:07Z +issue,125734,☂️ `torch.compile` generates slower code for LLMs than eager on ARM platform (M1/AARCH64),https://github.com/pytorch/pytorch/issues/125734,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: performance, oncall: cpu inductor, oncall: pt2",categorize and decide next maintainer owner/action,2024-05-08T01:18:19Z +issue,110259,Cannot avoid kineto_LIBRARY-NOTFOUND error when using pre-built pytorch,https://github.com/pytorch/pytorch/issues/110259,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: profiler",categorize and decide next maintainer owner/action,2024-05-07T10:16:03Z +issue,125564,Issues in loading quantized weights with Version==None,https://github.com/pytorch/pytorch/issues/125564,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-07T06:16:04Z +issue,123345,torch.export does not support Sparse CSR tensors,https://github.com/pytorch/pytorch/issues/123345,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-05-06T20:28:02Z +issue,125003,bool as scalar in jit ir API not working as expected.,https://github.com/pytorch/pytorch/issues/125003,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-06T07:09:10Z +issue,66623,Constant folding in symbolic shape inference fails: expected scalar type Long but found Float,https://github.com/pytorch/pytorch/issues/66623,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-05-05T14:35:07Z +issue,125004,RuntimeError: Unsupported qscheme: per_channel_affine. When using quantise_fx(,https://github.com/pytorch/pytorch/issues/125004,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-05-01T03:03:22Z +issue,125245,[export] Add customization API for unflattener to control restored module types.,https://github.com/pytorch/pytorch/issues/125245,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-04-30T17:12:08Z +issue,124458,Understand the oneDNN graph fusion with torch script,https://github.com/pytorch/pytorch/issues/124458,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-30T03:33:17Z +issue,125007,Support exporting the compute graph JiT compiled by inductor?,https://github.com/pytorch/pytorch/issues/125007,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-04-29T16:02:19Z +issue,76881,RecursionError when running torch.jit.script inside JitTestCase,https://github.com/pytorch/pytorch/issues/76881,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-29T03:06:09Z +issue,125033,Tensorboard's SummaryWriter.add_graph() doesn't work with packedSequence,https://github.com/pytorch/pytorch/issues/125033,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2024-04-26T22:01:38Z +issue,124674,Improving format of communication metadata in PyTorch Execution Trace.,https://github.com/pytorch/pytorch/issues/124674,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-04-25T19:37:27Z +issue,80861,"jit.freeze throws RuntimeError: stack_out && stack_out->size() == 1 INTERNAL ASSERT FAILED at ""../torch/csrc/jit/passes/frozen_conv_folding.cpp"":281",https://github.com/pytorch/pytorch/issues/80861,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-25T13:37:38Z +issue,59530,Batch size is hardcoded using torch.jit.trace with LSTMCell,https://github.com/pytorch/pytorch/issues/59530,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-25T07:22:44Z +issue,124796,Dynamo Export Support for Qwen/Qwen-7B-Chat: Mutating module attribute _ntk_alpha_cached_list during export,https://github.com/pytorch/pytorch/issues/124796,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",categorize and decide next maintainer owner/action,2024-04-25T02:21:03Z +issue,124507,torch.export of ResNet with dynamic height fails due to constraint violation,https://github.com/pytorch/pytorch/issues/124507,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-04-21T16:46:40Z +issue,124461,Support constant value in torch.jit.script,https://github.com/pytorch/pytorch/issues/124461,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-19T15:23:51Z +issue,76807,There is a bug with latest stable torch version and the following Nightly versions related to `optimize_for_mobile`,https://github.com/pytorch/pytorch/issues/76807,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-04-19T14:03:06Z +issue,123961,JIT reports RuntimeError when decorating forward with torch.jit.script,https://github.com/pytorch/pytorch/issues/123961,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-13T03:26:59Z +issue,52363,RuntimeError: Can't redefine method: forward on class: __torch__.torch.nn.quantized.modules.conv.Conv1d (of Python compilation unit at: 0x5571a223a770),https://github.com/pytorch/pytorch/issues/52363,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-12T19:41:05Z +issue,123824,Precision problem in `quantize_fx.convert_fx` and question about the `example_inputs` for `quantize_fx.prepare_fx`.,https://github.com/pytorch/pytorch/issues/123824,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-04-12T19:17:06Z +issue,69425,Integrate Libtorch into Unreal Engine 4: _ivalue_INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/69425,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-11T22:20:10Z +issue,84318,[Profiler] Generic Tensor summary,https://github.com/pytorch/pytorch/issues/84318,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: profiler",categorize and decide next maintainer owner/action,2024-04-11T04:14:15Z +issue,123464,torch._C.JitType annotation is busted,https://github.com/pytorch/pytorch/issues/123464,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: typing, oncall: jit",categorize and decide next maintainer owner/action,2024-04-09T18:54:23Z +issue,58845,Use jit in flask,https://github.com/pytorch/pytorch/issues/58845,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-07T08:58:45Z +issue,123066,[JIT archive] Add a flag to not include debug files,https://github.com/pytorch/pytorch/issues/123066,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-04-01T16:05:16Z +issue,122303,How to exclude some modules from quantization?,https://github.com/pytorch/pytorch/issues/122303,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-03-27T08:22:57Z +issue,66017,Torchscript `jit.script` breaks with OSError if dataclass used to pass results during model execution,https://github.com/pytorch/pytorch/issues/66017,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-03-26T16:22:01Z +issue,120437,fx::GraphModule::to_folder doesn't seem to handle _PyTreeCodeGen correctly.,https://github.com/pytorch/pytorch/issues/120437,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: export",categorize and decide next maintainer owner/action,2024-03-23T04:28:23Z +issue,121230,INTERNAL ASSERT FAILED in torch.pow,https://github.com/pytorch/pytorch/issues/121230,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-03-22T21:51:04Z +issue,122183,[FX Graph Mode Quantization] Prepare model fails with Linear-BN when `affine=False`,https://github.com/pytorch/pytorch/issues/122183,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-03-22T21:48:09Z +issue,122224,support QAT fusion for conv transpose + bn + relu,https://github.com/pytorch/pytorch/issues/122224,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-03-19T22:26:36Z +issue,117239,```torch.nn.Upsample``` class: Internal assert failed,https://github.com/pytorch/pytorch/issues/117239,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-03-19T21:52:07Z +issue,87145,Unrecognized data format when using release libtorch libraries in debug build,https://github.com/pytorch/pytorch/issues/87145,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-03-19T16:04:31Z +issue,122020,torch.jit.script(model) and torch.jit.trace(model) performance degradation.,https://github.com/pytorch/pytorch/issues/122020,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-03-18T21:52:38Z +issue,69782,Torch Profiler does not count FLOPs for backward pass,https://github.com/pytorch/pytorch/issues/69782,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2024-03-18T14:16:18Z +issue,121597,LibTorch-Lite 2.1.0 Libs missing x86_64,https://github.com/pytorch/pytorch/issues/121597,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: mobile",categorize and decide next maintainer owner/action,2024-03-11T20:15:23Z +issue,75556,[JIT] `torch.jit.ignore` is not working on hooks,https://github.com/pytorch/pytorch/issues/75556,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-03-07T05:50:54Z +issue,120705,Trying doing static_quantization with Pytorch 2.1.0 torch.ao.quantization,https://github.com/pytorch/pytorch/issues/120705,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-03-05T09:02:43Z +issue,96639,Completely different output between .pt and .ptl,https://github.com/pytorch/pytorch/issues/96639,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2024-03-04T17:52:59Z +issue,63929,[Poll] Support higher-order gradient computation in DDP,https://github.com/pytorch/pytorch/issues/63929,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2024-03-02T00:29:57Z +issue,70051,[feature request] Support python decorators in TorchScript,https://github.com/pytorch/pytorch/issues/70051,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-02-29T09:58:29Z +issue,59690,Inference ran on new thread leak memory on Android,https://github.com/pytorch/pytorch/issues/59690,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, module: memory usage, oncall: mobile",categorize and decide next maintainer owner/action,2024-02-26T03:13:33Z +issue,120485,Try to make `jit.trace` working with dynamic control flow but it does not work,https://github.com/pytorch/pytorch/issues/120485,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-02-23T14:26:07Z +issue,43388,"Abort message: ‘terminating with uncaught exception of type c10::Error: _ivalue_ INTERNAL ASSERT FAILED at ../torch/csrc/jit/api/object.cpp:19, please report a bug to PyTorch. (_ivalue at ../torch/csrc/jit/api/object.cpp:19)",https://github.com/pytorch/pytorch/issues/43388,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2024-02-23T09:11:01Z +issue,89386,How can i patch the torch.jit in the second solution? Could not figure out entrypoint ?,https://github.com/pytorch/pytorch/issues/89386,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-02-22T17:10:32Z +issue,119484,Follow-up #119356: Preserve node names for torch.export.export,https://github.com/pytorch/pytorch/issues/119484,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-02-12T17:45:48Z +issue,119454,Python 3.12 CPU test_fx.py::TestFunctionalTracing::test_nn_functional_adaptive_max_pool is failing,https://github.com/pytorch/pytorch/issues/119454,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: fx, module: tests, oncall: fx",categorize and decide next maintainer owner/action,2024-02-08T16:47:07Z +issue,119192,[JIT] Deserialized `RecursiveScriptFunction` forgets default values for arguments.,https://github.com/pytorch/pytorch/issues/119192,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-02-08T15:27:29Z +issue,118410,[export] Error loudly when tracing over a module not in the module hierarchy,https://github.com/pytorch/pytorch/issues/118410,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2024-02-06T23:28:17Z +issue,14867,[C++/Pytorch] Get input shapes,https://github.com/pytorch/pytorch/issues/14867,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-02-06T01:41:33Z +issue,118463,Inconsistency in Output for `torch.argmin` in JIT Compiled Model,https://github.com/pytorch/pytorch/issues/118463,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-29T19:07:42Z +issue,118462,Inconsistency in Output for `torch.nn.functional.poisson_nll_loss` in JIT Compiled Model,https://github.com/pytorch/pytorch/issues/118462,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: loss, oncall: jit",categorize and decide next maintainer owner/action,2024-01-29T19:07:30Z +issue,118297,Inconsistent outputs of `cosine_similarity` and `linalg_svdvals` between eagermode and torch.jit.trace(),https://github.com/pytorch/pytorch/issues/118297,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-29T17:19:45Z +issue,118401,Inconsistency in Output When Using `torch.nn.functional.selu` with JIT Compilation,https://github.com/pytorch/pytorch/issues/118401,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-29T15:12:05Z +issue,118400,Inconsistent output of `torch.lodget` with JIT compilation,https://github.com/pytorch/pytorch/issues/118400,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-29T15:11:57Z +issue,118189,Cannot load a jit-traced model if it uses nan_to_num with too large a constant.,https://github.com/pytorch/pytorch/issues/118189,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-29T14:56:29Z +issue,118303,Inconsistent outputs of `mish` and `log10` between eagermode and torch.jit.trace,https://github.com/pytorch/pytorch/issues/118303,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-27T10:00:08Z +issue,118302,Inconsistent outputs of `torch.adaptive_avg_pool1d` and `torch.threshold_` between eagermode and torch.jit.trace,https://github.com/pytorch/pytorch/issues/118302,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-27T10:00:01Z +issue,118300,Inconsistent outputs in `linalg_ldl_factor_ex` and `special_spheriacal_bessel_j0` between eagermode and torch.jit.trace,https://github.com/pytorch/pytorch/issues/118300,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-27T09:59:47Z +issue,82417,[JIT] _unsafe_view returns alias when size(input) = size argument,https://github.com/pytorch/pytorch/issues/82417,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-25T04:48:42Z +issue,118086,Different behaviors with torch.isfinite and torch.tril between eagermode and JIT,https://github.com/pytorch/pytorch/issues/118086,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-24T09:24:08Z +issue,117920,Could not run 'quantized::embedding_byte' with arguments from the 'QuantizedCPU' backend.,https://github.com/pytorch/pytorch/issues/117920,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-01-22T13:28:02Z +issue,113945,torch.jit.trace has incorrect execution for += operation during compilation,https://github.com/pytorch/pytorch/issues/113945,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-18T11:38:41Z +issue,117541,Jit model running on cuda:1 uses memory on cuda:0,https://github.com/pytorch/pytorch/issues/117541,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-16T14:38:04Z +issue,49538,jit tracer doesn't work with unflatten layer,https://github.com/pytorch/pytorch/issues/49538,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-12T15:06:42Z +issue,54140,[Torchscript] Error detected in torch::jit::(anonymous namespace)::DifferentiableGraphBackward on custom RNN model,https://github.com/pytorch/pytorch/issues/54140,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-12T03:08:58Z +issue,49444,[RFC] Add JIT graph fuser for oneDNN Graph API,https://github.com/pytorch/pytorch/issues/49444,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-12T03:07:52Z +issue,5489,"torch.jit.trace(network, data) fails if data is an OrderedDict",https://github.com/pytorch/pytorch/issues/5489,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2024-01-12T03:07:42Z +issue,48325,[JIT] Constant Propagation Shouldn't Run on Values that Escape the Graph,https://github.com/pytorch/pytorch/issues/48325,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-12T03:06:39Z +issue,43501,[jit] Shape hints for TorchScript,https://github.com/pytorch/pytorch/issues/43501,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-12T03:06:16Z +issue,41909,Dropout in TorchScript under torch.no_grad() produces incoherent results,https://github.com/pytorch/pytorch/issues/41909,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-12T03:06:12Z +issue,117229,```torch.nn.functional.pad```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117229,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-01-11T19:43:44Z +issue,117102,conver_fx report len(device) > 1 error,https://github.com/pytorch/pytorch/issues/117102,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-01-11T15:09:57Z +issue,116992,torch.fake_quantize_per_tensor_affine with inf input provides wrong output value,https://github.com/pytorch/pytorch/issues/116992,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-01-09T20:39:52Z +issue,116865,Optimizing Per-Channel Quantization for Improved Inference Performance,https://github.com/pytorch/pytorch/issues/116865,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-01-09T02:05:44Z +issue,116253,`torch.Tensor.flipud`: Heap-buffer-overflow,https://github.com/pytorch/pytorch/issues/116253,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2024-01-05T21:57:07Z +issue,116612,Loaded torch.jit traced module treats dtypes incorrectly in autograd,https://github.com/pytorch/pytorch/issues/116612,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2024-01-03T08:34:22Z +issue,82065,optimize_for_mobile vulkan_prepack::conv2d_clamp_prepack,https://github.com/pytorch/pytorch/issues/82065,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: vulkan, oncall: mobile",categorize and decide next maintainer owner/action,2024-01-02T21:58:06Z +issue,116544,[export] namedtuple inputs can't be retraced,https://github.com/pytorch/pytorch/issues/116544,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2023-12-30T19:41:03Z +issue,116498,torch 1.8 cannot torch.jit.load for script model,https://github.com/pytorch/pytorch/issues/116498,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-29T00:50:59Z +issue,116240,[inductor][cpu] RuntimeError: quantized_resize_cpu_ does not have a deterministic implementation,https://github.com/pytorch/pytorch/issues/116240,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpu, oncall: quantization",categorize and decide next maintainer owner/action,2023-12-24T14:36:04Z +issue,116250,`torch.Tensor.rot90`: Heap-buffer-overflow,https://github.com/pytorch/pytorch/issues/116250,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2023-12-21T16:07:05Z +issue,116252,`torch.Tensor.fliplr`: Heap-buffer-overflow,https://github.com/pytorch/pytorch/issues/116252,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2023-12-21T16:06:58Z +issue,116251,`torch.Tensor.flip`: Heap-buffer-overflow,https://github.com/pytorch/pytorch/issues/116251,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2023-12-21T16:06:51Z +issue,116254,C++ API `at::quantized_max_pool2d`: Heap-buffer-overflow,https://github.com/pytorch/pytorch/issues/116254,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,categorize and decide next maintainer owner/action,2023-12-21T16:06:37Z +issue,116010,Why has no torch::jit::load() in Pytorch Android's libtorch_cpu.so,https://github.com/pytorch/pytorch/issues/116010,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-19T05:16:21Z +issue,115914,torch.jit.script fails when parsing torch.SymInt,https://github.com/pytorch/pytorch/issues/115914,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-17T05:35:16Z +issue,115736,Precision Loss in JIT ScriptModule at float64 Tensor Init,https://github.com/pytorch/pytorch/issues/115736,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-15T10:34:26Z +issue,115733,RuntimeError: thread_predicates_.find(tv_inp) != thread_predicates_.end(),https://github.com/pytorch/pytorch/issues/115733,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-13T16:15:43Z +issue,115622,Torch jit load model with enums,https://github.com/pytorch/pytorch/issues/115622,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-13T16:15:19Z +issue,115639,"`torch.jit.save` fails for static function from C++ TorchScript extension with ""Failed to downcast a Function to a GraphFunction""",https://github.com/pytorch/pytorch/issues/115639,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-13T13:47:47Z +issue,115488,torch.export.export is overriding source_fn inside higher order op submodules,https://github.com/pytorch/pytorch/issues/115488,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2023-12-11T18:23:38Z +issue,115055,Better test suite for torch.export.export,https://github.com/pytorch/pytorch/issues/115055,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2023-12-07T20:25:56Z +issue,62962,[JIT] NVRTC unknown error,https://github.com/pytorch/pytorch/issues/62962,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-07T03:11:09Z +issue,55603,Faster Inference time for first .pt model prediction,https://github.com/pytorch/pytorch/issues/55603,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-05T17:42:07Z +issue,75788,`torch.jit.script` Script functions do return `requires_grad = False` if `torch.no_grad()` has been used,https://github.com/pytorch/pytorch/issues/75788,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-12-05T17:39:42Z +issue,114967,Inplace update to buffers doesn't work with `aot_compile`,https://github.com/pytorch/pytorch/issues/114967,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: export,categorize and decide next maintainer owner/action,2023-12-04T21:24:20Z +issue,114102,"""aten::prelu"" can not be used in the metal backend",https://github.com/pytorch/pytorch/issues/114102,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-12-03T03:51:00Z +issue,52471,all reduce hangs and does not throw exception when CUDA_VISIBLE_DEVICES is not set properly using the NCCL backend,https://github.com/pytorch/pytorch/issues/52471,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",categorize and decide next maintainer owner/action,2023-11-30T13:38:05Z +issue,114543,[torch.jit.script] Torchscript produces incorrect result when argmax result is used in indexing,https://github.com/pytorch/pytorch/issues/114543,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-28T07:53:03Z +issue,114283,CppException in Android Studio with faster_rcnn_fbnetv3g_fpn.yaml Configuration,https://github.com/pytorch/pytorch/issues/114283,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-11-21T23:13:24Z +issue,92073,compilig MultiHeadAttention,https://github.com/pytorch/pytorch/issues/92073,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-14T02:47:48Z +issue,90885,TorchScript with complex abs doesn't work in backward,https://github.com/pytorch/pytorch/issues/90885,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-13T17:21:49Z +issue,66930,JIT vs. eager mismatches for jit.traced `int8` to `int32` casting,https://github.com/pytorch/pytorch/issues/66930,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-11T09:33:32Z +issue,92987,Test Failure: TestUpgraders.test_aten_div_scalar_at_3 on a big-endian machine (issue in torch.jit.load()),https://github.com/pytorch/pytorch/issues/92987,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-08T10:39:06Z +issue,80561,Cannot run scripted BERT_Pytorch,https://github.com/pytorch/pytorch/issues/80561,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-03T20:06:16Z +issue,112425,torch.jit.trace output changes when I add comments to the file,https://github.com/pytorch/pytorch/issues/112425,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-02T14:18:40Z +issue,89114,jit.script() fails to resolve/cast Optional[Tensor] fields of sub-modules or base classes of the object being scripted,https://github.com/pytorch/pytorch/issues/89114,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-01T08:03:04Z +issue,76295,TorchScript: jit.script fails when using 'tolist' with int32 tensors.,https://github.com/pytorch/pytorch/issues/76295,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-11-01T04:05:47Z +issue,69609,[mobile_optimizer] RuntimeError: We don't have an op for metal_prepack::conv2d_prepack,https://github.com/pytorch/pytorch/issues/69609,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-10-29T16:18:35Z +issue,111792,torch.jit.trace is not able to trace torch extension,https://github.com/pytorch/pytorch/issues/111792,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-24T02:31:17Z +issue,111863,Jit scripting support for `|` and mixing `typing`.,https://github.com/pytorch/pytorch/issues/111863,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-23T23:58:55Z +issue,97047,Optimize for mobile produces incorrect result with INSERT_FOLD_PREPACK_OPS optimization,https://github.com/pytorch/pytorch/issues/97047,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2023-10-13T09:55:31Z +issue,111135,"No op for aten::where with argument types: Tensor, Tensor, bool.",https://github.com/pytorch/pytorch/issues/111135,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-12T14:06:34Z +issue,98695,Torchscript: Name Mangling prevents Type Refinement,https://github.com/pytorch/pytorch/issues/98695,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-11T21:15:44Z +issue,68568,"About ModuleDict indexing when save model based on ""torch.jit.script()"" method",https://github.com/pytorch/pytorch/issues/68568,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-09T16:57:05Z +issue,103070,Unexpected failure in LLVM JIT when running TorchScript model in C++,https://github.com/pytorch/pytorch/issues/103070,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-08T03:05:12Z +issue,29843,torch.distributions.normal.Normal is not JIT supported,https://github.com/pytorch/pytorch/issues/29843,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-06T18:35:26Z +issue,110379,Pytorch LoadNativeLibrary issue,https://github.com/pytorch/pytorch/issues/110379,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-10-04T19:52:30Z +issue,42885,Unable to call `super` method with TorchScript,https://github.com/pytorch/pytorch/issues/42885,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-04T17:26:32Z +issue,48932,Unknown builtin op: torchvision::nms when loading scripted FasterRCNN,https://github.com/pytorch/pytorch/issues/48932,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-10-03T17:35:04Z +issue,89185,[feature request] Add ability to preserve traced shape during torch.jit.save and torch.jit.load,https://github.com/pytorch/pytorch/issues/89185,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-30T04:35:18Z +issue,109488,[bug] FALLBACK path has been taken inside: runCudaFusionGroup,https://github.com/pytorch/pytorch/issues/109488,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-25T09:33:10Z +issue,67146,`torch.jit.is_scripting()` not set when scripting a Module,https://github.com/pytorch/pytorch/issues/67146,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-20T16:54:58Z +issue,109699,[TorchScript] Support ScriptFunction arguments in torch.jit.script calls.,https://github.com/pytorch/pytorch/issues/109699,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-20T15:02:17Z +issue,109719,Make torch.cuda.graphs.is_current_stream_capturing() available in TorchScript,https://github.com/pytorch/pytorch/issues/109719,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, oncall: jit",categorize and decide next maintainer owner/action,2023-09-20T15:00:36Z +issue,109579,[Android: React Native] couldn't find DSO to load: libtorch-code-gen.so when loading model,https://github.com/pytorch/pytorch/issues/109579,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-09-20T00:11:45Z +issue,109113,"The API ""torch::jit::_load_for_mobile"" is limited to create an object living on the stack.",https://github.com/pytorch/pytorch/issues/109113,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-12T15:19:06Z +issue,84831,INTERNAL ASSERT FAILED for _jit_pass_vulkan_optimize_for_mobile (Google Colab),https://github.com/pytorch/pytorch/issues/84831,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: vulkan, oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2023-09-11T06:14:33Z +issue,93913,[BUG] jit.trace not working for torchvision ViT models,https://github.com/pytorch/pytorch/issues/93913,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-08T21:25:43Z +issue,108665,INTERNAL ASSERT FAILED in `shape_type_inference.cpp`,https://github.com/pytorch/pytorch/issues/108665,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-08T14:50:29Z +issue,108865,Heap buffer overflow with `torch::load` on fuzzy data,https://github.com/pytorch/pytorch/issues/108865,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-08T14:14:31Z +issue,108670,torch.jit.script produces incorrect gradients,https://github.com/pytorch/pytorch/issues/108670,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-07T01:17:37Z +issue,108332,RuntimeError: dims.value().size() == self->getMaybeRFactorDomain().size(),https://github.com/pytorch/pytorch/issues/108332,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-04T22:29:02Z +issue,105821,JIT input aliasing does not support aten::fill_,https://github.com/pytorch/pytorch/issues/105821,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-09-01T11:47:17Z +issue,41869,TorchScript pack_padded_sequence and pad_packed_sequence run time error,https://github.com/pytorch/pytorch/issues/41869,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-31T20:30:33Z +issue,52047,Detect OpenMP Loop and this application may hang warnings,https://github.com/pytorch/pytorch/issues/52047,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: tests, oncall: jit",categorize and decide next maintainer owner/action,2023-08-31T08:14:27Z +issue,108226,torch.nn.functional.pad() with value type bool,https://github.com/pytorch/pytorch/issues/108226,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-30T14:42:47Z +issue,107771,libtorch infer error : CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/107771,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-24T22:58:42Z +issue,60937,[jit] About traced script module using in c++,https://github.com/pytorch/pytorch/issues/60937,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-23T04:38:35Z +issue,40561,Cannot manually assign a tensor to .grad from TorchScript,https://github.com/pytorch/pytorch/issues/40561,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-16T10:37:22Z +issue,106699,torch.jit.script: scripting doesn't work with wraps,https://github.com/pytorch/pytorch/issues/106699,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-10T11:19:47Z +issue,48082,Microsoft C++ exception: torch::jit::ErrorReport at memory location 0x0000001C0C3BBB40.,https://github.com/pytorch/pytorch/issues/48082,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-09T19:39:42Z +issue,32976,[JIT] pytorch 1.4 breaks torch.jit.script(LSTM/GRU),https://github.com/pytorch/pytorch/issues/32976,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-09T02:25:54Z +issue,106764,[Feature request] Add new API Tensor.device_as,https://github.com/pytorch/pytorch/issues/106764,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend, oncall: jit",categorize and decide next maintainer owner/action,2023-08-08T23:56:04Z +issue,68014,Add TORCHELASTIC_RESTART_COUNT env variable to DDP logging,https://github.com/pytorch/pytorch/issues/68014,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2023-08-07T13:48:54Z +issue,104450,"Numpy/scipy module works fine with Torch modules, but not TorchScript. How to torchscript a numpy/scipy module?",https://github.com/pytorch/pytorch/issues/104450,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-02T17:55:14Z +issue,105997,Flatbuffer torchscipt files don't load in PyTorch Android Lite 1.13.1,https://github.com/pytorch/pytorch/issues/105997,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-02T17:29:17Z +issue,106435,"[JIT] .item() dict keys cause `RuntimeError: Cannot create dict for key type 'Scalar', only int, float, complex, Tensor, device and string keys are supported`",https://github.com/pytorch/pytorch/issues/106435,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-02T13:51:35Z +issue,106128,Torchscript optimizer incorrectly applies constant propagation to convert prim::ListConstruct() into prim::Constant,https://github.com/pytorch/pytorch/issues/106128,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-01T16:17:05Z +issue,106129,Allow settingsetGraphExecutorOptimize default for all threads,https://github.com/pytorch/pytorch/issues/106129,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-08-01T16:17:00Z +issue,90365,[JIT] INTERNAL ASSERT FAILED when dispatching for `torch.Tensor.view`,https://github.com/pytorch/pytorch/issues/90365,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-27T21:51:50Z +issue,105917,Torch.jit.frontend.NotSupportedError: not supporting functions with variable number of arguments.,https://github.com/pytorch/pytorch/issues/105917,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-27T14:38:01Z +issue,105918,Torch.jit : RuntimeError: Unable to extract string literal index for ModuleDict,https://github.com/pytorch/pytorch/issues/105918,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-27T14:37:58Z +issue,105900,torch.jit.frontend.NotSupported when compiling stable-diffusion,https://github.com/pytorch/pytorch/issues/105900,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-27T14:27:36Z +issue,104959,Saving a LightningModule torch.jit.ScriptModule is incompatible with torch.amp.autocast,https://github.com/pytorch/pytorch/issues/104959,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), oncall: jit",categorize and decide next maintainer owner/action,2023-07-20T02:32:21Z +issue,43520,[jit] cannot use functools.partial with TorchScript,https://github.com/pytorch/pytorch/issues/43520,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-17T13:22:03Z +issue,105196,Error loading TorchScript model with torchvision::nms operation in libtorch,https://github.com/pytorch/pytorch/issues/105196,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-14T02:58:47Z +issue,104906,torch.jit.frontend.NotSupportedError: keyword-arg expansion is not supported: for dgl.nn.HeteroGraphConv(),https://github.com/pytorch/pytorch/issues/104906,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-13T02:28:38Z +issue,51135,[TorchScript Usability] Behaviors of used-defined subclasses of torch.tensor are silently ignored,https://github.com/pytorch/pytorch/issues/51135,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-10T10:00:16Z +issue,104712,torch.jit slicing error (styleganv2),https://github.com/pytorch/pytorch/issues/104712,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-07T07:55:26Z +issue,104732,"torch.jit.trace says ""Arguments for call are invalid"" on torch.ops.aten.sub(3, x, alpha=3)",https://github.com/pytorch/pytorch/issues/104732,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-07-07T01:23:04Z +issue,92294,jit.fork stalls multiprocessing dataloader,https://github.com/pytorch/pytorch/issues/92294,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader, oncall: jit",categorize and decide next maintainer owner/action,2023-07-06T07:06:00Z +issue,92246,test_qnnpack_add fails,https://github.com/pytorch/pytorch/issues/92246,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: xnnpack, oncall: mobile",categorize and decide next maintainer owner/action,2023-06-30T20:58:12Z +issue,104458,[Feature Request] Add a new overload of torch::jit::load to restore traced shape and type,https://github.com/pytorch/pytorch/issues/104458,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-06-30T19:56:02Z +issue,103788,"[Torch Mlir] avg_pool1d function padding init value should be (0,)",https://github.com/pytorch/pytorch/issues/103788,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-06-29T13:43:43Z +issue,51969,Support backward hooks in JIT,https://github.com/pytorch/pytorch/issues/51969,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-06-29T12:58:01Z +issue,104041,Scripted model is loaded on GPU but the inference seems to utilize the CPU with zero GPU utilization,https://github.com/pytorch/pytorch/issues/104041,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-06-22T20:56:21Z +issue,100730,`torch::jit::EliminateExceptions` lowering pass never completes on specific model,https://github.com/pytorch/pytorch/issues/100730,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-06-20T20:32:31Z +issue,62221,Torchscript Optimize for Mobile RuntimeError: Can't deepcopy IValue with tag: Enum,https://github.com/pytorch/pytorch/issues/62221,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-06-14T09:43:54Z +issue,103250,torch.jit.script mean(keepdim=True) segfaults on GPU,https://github.com/pytorch/pytorch/issues/103250,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-06-09T23:34:29Z +issue,47887,"ONNX export of a scripted submodule fails with ""Modules that are called during a trace must be registered as submodules of the thing being traced""",https://github.com/pytorch/pytorch/issues/47887,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: jit",categorize and decide next maintainer owner/action,2023-06-06T07:39:31Z +issue,102653,torchscript dataclasses have bad support for class types as fields,https://github.com/pytorch/pytorch/issues/102653,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-31T21:49:11Z +issue,33491,jit.trace checker fails for LSTM,https://github.com/pytorch/pytorch/issues/33491,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-30T09:27:03Z +issue,68340,Torch 1.10.0 - RuntimeError: requires_grad_ is not supported on ScriptModules,https://github.com/pytorch/pytorch/issues/68340,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-22T16:59:09Z +issue,101855,torch.jit.trace() Floating point exception,https://github.com/pytorch/pytorch/issues/101855,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-19T20:23:58Z +issue,58789,Type annotations for torch.jit.* decorators,https://github.com/pytorch/pytorch/issues/58789,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-18T19:13:35Z +issue,70058,Unknown builtin op: spconv::get_indice_pairs,https://github.com/pytorch/pytorch/issues/70058,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-18T06:15:05Z +issue,101055,DISABLE libtorch-2.0.0+cu117 destructor exception,https://github.com/pytorch/pytorch/issues/101055,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-11T03:52:30Z +issue,100636,tracing does not work when torch.distributions is involved,https://github.com/pytorch/pytorch/issues/100636,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-05-05T13:38:56Z +issue,99821,Tracer cannot infer type of Seq2SeqLMOutput,https://github.com/pytorch/pytorch/issues/99821,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-04-24T17:29:48Z +issue,99561,Pytorch mobile crashes on Android when loading a custom model,https://github.com/pytorch/pytorch/issues/99561,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2023-04-21T17:12:56Z +issue,96185,`ld: error: unknown argument '-force_load'` when linking libtorch on Android,https://github.com/pytorch/pytorch/issues/96185,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: mobile",categorize and decide next maintainer owner/action,2023-04-19T04:30:46Z +issue,98888,Changes to TorchScript autodiff changing default behavior are no longer accepted,https://github.com/pytorch/pytorch/issues/98888,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-04-12T14:44:00Z +issue,73142,Support passing Python code as a string to torch.jit.script,https://github.com/pytorch/pytorch/issues/73142,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-04-11T20:51:11Z +issue,98064,Module 'Sequential' has no attribute '_modules' :,https://github.com/pytorch/pytorch/issues/98064,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-04-10T13:36:29Z +issue,91387,getting issue 'typeindex' file not found in Littorch-Lite/install/include/ATen/core/custom_class.h,https://github.com/pytorch/pytorch/issues/91387,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-04-07T02:12:54Z +issue,98422,[FX] Symbolic trace over `torch.Tensor.${fn}` APIs,https://github.com/pytorch/pytorch/issues/98422,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: fx,categorize and decide next maintainer owner/action,2023-04-05T20:28:09Z +issue,52312,TorchScript can't wrap Python constants in current module,https://github.com/pytorch/pytorch/issues/52312,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-04-05T01:16:56Z +issue,98210,torch.jit.script + legacy executor mode has diff in some pattern,https://github.com/pytorch/pytorch/issues/98210,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-04-05T00:40:48Z +issue,98124,PyTorch Profiler fails recording functions,https://github.com/pytorch/pytorch/issues/98124,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2023-04-04T14:10:23Z +issue,86667,Adding a linear layer leads to failure of `optimize_for_mobile`,https://github.com/pytorch/pytorch/issues/86667,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, oncall: jit",categorize and decide next maintainer owner/action,2023-04-02T20:02:33Z +issue,97913,BUG torch.jit.annotate on List + torch.stack give wrong DTYPE,https://github.com/pytorch/pytorch/issues/97913,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-30T14:42:11Z +issue,94908,ASSERT(initialized()) Debug Error after JIT fusion on Windows,https://github.com/pytorch/pytorch/issues/94908,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, oncall: jit",categorize and decide next maintainer owner/action,2023-03-30T07:56:58Z +issue,52517,[doc] JIT-compatible c++ custom ops,https://github.com/pytorch/pytorch/issues/52517,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: jit",categorize and decide next maintainer owner/action,2023-03-28T08:07:58Z +issue,82312,"A/libc: Fatal signal 6 (SIGABRT), code -1 (SI_QUEUE) in tid 9792 (Background), pid 9674 (ample.testtorch)",https://github.com/pytorch/pytorch/issues/82312,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-03-27T05:33:20Z +issue,96448,Unable to move torch.jit.load-ed models to XLA devices,https://github.com/pytorch/pytorch/issues/96448,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: xla, oncall: jit",categorize and decide next maintainer owner/action,2023-03-22T23:13:19Z +issue,97343,Traced module shows non-deterministic behaviour on CUDA,https://github.com/pytorch/pytorch/issues/97343,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-22T20:04:12Z +issue,95858,[JIT] Support string type annotations in NamedTuples,https://github.com/pytorch/pytorch/issues/95858,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-22T15:25:28Z +issue,68921,[Bug] forward function error occurs when the scriptmodule (made in Python) including the hook function is loaded from C++(Libtorch1.8.2),https://github.com/pytorch/pytorch/issues/68921,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-18T02:49:36Z +issue,96735,when run python run_test.py -i test_ops_jit error like this. ValueError: option names {'--junit-xml-reruns'} already added,https://github.com/pytorch/pytorch/issues/96735,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-16T03:04:43Z +issue,69741,JIT: Support for `torch.autograd.functional.jacobian` in TorchScript,https://github.com/pytorch/pytorch/issues/69741,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-15T23:07:06Z +issue,96420,Add location information when exception are thrown in `torch.jit.annotations.try_ann_to_type`,https://github.com/pytorch/pytorch/issues/96420,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-14T21:13:44Z +issue,95960,`@torch.jit.unused` does not properly ignore unsupported function signature,https://github.com/pytorch/pytorch/issues/95960,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-14T18:11:14Z +issue,80007,when forward use **kwargs，how to construct the example_ Inputs parameter in jit.trace?,https://github.com/pytorch/pytorch/issues/80007,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-11T03:33:15Z +issue,79875,"A/libc: Fatal signal 6 (SIGABRT), code -6 in tid 11742 (objectdetection)",https://github.com/pytorch/pytorch/issues/79875,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2023-03-09T11:53:43Z +issue,76181,Allow `torch.fx` tracing on TorchScript models,https://github.com/pytorch/pytorch/issues/76181,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: jit",categorize and decide next maintainer owner/action,2023-03-08T16:28:19Z +issue,95789,drastic speed regression of torch.jit.load starting with the 20230301 nightly,https://github.com/pytorch/pytorch/issues/95789,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-08T09:32:49Z +issue,96098,Error during inference on iOS: INTERNAL ASSERT FAILED at it_type_base.h:535,https://github.com/pytorch/pytorch/issues/96098,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-03-07T02:11:46Z +issue,66547,"destroy_process_group() does not clear worker count, leading to ""Timed out initializing process group in store based barrier""",https://github.com/pytorch/pytorch/issues/66547,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2023-03-03T19:14:16Z +issue,75336,Libtorch Crash after Attempting to Evaluate Model after Opening an OpenGL Context,https://github.com/pytorch/pytorch/issues/75336,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-02-28T19:25:36Z +issue,95497,Import parameters from jit,https://github.com/pytorch/pytorch/issues/95497,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-02-28T16:25:08Z +issue,95560,torch.jit.load documentation doesn't specify if it is safe to load untrusted models or not,https://github.com/pytorch/pytorch/issues/95560,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: jit",categorize and decide next maintainer owner/action,2023-02-27T16:28:02Z +issue,89868,No pytorch_jni.dll file in libtorch 1.13.0 lib folder,https://github.com/pytorch/pytorch/issues/89868,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: java,categorize and decide next maintainer owner/action,2023-02-25T00:26:03Z +issue,78483,is the issue resolved? windows not pytorch_jni in path,https://github.com/pytorch/pytorch/issues/78483,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: java,categorize and decide next maintainer owner/action,2023-02-24T22:00:50Z +issue,90374,torch.utils.tensorboard import fails if a new protobuf > 3.20 is installed (bug in tensorboard/tensorflow but better guard against it),https://github.com/pytorch/pytorch/issues/90374,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2023-02-23T01:02:30Z +issue,95194,High Cuda Memory Consumption for Simple ResNet50 Inference,https://github.com/pytorch/pytorch/issues/95194,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-02-22T22:55:47Z +issue,36910,jit.script leads to RuntimeError: attribute lookup is not defined on python value of type in some cases,https://github.com/pytorch/pytorch/issues/36910,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-02-15T14:25:30Z +issue,50533,Non-avairable nightly version for converting models to NNAPI,https://github.com/pytorch/pytorch/issues/50533,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2023-02-08T17:03:06Z +issue,92740,[JIT] Applying `conv2d` over Constants Leads to Exception,https://github.com/pytorch/pytorch/issues/92740,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-02-08T06:56:27Z +issue,93350,Tensorboard SummaryWriter with cloud storage does not work on Mac,https://github.com/pytorch/pytorch/issues/93350,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2023-02-07T10:06:51Z +issue,93826,torch.jit.script does not work with DataParallel,https://github.com/pytorch/pytorch/issues/93826,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-02-03T15:54:36Z +issue,90608,Operator overload priority should not rely on static initialization order,https://github.com/pytorch/pytorch/issues/90608,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-02-02T21:34:56Z +issue,92812,"torch.fx fails to trace through ""+"" op between torch.Size and torch.fx.proxy.Proxy",https://github.com/pytorch/pytorch/issues/92812,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: fx,categorize and decide next maintainer owner/action,2023-02-02T03:05:22Z +issue,93255,MaskRCNN model loaded fail with torch::jit::load(model_path) (C++ API),https://github.com/pytorch/pytorch/issues/93255,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-31T21:23:47Z +issue,72498,torch.jit.script + torch.split + onnxruntime incompatibility,https://github.com/pytorch/pytorch/issues/72498,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-30T11:37:34Z +issue,84340,Floordiv is deprecated.,https://github.com/pytorch/pytorch/issues/84340,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-27T20:27:16Z +issue,93002,Replace pattern fails on incompatible function arguments,https://github.com/pytorch/pytorch/issues/93002,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: fx,categorize and decide next maintainer owner/action,2023-01-27T09:17:40Z +issue,92910,torch.jit.save() generates different contents in a file among different endian machines,https://github.com/pytorch/pytorch/issues/92910,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: power, oncall: jit",categorize and decide next maintainer owner/action,2023-01-27T08:46:23Z +issue,91293,torch.fx tracer emits type error when tracing module that directly contains and uses the torch.cat() function,https://github.com/pytorch/pytorch/issues/91293,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: fx,categorize and decide next maintainer owner/action,2023-01-26T13:10:04Z +issue,92888,[LibTorch] pickle_save output cannot be reloaded using pickle_load in Windows,https://github.com/pytorch/pytorch/issues/92888,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-25T06:37:25Z +issue,92866,JIT Function Fails when run a second time,https://github.com/pytorch/pytorch/issues/92866,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-25T06:26:38Z +issue,92742,[JIT] Consecutive use of `addmm` Leads to Exception,https://github.com/pytorch/pytorch/issues/92742,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-23T15:22:14Z +issue,51445,Saving embedding vectors as TSV for Tensorboard is not efficient,https://github.com/pytorch/pytorch/issues/51445,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2023-01-22T12:15:56Z +issue,92674,[JIT] `Linear` + `BatchNorm2d` Trigger Inconsistency between Eager Mode and JIT,https://github.com/pytorch/pytorch/issues/92674,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-20T17:21:23Z +issue,92375,operations failed in TorchScript interpreter,https://github.com/pytorch/pytorch/issues/92375,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-20T01:47:03Z +issue,92563,[JIT] INTERNAL ASSERT FAILED when `Conv2d` and `clamp` used together,https://github.com/pytorch/pytorch/issues/92563,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-20T01:42:14Z +issue,92548,[JIT] Inconsistency in tensor shape between eager mode and JIT,https://github.com/pytorch/pytorch/issues/92548,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-20T00:42:58Z +issue,92558,"[JIT][TracingCheckError] inplace ops incompatible with `contiguous(.., channels_last)`",https://github.com/pytorch/pytorch/issues/92558,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-20T00:28:22Z +issue,92654,Traced model output differs on C++ and Python,https://github.com/pytorch/pytorch/issues/92654,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-20T00:27:46Z +issue,92230,Add torch::jit::ScriptModule to the C++ API,https://github.com/pytorch/pytorch/issues/92230,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-18T13:11:31Z +issue,91396,[JIT] Zero-channel conv2d cannot be applied with `optimize_for_inference`,https://github.com/pytorch/pytorch/issues/91396,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-18T06:53:01Z +issue,92175,test_fx_passes generate bad test names,https://github.com/pytorch/pytorch/issues/92175,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: fx",categorize and decide next maintainer owner/action,2023-01-13T21:46:46Z +issue,86662,libtorch throws `required keyword attribute 'profiled_view_size' has the wrong type` on Linux,https://github.com/pytorch/pytorch/issues/86662,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-11T19:02:43Z +issue,91098,[TorchScript] Failed to Forward Correct Number of Arguments to Different Functions,https://github.com/pytorch/pytorch/issues/91098,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-11T06:54:05Z +issue,91903,Profiling with stack enabled results in error when Python's cProfile is also running,https://github.com/pytorch/pytorch/issues/91903,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2023-01-10T14:01:54Z +issue,91738,tracing torchvision detection model results in an error,https://github.com/pytorch/pytorch/issues/91738,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: vision, oncall: jit",categorize and decide next maintainer owner/action,2023-01-06T04:02:42Z +issue,91686,Sparse tensor not supported (Minkowski Engine),https://github.com/pytorch/pytorch/issues/91686,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-05T23:47:02Z +issue,64492,AttributeError: Can't pickle local object 'schedule..schedule_fn',https://github.com/pytorch/pytorch/issues/64492,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2023-01-04T16:15:19Z +issue,91358,[JIT] .backward() not supported by JIT trace,https://github.com/pytorch/pytorch/issues/91358,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-03T02:12:33Z +issue,91562,`torch::jit::optimize_for_inference` doesn't preserve exported methods when calling `freeze`,https://github.com/pytorch/pytorch/issues/91562,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-03T02:06:20Z +issue,91504,torch.jit.script ERR: RuntimeError: Can't redefine method: forward on class: __torch__.SoSadModule,https://github.com/pytorch/pytorch/issues/91504,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2023-01-03T01:42:58Z +issue,69352,I want to know how to read the LMDB file once when using DDP,https://github.com/pytorch/pytorch/issues/69352,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",categorize and decide next maintainer owner/action,2022-12-29T14:32:17Z +issue,89241,"Can't use JIT modules traced with AMP autocast, with Triton Server (or any C++ environment) - freeze() issue ?",https://github.com/pytorch/pytorch/issues/89241,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-29T03:42:56Z +issue,91360,Offer `get_buffer` and `get_submodule` in `ScriptModule`?,https://github.com/pytorch/pytorch/issues/91360,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-27T23:29:05Z +issue,91229,JIT mishandles torch.round() in PyTorch 1.10.1,https://github.com/pytorch/pytorch/issues/91229,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-22T21:23:05Z +issue,91311,Profiler is not properly recording seq number when any key above autograd is used,https://github.com/pytorch/pytorch/issues/91311,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2022-12-22T15:30:42Z +issue,69006,[JIT] allow `Optional[torch.Generator]` be tracable in Python,https://github.com/pytorch/pytorch/issues/69006,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-20T12:35:28Z +issue,90915,Torch SummaryWriter import fails with torch 2.0 with an error on numpy.object,https://github.com/pytorch/pytorch/issues/90915,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2022-12-15T22:30:19Z +issue,90305,Inconsistent Hash of IValue between aten/src/ATen/core/ivalue.cpp and aten/src/ATen/core/Dict_inl.h,https://github.com/pytorch/pytorch/issues/90305,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-12T18:12:45Z +issue,90367,[JIT] INTERNAL ASSERT FAILED `torch.add` with boolean primitive constant,https://github.com/pytorch/pytorch/issues/90367,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-09T16:51:57Z +issue,90366,[JIT] INTERNAL ASSERT FAILED `torch.mul` with boolean primitive constant,https://github.com/pytorch/pytorch/issues/90366,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-09T16:51:52Z +issue,90369,[JIT] Wrong type inference leads to misleading error message,https://github.com/pytorch/pytorch/issues/90369,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-09T16:51:44Z +issue,90466,Saving a scripted module to a buffer does not work.,https://github.com/pytorch/pytorch/issues/90466,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-09T16:12:12Z +issue,67902,torch::jit::pickle_load is unable to read some files saved by torch.save,https://github.com/pytorch/pytorch/issues/67902,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-09T03:42:03Z +issue,90301,Unknown buildin op: aten::pad,https://github.com/pytorch/pytorch/issues/90301,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-07T03:15:47Z +issue,90181,Graph is renamed in torch.jit,https://github.com/pytorch/pytorch/issues/90181,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-06T18:41:26Z +issue,37883,Support @property decorator in TorchScript,https://github.com/pytorch/pytorch/issues/37883,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-05T20:07:28Z +issue,21064,[JIT] Expose subgraph execution for intermediate output extraction,https://github.com/pytorch/pytorch/issues/21064,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-03T08:06:41Z +issue,58109,torch.jit.trace memory leak,https://github.com/pytorch/pytorch/issues/58109,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-12-02T16:23:39Z +issue,89076,Transformers model tracing not working,https://github.com/pytorch/pytorch/issues/89076,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-11-30T20:39:50Z +issue,68385,ImportError: cannot import name 'ProcessGroup' from 'torch.distributed',https://github.com/pytorch/pytorch/issues/68385,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: macos, oncall: distributed",categorize and decide next maintainer owner/action,2022-11-30T19:05:28Z +issue,71261,The jit model will fail when calling the torch.autograd.functional.jacobian with multiple inputs and setting the vectorize to true.,https://github.com/pytorch/pytorch/issues/71261,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2022-11-25T09:39:04Z +issue,89483,Unexpected behavior from torchscript (mixing trace with script),https://github.com/pytorch/pytorch/issues/89483,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-11-22T15:34:14Z +issue,89484,"kind_.is_prim() INTERNAL ASSERT FAILED at ""../torch/csrc/jit/ir/ir.cpp"":1098",https://github.com/pytorch/pytorch/issues/89484,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: jit",categorize and decide next maintainer owner/action,2022-11-22T15:13:35Z +issue,84304,op for aten::bitwise_and during torch.jit.trace,https://github.com/pytorch/pytorch/issues/84304,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-11-16T16:02:55Z +issue,89065,Improve clarity of meaning of `torch.jit.trace`'s `example_inputs`,https://github.com/pytorch/pytorch/issues/89065,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-11-15T23:08:28Z +issue,86791,We don't have an op for vulkan_prepack::conv2d_clamp_prepack but it isn't a special case.,https://github.com/pytorch/pytorch/issues/86791,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: vulkan, oncall: mobile",categorize and decide next maintainer owner/action,2022-11-14T08:54:08Z +issue,88765,torch.jit.trace() - AttributeError: 'NoneType' object has no attribute '__module__,https://github.com/pytorch/pytorch/issues/88765,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-11-14T01:37:48Z +issue,88735,RuntimeError: method '__torch__.___torch_mangle_0.MyModule.sin' already defined.,https://github.com/pytorch/pytorch/issues/88735,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-11-10T16:42:08Z +issue,75673,Tensorboard Issue with visualizing the connections of encoder-decoder network,https://github.com/pytorch/pytorch/issues/75673,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2022-11-10T08:10:23Z +issue,67972,Distributed.TCPStore,https://github.com/pytorch/pytorch/issues/67972,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2022-11-08T16:59:29Z +issue,88574,Add support for `torch.Generator` in the FX IR,https://github.com/pytorch/pytorch/issues/88574,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: fx,categorize and decide next maintainer owner/action,2022-11-08T09:53:21Z +issue,39250,SummaryWriter `add_hparams` should support adding new hyperparameters,https://github.com/pytorch/pytorch/issues/39250,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2022-11-06T21:31:17Z +issue,88491,torch.set_grad_enabled results in RuntimeError with torch.jit.script,https://github.com/pytorch/pytorch/issues/88491,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-11-04T21:30:15Z +issue,88448,linear mm weight and bias dtypes mismatch bypasses,https://github.com/pytorch/pytorch/issues/88448,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-11-04T21:28:16Z +issue,88301,1.12.1 incompatible with c++ built for 1.12.0 and vice versa,https://github.com/pytorch/pytorch/issues/88301,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: jit",categorize and decide next maintainer owner/action,2022-11-02T15:28:36Z +issue,51376,Huggingface model not being torch scriptable,https://github.com/pytorch/pytorch/issues/51376,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-31T09:49:48Z +issue,86572,Instructions for Selective Build for Mobile Linux Platform,https://github.com/pytorch/pytorch/issues/86572,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-10-31T00:35:52Z +issue,71277,JIT support for `torch.__version__` & str comparison operations,https://github.com/pytorch/pytorch/issues/71277,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2022-10-29T20:26:17Z +issue,87781,`torchtyping` annotations make saving to Torchscript fail,https://github.com/pytorch/pytorch/issues/87781,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-27T14:24:31Z +issue,87787,`BatchNorm` a 0-shape tensor will crash in JIT trace w/o profiling executor on cuda,https://github.com/pytorch/pytorch/issues/87787,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-27T14:23:06Z +issue,48452,After updating to Xcode 12 and LibTorch to 1.7.0. Facing issue when running unit test.,https://github.com/pytorch/pytorch/issues/48452,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: ios, oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2022-10-27T12:19:33Z +issue,87792,`AvgPool` and `MaxPool` will crash in JIT w/o profiling executor,https://github.com/pytorch/pytorch/issues/87792,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-26T17:52:05Z +issue,87487,[JIT] Inconsistent handling of tracing dict output leads to assertion,https://github.com/pytorch/pytorch/issues/87487,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-23T19:26:17Z +issue,60333,Two-element ModuleList results in error in inference_mode when jit'ed,https://github.com/pytorch/pytorch/issues/60333,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-23T06:58:16Z +issue,87268,register_package has no further documentation,https://github.com/pytorch/pytorch/issues/87268,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: package/deploy",categorize and decide next maintainer owner/action,2022-10-21T13:03:35Z +issue,87390,`chunk` a 0-dim tensor will crash in JIT script w/o profiling executor,https://github.com/pytorch/pytorch/issues/87390,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-21T12:55:27Z +issue,85513,"Measure impact of JIT decompositions, reconsider the design",https://github.com/pytorch/pytorch/issues/85513,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-20T21:29:31Z +issue,87289,'str' object has no attribute '__module__' in jit is_final,https://github.com/pytorch/pytorch/issues/87289,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-20T14:45:54Z +issue,77167,nn.functional.pad accepts bool values but raises internal assert when converted to JIT,https://github.com/pytorch/pytorch/issues/77167,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-20T05:50:10Z +issue,86818,Forward hooks for ScriptModules,https://github.com/pytorch/pytorch/issues/86818,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-18T01:45:48Z +issue,84321,[Profiler] Snapshot CudaCachingAllocator on profile begin,https://github.com/pytorch/pytorch/issues/84321,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: profiler",categorize and decide next maintainer owner/action,2022-10-15T17:56:39Z +issue,86816,JIT model returns different value on cpu with uniform-initialized input,https://github.com/pytorch/pytorch/issues/86816,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-13T17:17:32Z +issue,86804,JIT model will have a different jacobian after the first computation,https://github.com/pytorch/pytorch/issues/86804,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-13T17:16:31Z +issue,85877,JIT model could return 'NaN' gradient after the first execution,https://github.com/pytorch/pytorch/issues/85877,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-12T17:35:53Z +issue,86552,`max_unpool` and `max_pool` will trigger INTERNAL ASSERT FAIL in JIT,https://github.com/pytorch/pytorch/issues/86552,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-12T16:37:29Z +issue,63725,[Feature Request] add graph to hparams for tensorboard,https://github.com/pytorch/pytorch/issues/63725,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2022-10-11T15:16:07Z +issue,86287,JIT `lgamma` will return `inf` only with dual input in forward mode,https://github.com/pytorch/pytorch/issues/86287,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-10T18:22:20Z +issue,86616,JIT returns different values for a model on cuda and returns a strange error message on cpu,https://github.com/pytorch/pytorch/issues/86616,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), oncall: jit",categorize and decide next maintainer owner/action,2022-10-10T17:22:53Z +issue,86598,`squeeze_` fails with JIT but succeeds without it,https://github.com/pytorch/pytorch/issues/86598,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-10T17:06:11Z +issue,86597,JIT returns different values for `cos + frac` on cpu,https://github.com/pytorch/pytorch/issues/86597,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), oncall: jit",categorize and decide next maintainer owner/action,2022-10-10T15:08:59Z +issue,86553,`topk` will return the wrong value and could read out-of-bound value after jit,https://github.com/pytorch/pytorch/issues/86553,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-10T14:52:35Z +issue,86455,(JIT) x:Optional[T] cannot not expect content type after `if x is None or x.shape[0]==1`,https://github.com/pytorch/pytorch/issues/86455,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-08T04:23:57Z +issue,86351,Jetson JIT: Memory Leak on inference after optimize_for_inference,https://github.com/pytorch/pytorch/issues/86351,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-06T05:51:11Z +issue,86276,JIT miss the argument `as_tuple` for API `nonzero`,https://github.com/pytorch/pytorch/issues/86276,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-05T16:09:48Z +issue,81337,JIT trace takes forever on a simple method,https://github.com/pytorch/pytorch/issues/81337,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-04T05:02:46Z +issue,86124,torch.jit.trace throwing Invalid name for qualified name eror,https://github.com/pytorch/pytorch/issues/86124,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-03T16:09:30Z +issue,47336,[jit] Properly support slicing of ModuleList,https://github.com/pytorch/pytorch/issues/47336,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-03T16:09:12Z +issue,86116,JIT fails to trace binary cross entropy with a strange error msg,https://github.com/pytorch/pytorch/issues/86116,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-03T15:18:54Z +issue,86112,`cdist` should succeed when `p` is integer in JIT,https://github.com/pytorch/pytorch/issues/86112,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-03T13:19:09Z +issue,86110,JIT return a tensor with different datatype from the tensor w/o gradient and normal function,https://github.com/pytorch/pytorch/issues/86110,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-10-03T12:50:01Z +issue,79003,Libtorch C++ mobile build linking error,https://github.com/pytorch/pytorch/issues/79003,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: mobile",categorize and decide next maintainer owner/action,2022-10-01T06:41:03Z +issue,85706,RuntimeError: [enforce fail at CPUAllocator.cpp:68] . DefaultCPUAllocator: can't allocate memory: you tried to allocate 4860032 bytes. Error code 12 (Cannot allocate memory),https://github.com/pytorch/pytorch/issues/85706,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-29T17:39:33Z +issue,85893,TorchScript does not recognize mix-in types with `Enum`,https://github.com/pytorch/pytorch/issues/85893,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-29T14:14:47Z +issue,85792,TorchScript causes range_check error after a few iterations of forward-backward passes,https://github.com/pytorch/pytorch/issues/85792,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-29T08:18:01Z +issue,85149,Pytorch on iOS (iPhone X & XR) throwing can't allocate memory exception. Ref Logs:,https://github.com/pytorch/pytorch/issues/85149,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",categorize and decide next maintainer owner/action,2022-09-29T05:31:15Z +issue,84681,JIT will affect the gradient computation of forward mode,https://github.com/pytorch/pytorch/issues/84681,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-28T16:41:17Z +issue,85813,TorchScript error for `Enum` inside a module,https://github.com/pytorch/pytorch/issues/85813,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-28T15:16:04Z +issue,85805,"`vector_norm` will trigger ""Tracing failed sanity checks"" for JIT when ord is boolean tensor",https://github.com/pytorch/pytorch/issues/85805,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-28T14:08:50Z +issue,85804,JIT fails to trace `sparse.mm` with a strange error,https://github.com/pytorch/pytorch/issues/85804,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-28T14:01:03Z +issue,69616,Multi-GPU training stuck when using grad scaler,https://github.com/pytorch/pytorch/issues/69616,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, oncall: distributed",categorize and decide next maintainer owner/action,2022-09-28T05:28:40Z +issue,44951,copy.deepcopy not working properly for jit.TopLevelTracedModule,https://github.com/pytorch/pytorch/issues/44951,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-27T15:02:27Z +issue,85588,`linalg.norm` cannot compute the grad in forward mode after script,https://github.com/pytorch/pytorch/issues/85588,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-26T19:52:07Z +issue,85505,The reload model has different (and strange) forward computation from original model with `LSTMCell`,https://github.com/pytorch/pytorch/issues/85505,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-26T15:11:34Z +issue,85514,The reload `MultiLabelMarginLoss` will have different gradients on cuda,https://github.com/pytorch/pytorch/issues/85514,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-26T15:11:20Z +issue,85585,`as_tensor` will return a different dtype with script,https://github.com/pytorch/pytorch/issues/85585,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-26T15:03:24Z +issue,85573,`jit` could make some undifferentiable APIs differentiable,https://github.com/pytorch/pytorch/issues/85573,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-23T23:38:23Z +issue,85570,`mvlgamma_` will fail when compiling with trace `jit`,https://github.com/pytorch/pytorch/issues/85570,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-23T21:29:50Z +issue,85568,`detach_` behaves differently when computing the gradients in forward mode w/ `jit`,https://github.com/pytorch/pytorch/issues/85568,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-23T20:50:28Z +issue,85342,"JIT fuser issues with {ceil,floor,round,trunc}(int8)",https://github.com/pytorch/pytorch/issues/85342,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-23T17:40:48Z +issue,83250,Using Pytorch and Mapbox in the same project,https://github.com/pytorch/pytorch/issues/83250,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-09-15T21:03:18Z +issue,85072,Segmentation fault in `torch.jit.wait`,https://github.com/pytorch/pytorch/issues/85072,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-15T06:02:53Z +issue,61344,Get TensorType's device in python,https://github.com/pytorch/pytorch/issues/61344,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2022-09-14T01:20:02Z +issue,84685,[TensorExpr] applying `rfactor` for a Mul Reducer with init value different than 1 results in wrong results,https://github.com/pytorch/pytorch/issues/84685,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-13T21:46:41Z +issue,84495,functionalize: Does not compose cleanly with torch.jit.script/torch.jit.trace,https://github.com/pytorch/pytorch/issues/84495,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: functionalization, oncall: jit",categorize and decide next maintainer owner/action,2022-09-13T00:37:34Z +issue,84673,Autograd will take `init` module API into account when using `jit`,https://github.com/pytorch/pytorch/issues/84673,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-12T21:56:56Z +issue,84628,INTERNAL ASSERT when the type of argument is not considered in JIT,https://github.com/pytorch/pytorch/issues/84628,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-12T21:54:42Z +issue,84615,Serialize the warmed up torchscript module,https://github.com/pytorch/pytorch/issues/84615,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-12T20:41:09Z +issue,84588,torch.jit.script IndentationError: unexpected indent,https://github.com/pytorch/pytorch/issues/84588,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-11T16:03:15Z +issue,61966,Bootcamp Task: Add `prim::to_mkldnn` to convert from aten's nchw to mkldnn's nChw8c directly,https://github.com/pytorch/pytorch/issues/61966,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2022-09-10T18:56:54Z +issue,84646,JIT script calculation/dtype inconsistent depending on operator expression,https://github.com/pytorch/pytorch/issues/84646,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-09-07T16:59:45Z +issue,84055,scripted fasterRCNN model cannot be loaded with libtorch c++ API,https://github.com/pytorch/pytorch/issues/84055,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: jit",categorize and decide next maintainer owner/action,2022-08-31T11:09:21Z +issue,46375,JIT: error in LSTM with `flatten_parameters`,https://github.com/pytorch/pytorch/issues/46375,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-30T07:44:51Z +issue,84189,"RuntimeError: outputs_[i]->uses().empty() INTERNAL ASSERT FAILED at /pytorch/torch/csrc/jit/ir.cpp:1027, please report a bug to PyTorch. (eraseOutput at /pytorch/torch/csrc/jit/ir.cpp:1027)",https://github.com/pytorch/pytorch/issues/84189,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-30T02:49:56Z +issue,81635,Windows Debug binaries crash on forward: assert fail on IListRefIterator destructor,https://github.com/pytorch/pytorch/issues/81635,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-29T11:21:57Z +issue,84187,[jit] WithInsertPoint can't get back to the prev_ node if the prev_ node has been destroyed,https://github.com/pytorch/pytorch/issues/84187,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-29T06:58:01Z +issue,84135,[jit] ignored method calling static method results in an error,https://github.com/pytorch/pytorch/issues/84135,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-26T18:52:02Z +issue,84049,TorchScript unsupport tuple unpacking as function inputs.,https://github.com/pytorch/pytorch/issues/84049,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-25T10:45:59Z +issue,82598,Deep copy models with `create_feature_extractor` produces different parameters,https://github.com/pytorch/pytorch/issues/82598,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: vision, oncall: fx",categorize and decide next maintainer owner/action,2022-08-22T20:02:20Z +issue,62381,torch.distributed and subprocess do not work together?,https://github.com/pytorch/pytorch/issues/62381,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2022-08-22T13:10:12Z +issue,83769,[TorchTidy] Check if `set_to_none` would change optimizer semantics.,https://github.com/pytorch/pytorch/issues/83769,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2022-08-19T20:49:21Z +issue,76013,test_jit.py TestWarn.test_warn and friends doesn't work under pytest,https://github.com/pytorch/pytorch/issues/76013,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-19T16:05:20Z +issue,83551,OpenJDK libtorch_cpu.so stack guard warning,https://github.com/pytorch/pytorch/issues/83551,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: java,categorize and decide next maintainer owner/action,2022-08-17T14:18:30Z +issue,83589,Speedup for adding images to tensorboard,https://github.com/pytorch/pytorch/issues/83589,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2022-08-17T14:16:16Z +issue,83585,Segfault when profiling with_stack=True on model with jit.optimize_for_inference,https://github.com/pytorch/pytorch/issues/83585,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-17T11:43:27Z +issue,28206,torch.utils.tensorboard.SummaryWriter.add_graph do not support non-tensor inputs,https://github.com/pytorch/pytorch/issues/28206,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2022-08-14T10:43:52Z +issue,83349,checkpoint function is not jit compatible,https://github.com/pytorch/pytorch/issues/83349,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-12T20:25:16Z +issue,83143,Bug in building pytorch deploy from source in macos USE_DEPLOY=1,https://github.com/pytorch/pytorch/issues/83143,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-08-11T21:56:02Z +issue,83161,Optimize for mobile metal model,https://github.com/pytorch/pytorch/issues/83161,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-08-11T21:54:06Z +issue,75903,torch.jit.script'd function very slow on first invocation on latest nightly,https://github.com/pytorch/pytorch/issues/75903,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, oncall: jit",categorize and decide next maintainer owner/action,2022-08-11T17:28:29Z +issue,82951,torch.profiler's FLOPs measure only counts operations involving '+' and '*' .,https://github.com/pytorch/pytorch/issues/82951,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2022-08-10T13:23:47Z +issue,46460,C++ API: Training and inference of torchscript modules on multiple GPU,https://github.com/pytorch/pytorch/issues/46460,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-10T13:02:56Z +issue,83045,[JIT] Scripting modules fails for modules that contain nested NamedTuples,https://github.com/pytorch/pytorch/issues/83045,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-09T01:11:59Z +issue,82872,Unknown builtin op: torchvision::deform_conv2d,https://github.com/pytorch/pytorch/issues/82872,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-08T22:27:43Z +issue,82762,Fill in a bool Tensor not supported in jit,https://github.com/pytorch/pytorch/issues/82762,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-03T22:37:51Z +issue,81460,[JIT] Request Constant Propagation to keep fake_quantize_per_tensor_affine and fake_quantize_per_channel_affine on the graph,https://github.com/pytorch/pytorch/issues/81460,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-02T21:42:02Z +issue,82635,[Torchscript] torch.min returns wrong gradient when inputs are equal,https://github.com/pytorch/pytorch/issues/82635,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-02T05:16:58Z +issue,82634,[Torchscript] some activations backward are not fused when used with linear,https://github.com/pytorch/pytorch/issues/82634,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-02T05:11:23Z +issue,69931,[JIT] torch.exp roughly 20 times slower in TorchScript vs. PyTorch,https://github.com/pytorch/pytorch/issues/69931,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-08-01T14:03:29Z +issue,82451,add support for bitwise operations with floating point numbers,https://github.com/pytorch/pytorch/issues/82451,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-31T18:02:49Z +issue,62313,"On Triton inference server, the TorchScript .pt file, exported by 1.9.0, serves unsuccessfully",https://github.com/pytorch/pytorch/issues/62313,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-29T07:43:37Z +issue,68256,RuntimeError: CUDA error: initialization error when calling torch.distributed.init_process_group using torch multiprocessing,https://github.com/pytorch/pytorch/issues/68256,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2022-07-29T00:16:24Z +issue,81651,optimize_for_mobile has an issue with constant operations at the end of a loop,https://github.com/pytorch/pytorch/issues/81651,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-07-27T22:12:07Z +issue,82324,[JIT] SchemaInfo warning appears out in the wild,https://github.com/pytorch/pytorch/issues/82324,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-27T15:47:50Z +issue,30365,TorchScript Performance: 150x gap between TorchScript and Native Python,https://github.com/pytorch/pytorch/issues/30365,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-26T23:00:36Z +issue,82216,iOS TestApp from mobile performance recipes tutorial doesn't build on macOS,https://github.com/pytorch/pytorch/issues/82216,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",categorize and decide next maintainer owner/action,2022-07-26T16:49:11Z +issue,60844,DDP fails if you have multiple forward passes and a single backwards pass with `find_unused_parameters=True`,https://github.com/pytorch/pytorch/issues/60844,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2022-07-26T14:55:15Z +issue,82084,Untangle TorchScript prim ops in aten namespace,https://github.com/pytorch/pytorch/issues/82084,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-25T03:30:32Z +issue,67570,RFC: Overlap optimizer computation with DDP/FSDP backward,https://github.com/pytorch/pytorch/issues/67570,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2022-07-19T02:45:34Z +issue,81245,"Libtorch cannot load TrochScript Module correctly, when a network contains conv2d(inchannels=64, outchannels=128, kernelsize=1) .",https://github.com/pytorch/pytorch/issues/81245,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-12T16:48:50Z +issue,65305,Add support for HF NLP models in torch.package/deploy,https://github.com/pytorch/pytorch/issues/65305,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-07-11T15:46:27Z +issue,81100,[jit] Failed to load a saved scripted function,https://github.com/pytorch/pytorch/issues/81100,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-08T09:47:59Z +issue,80939,[jit] script backward wrong gradient,https://github.com/pytorch/pytorch/issues/80939,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-06T05:53:23Z +issue,80829,Can torchscript dump backward graph?,https://github.com/pytorch/pytorch/issues/80829,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-06T00:08:47Z +issue,80863,SummaryWriter add_embedding issue with label_img,https://github.com/pytorch/pytorch/issues/80863,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2022-07-05T17:46:36Z +issue,70511,Strange behavior of torch.jit.trace when moving parameters across device,https://github.com/pytorch/pytorch/issues/70511,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-07-04T06:44:43Z +issue,80606,[jit.script] jit.script give uncertain results using torch.half,https://github.com/pytorch/pytorch/issues/80606,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, oncall: jit",categorize and decide next maintainer owner/action,2022-06-30T22:04:37Z +issue,80321,java.lang.ExceptionInInitializerError at org.pytorch.NativePeer.initHybrid(Native Method),https://github.com/pytorch/pytorch/issues/80321,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-06-29T06:25:27Z +issue,80142,CPUProfilingAllocator greedy allocation plan generation failed,https://github.com/pytorch/pytorch/issues/80142,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-06-28T16:10:58Z +issue,61709,[pytorch android] use Vulkan backend crash,https://github.com/pytorch/pytorch/issues/61709,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, module: vulkan, oncall: mobile",categorize and decide next maintainer owner/action,2022-06-22T06:46:29Z +issue,67142,"enumerate(..., start=idx) is not working correctly while scripting",https://github.com/pytorch/pytorch/issues/67142,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-20T16:21:50Z +issue,75502,TensorBoard frontend fails to display embeddings when `add_embedding()` writes large `label_img`,https://github.com/pytorch/pytorch/issues/75502,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2022-06-18T20:12:13Z +issue,79751,[JIT] failures with nested with blocks + loop continuation,https://github.com/pytorch/pytorch/issues/79751,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-16T21:48:21Z +issue,79684,Does Torch JIT Support Trace High-level Custom Op?,https://github.com/pytorch/pytorch/issues/79684,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-16T06:12:47Z +issue,79375,TorchScript bidirectional lnlstm from example doesn't work,https://github.com/pytorch/pytorch/issues/79375,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-14T00:23:31Z +issue,47162,Custon C++ classes are not traceable,https://github.com/pytorch/pytorch/issues/47162,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-13T08:47:22Z +issue,79275,testSerializationInterop in test/cpp/jit/torch_python_test.cpp has not run in over two years,https://github.com/pytorch/pytorch/issues/79275,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-10T13:50:42Z +issue,68871,[LibTorch-Lite] Add a custom flag to build LibTorch-Lite with LAPACK included,https://github.com/pytorch/pytorch/issues/68871,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",categorize and decide next maintainer owner/action,2022-06-08T17:33:22Z +issue,75935,torch.jit.trace error when custom autograd function used in the model,https://github.com/pytorch/pytorch/issues/75935,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-08T12:22:54Z +issue,79008,torchscript jit trace support custom op without specific csrc and .so,https://github.com/pytorch/pytorch/issues/79008,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-07T11:43:03Z +issue,60971,jit scripting for the parametrizations,https://github.com/pytorch/pytorch/issues/60971,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nn.utils.parametrize, oncall: jit",categorize and decide next maintainer owner/action,2022-06-06T17:07:00Z +issue,78920,Strange tracing result with torchscript,https://github.com/pytorch/pytorch/issues/78920,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-06T07:26:48Z +issue,62300,ProcessGroupGloo CUDA Comm Synchronization Looks Wrong,https://github.com/pytorch/pytorch/issues/62300,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2022-06-03T20:47:17Z +issue,77837,TorchScript attempts to compile dead branch of torch.jit.is_scripting,https://github.com/pytorch/pytorch/issues/77837,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-03T19:49:44Z +issue,78812,TorchScript inference get intermediate result?,https://github.com/pytorch/pytorch/issues/78812,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-03T18:36:16Z +issue,78805,"torch.jit.script segmentation fault (pytorch debayer module) 1.10, 1.11 and nightly",https://github.com/pytorch/pytorch/issues/78805,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-03T13:43:08Z +issue,78486,Exception in torch.jit.script doesn't indicate where in the code the problem lies.,https://github.com/pytorch/pytorch/issues/78486,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-06-02T22:39:52Z +issue,78559,[JIT] autodiff implementation of rand_like function is outdated,https://github.com/pytorch/pytorch/issues/78559,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-31T19:09:29Z +issue,72220,[JIT] Assert that the autodiff implementation of backward() returns the correct number of values,https://github.com/pytorch/pytorch/issues/72220,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-31T18:16:36Z +issue,75280,[JIT] log_extract.py improvements,https://github.com/pytorch/pytorch/issues/75280,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-31T17:39:19Z +issue,68987,Operator is not supported in mobile module.,https://github.com/pytorch/pytorch/issues/68987,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-05-29T19:33:22Z +issue,64419,Windows Torch.distributed Multi-GPU training with Gloo backend not working,https://github.com/pytorch/pytorch/issues/64419,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2022-05-28T11:32:36Z +issue,78172,Allow specifying pickle module for torch.package,https://github.com/pytorch/pytorch/issues/78172,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-05-25T18:40:03Z +issue,61221,Unable to add a scalar to an MKLDNN tensor,https://github.com/pytorch/pytorch/issues/61221,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: intel, module: mkldnn, oncall: jit",categorize and decide next maintainer owner/action,2022-05-24T11:36:28Z +issue,78133,Enhancements to AliasDB to handle in-place operations,https://github.com/pytorch/pytorch/issues/78133,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-23T22:45:24Z +issue,50714,[PyTorch Mobile] Android speed benchmark binary crashes when using vulkan,https://github.com/pytorch/pytorch/issues/50714,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-05-23T15:04:37Z +issue,77840,Legacy model format is not supported on mobile,https://github.com/pytorch/pytorch/issues/77840,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-05-23T03:27:04Z +issue,76672,C++ CUDA assign existing memory to forward method.,https://github.com/pytorch/pytorch/issues/76672,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-18T16:45:39Z +issue,77354,"outputs_[i]->uses().empty()INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1646755853042/work/torch/csrc/jit/ir/ir.cpp"":1314, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/77354,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-12T21:21:43Z +issue,73608,`torch.jit.load` fails when function parameters use non-ASCII characters,https://github.com/pytorch/pytorch/issues/73608,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-11T02:38:14Z +issue,67241,"init_process_group hangs for multi-node, Pytorch > v1.3.1 and file init_method",https://github.com/pytorch/pytorch/issues/67241,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: nccl, oncall: distributed",categorize and decide next maintainer owner/action,2022-05-10T13:25:04Z +issue,54133,[TorchScript] kind_.is_prim() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/54133,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-09T15:46:06Z +issue,76578,[JIT][Autocast] Batchnorm folding pass during freezing doesn't preserve types,https://github.com/pytorch/pytorch/issues/76578,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-08T20:29:50Z +issue,77027,Kill use of TensorImpl::ShareExternalPointer in torch/csrc/jit/tensorexpr/external_functions.cpp,https://github.com/pytorch/pytorch/issues/77027,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-07T17:18:48Z +issue,74799,Irrelevant warning during ONNX export of torch.jit.ScriptFunction: Model has no forward function,https://github.com/pytorch/pytorch/issues/74799,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: jit",categorize and decide next maintainer owner/action,2022-05-04T20:00:21Z +issue,76686,[torch::deploy] Remove `manager_` from the constructor and deconstructor of `InterpreterSession`,https://github.com/pytorch/pytorch/issues/76686,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: deploy, oncall: package/deploy",categorize and decide next maintainer owner/action,2022-05-03T20:01:05Z +issue,76683,[torch::deploy] remove reliance on manager_ for unload,https://github.com/pytorch/pytorch/issues/76683,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-05-03T08:48:11Z +issue,76682,[torch::deploy] Remove manager_ from AquireSession,https://github.com/pytorch/pytorch/issues/76682,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-05-03T08:48:05Z +issue,76685,[torch::deploy] move create_movable to interpreter_manager,https://github.com/pytorch/pytorch/issues/76685,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-05-03T08:47:49Z +issue,75956,[JIT] [Autocast] JIT Autocast Pass operations' list should be extendable and consistent with imperative path,https://github.com/pytorch/pytorch/issues/75956,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-05-02T15:24:45Z +issue,76309,More understandable name column of the table of the profiling result.,https://github.com/pytorch/pytorch/issues/76309,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2022-04-30T00:09:36Z +issue,76410,TorchFunction handling and overload resolution very slow in `torch.ops`,https://github.com/pytorch/pytorch/issues/76410,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-27T23:14:42Z +issue,76503,[JIT] make IRAttributeError extend jit::Error,https://github.com/pytorch/pytorch/issues/76503,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-27T23:11:04Z +issue,72330,Add and Mul torch tensors on Metal (IOS),https://github.com/pytorch/pytorch/issues/72330,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",categorize and decide next maintainer owner/action,2022-04-27T18:25:54Z +issue,50943,Exporting opnames fails when using custom classes,https://github.com/pytorch/pytorch/issues/50943,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-04-23T18:06:33Z +issue,75794,Different result with JIT,https://github.com/pytorch/pytorch/issues/75794,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-19T23:23:39Z +issue,75655,`jit(Function)` results in double execution,https://github.com/pytorch/pytorch/issues/75655,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-19T17:43:27Z +issue,76029,Mobile assets upload could break third party mirrors due to binary data size,https://github.com/pytorch/pytorch/issues/76029,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-04-19T17:35:28Z +issue,68330,"isObject()INTERNAL ASSERT FAILED at ""../aten/src/ATen/core/ivalue_inl.h"":115, please report a bug to PyTorch. Expected Object but got Tensor",https://github.com/pytorch/pytorch/issues/68330,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2022-04-16T12:29:45Z +issue,75864,"INTERNAL ASSERT FAILED at ""vulkan_rewrite.cpp"":272",https://github.com/pytorch/pytorch/issues/75864,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-04-15T21:45:32Z +issue,58744,Using pytorch 1.1 model on pytorch 1.8 environment,https://github.com/pytorch/pytorch/issues/58744,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-13T02:00:28Z +issue,75654,jit fails when trying to assign values to model via hook,https://github.com/pytorch/pytorch/issues/75654,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-12T09:17:58Z +issue,72516,test_del (jit.test_builtins.TestBuiltins) fails due to highlight assertions,https://github.com/pytorch/pytorch/issues/72516,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-10T00:15:35Z +issue,75309,[jit] failing OpInfo JIT tests for conv1d and complex input,https://github.com/pytorch/pytorch/issues/75309,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-08T18:30:13Z +issue,75308,Using Adaptive Model for Pytorch Mobile,https://github.com/pytorch/pytorch/issues/75308,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-04-08T04:12:00Z +issue,75467,torch.package: log.info when loading packages w/ externed modules + documentation,https://github.com/pytorch/pytorch/issues/75467,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-04-07T22:17:28Z +issue,75217,Storing LTC tensor shape information in jit::Value,https://github.com/pytorch/pytorch/issues/75217,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-06T18:13:31Z +issue,72112,Unable to parse IR generated by Lazy Tensor Core with TorchScript Backend,https://github.com/pytorch/pytorch/issues/72112,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: lazy, oncall: jit",categorize and decide next maintainer owner/action,2022-04-06T17:25:20Z +issue,75005,Cannot create TorchScript file after converting into a quantized model,https://github.com/pytorch/pytorch/issues/75005,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-06T15:28:38Z +issue,46907,SummaryWriter deletes data automatically when there are too many,https://github.com/pytorch/pytorch/issues/46907,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2022-04-05T16:23:29Z +issue,75038,Generate source location information in TS LTC backend,https://github.com/pytorch/pytorch/issues/75038,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-04-04T00:39:16Z +issue,74964,Add Unit Tests for Torch.Package Subclasses [Waiting on changes to land],https://github.com/pytorch/pytorch/issues/74964,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-04-01T14:57:34Z +issue,74956,Allow `torch.package` to accept `*.pyc` files,https://github.com/pytorch/pytorch/issues/74956,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-04-01T14:57:11Z +issue,75002,JIT autocasting fails on Optional[Tensor],https://github.com/pytorch/pytorch/issues/75002,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-31T21:44:07Z +issue,65813,[RFC] Low-level speed optimizations for PowerSGD,https://github.com/pytorch/pytorch/issues/65813,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2022-03-30T04:15:49Z +issue,74909,torch.package fails to import if dataclass and __future__.annotations present,https://github.com/pytorch/pytorch/issues/74909,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-03-29T21:29:48Z +issue,73972,Forward_AD and Torchscript Functions results in Nones or wrong values.,https://github.com/pytorch/pytorch/issues/73972,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, oncall: jit",categorize and decide next maintainer owner/action,2022-03-29T18:50:49Z +issue,70159,[Android] Unknown builtin op: aten::reflection_pad3d,https://github.com/pytorch/pytorch/issues/70159,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-03-27T22:59:52Z +issue,74605,torch.profiler.profile does't work well for CPU model when not using torch.profiler.schedule,https://github.com/pytorch/pytorch/issues/74605,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2022-03-26T00:06:21Z +issue,72659,Add a unit test which uses an extension module + ordered importer,https://github.com/pytorch/pytorch/issues/72659,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2022-03-23T16:09:09Z +issue,74589,Test test_ops_jit is taking too much time on ASAN,https://github.com/pytorch/pytorch/issues/74589,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-23T00:19:00Z +issue,36718,Support alpha channel in tensorboard.add_figure,https://github.com/pytorch/pytorch/issues/36718,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2022-03-21T17:07:47Z +issue,74272,[PyTorch] Lightweight dispatch,https://github.com/pytorch/pytorch/issues/74272,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-03-16T00:59:01Z +issue,74221,`package.PackageExporter` does not actually appear to have a `file_structure` method,https://github.com/pytorch/pytorch/issues/74221,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: package/deploy",categorize and decide next maintainer owner/action,2022-03-15T15:30:51Z +issue,72110,Conversion Error in pytorch mobile with metal,https://github.com/pytorch/pytorch/issues/72110,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: memory format, oncall: mobile",categorize and decide next maintainer owner/action,2022-03-15T06:28:17Z +issue,62821,Enable fft support for mobile builds,https://github.com/pytorch/pytorch/issues/62821,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, module: arm, module: fft, oncall: mobile",categorize and decide next maintainer owner/action,2022-03-14T11:16:48Z +issue,74095,torchscript RNN modules cannot move between GPU,https://github.com/pytorch/pytorch/issues/74095,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-11T18:53:01Z +issue,64688,Backend capability chart for nccl GPU send/recv may be stale,https://github.com/pytorch/pytorch/issues/64688,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",categorize and decide next maintainer owner/action,2022-03-11T17:01:05Z +issue,71929,[JIT] Results are different when saving in TorchScript Format through train/eval mode,https://github.com/pytorch/pytorch/issues/71929,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-11T07:20:23Z +issue,69434,"JIT RuntimeError: 'Union[Tensor, List[float], List[int]]' object is not subscriptable",https://github.com/pytorch/pytorch/issues/69434,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-10T17:18:48Z +issue,74030,"[JIT] tuple of float is accepted in List[float] but not in Union[float, List[float]]",https://github.com/pytorch/pytorch/issues/74030,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-10T16:17:08Z +issue,41417,[jit] support class polymorphism,https://github.com/pytorch/pytorch/issues/41417,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-09T20:25:25Z +issue,69308,Enable DDP checkpointing tests for Gloo backend,https://github.com/pytorch/pytorch/issues/69308,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2022-03-02T16:54:47Z +issue,73571,conv with padding='same' fails on lazy torchscript nvfuser,https://github.com/pytorch/pytorch/issues/73571,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-01T16:49:10Z +issue,70533,JIT / TorchScript should support for sum(List[torch.Tensor]) like non-JIT PyTorch already does,https://github.com/pytorch/pytorch/issues/70533,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-03-01T10:03:29Z +issue,73419,test_jit_cuda_fuser fails on non-CUDA node for 1.11.0rc3,https://github.com/pytorch/pytorch/issues/73419,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-25T10:43:25Z +issue,70240,JIT: wrong list/tuple length if using Union,https://github.com/pytorch/pytorch/issues/70240,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-19T12:03:23Z +issue,70223,"jit: Confusing behavior w/ torch.autograd.grad, iterative loop, and printing?",https://github.com/pytorch/pytorch/issues/70223,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-19T05:02:48Z +issue,73072,Exporting model to TorchScript fails once I load the saved torchscript file,https://github.com/pytorch/pytorch/issues/73072,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-18T10:13:16Z +issue,72906,Replace deprecated distutils package in torch/utils/tensorboard/__init__.py,https://github.com/pytorch/pytorch/issues/72906,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2022-02-16T23:39:18Z +issue,69347,Memory leak when moving tensors between Metal and CPU on iOS,https://github.com/pytorch/pytorch/issues/69347,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-02-16T14:22:53Z +issue,70099,"Question: what is ""Parameter indices""?",https://github.com/pytorch/pytorch/issues/70099,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2022-02-15T15:02:44Z +issue,70102,torch jit script segm fault,https://github.com/pytorch/pytorch/issues/70102,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-15T12:03:02Z +issue,70037,Question about collect tensor in distributed dataparallel?,https://github.com/pytorch/pytorch/issues/70037,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2022-02-14T10:02:55Z +issue,71912,`torch._sources.normalize_source_lines()` and consequently `torch.jit.script` fail with lambda functions,https://github.com/pytorch/pytorch/issues/71912,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-14T08:04:31Z +issue,69972,torchscript does not work with `SyncBatchNorm` layers,https://github.com/pytorch/pytorch/issues/69972,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-13T22:02:38Z +issue,69830,Conversion error from pytorch model to libtorch model,https://github.com/pytorch/pytorch/issues/69830,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-13T06:03:01Z +issue,72746,CPU execution/dispatch time dominates and slows down small TorchScript GPU models,https://github.com/pytorch/pytorch/issues/72746,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-11T21:35:24Z +issue,69814,JIT / TorchScript docs missing any mention of typing.cast,https://github.com/pytorch/pytorch/issues/69814,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-11T09:02:58Z +issue,72295,[JIT][Autocasting] Add autocasting & constant propagation as part of freezing,https://github.com/pytorch/pytorch/issues/72295,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-10T21:36:02Z +issue,55211,[TorchScript] Can we support type refinement by Python type annotation?,https://github.com/pytorch/pytorch/issues/55211,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-10T19:16:12Z +issue,72617,TorchScript assertion failure for a `List[...]` inside a `NamedTuple`,https://github.com/pytorch/pytorch/issues/72617,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-09T21:06:41Z +issue,72252,Mobile: minimize inferencing latency by having better control over available CPU cores,https://github.com/pytorch/pytorch/issues/72252,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-02-08T23:25:23Z +issue,69591,Make DDP + Zero have the same communication volume as regular DDP,https://github.com/pytorch/pytorch/issues/69591,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2022-02-07T00:06:17Z +issue,69594,RuntimeError: Operation not supported,https://github.com/pytorch/pytorch/issues/69594,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2022-02-07T00:06:16Z +issue,69563,[Distributed Tests] Print stacktraces of all processes when one fails,https://github.com/pytorch/pytorch/issues/69563,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2022-02-05T22:03:29Z +issue,69506,[performance] a profiler util to show a rough break-down of the types of ops used by a model,https://github.com/pytorch/pytorch/issues/69506,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2022-02-05T18:02:46Z +issue,71849,"addcmul, byte_channels_last fail test_nnc_correctness opinfo tests under UBSAN",https://github.com/pytorch/pytorch/issues/71849,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,categorize and decide next maintainer owner/action,2022-02-05T02:26:09Z +issue,72338,[JIT] Channels last optimization pass,https://github.com/pytorch/pytorch/issues/72338,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-04T19:29:43Z +issue,69435,"torch.is_tensor(obj) doesn't work with JIT, despite the fact that isinstance(obj, Tensor) already works with JIT",https://github.com/pytorch/pytorch/issues/69435,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-03T20:02:46Z +issue,69354,Missing instruction in recipe for Andoid/mobile interpreter,https://github.com/pytorch/pytorch/issues/69354,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-02-01T16:03:08Z +issue,72029,JIT function crashing or failing (depending on profiling),https://github.com/pytorch/pytorch/issues/72029,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-02-01T00:39:37Z +issue,69084,Unicode Normalization with Torchscript,https://github.com/pytorch/pytorch/issues/69084,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-29T17:05:16Z +issue,68984,speed_benchmark_torch crashes when trying to run a model using Vulkan backend on Android,https://github.com/pytorch/pytorch/issues/68984,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2022-01-29T02:10:23Z +issue,68940,Breakpoint training in a Tensorboard log,https://github.com/pytorch/pytorch/issues/68940,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2022-01-28T23:02:44Z +issue,68968,torch.jit.trace with pack_padded_sequence cannot do dynamic batch,https://github.com/pytorch/pytorch/issues/68968,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-28T02:13:26Z +issue,71784,[JIT] addmm differs from eager mode,https://github.com/pytorch/pytorch/issues/71784,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-26T05:15:47Z +issue,71286,"[JIT][tensorexpr] cat, batch_norm opinfo tests failing",https://github.com/pytorch/pytorch/issues/71286,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-25T23:26:49Z +issue,65643,[JIT] Start Using and Optimizing on Symbolic Shape Inference in Freezing,https://github.com/pytorch/pytorch/issues/65643,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-24T21:23:24Z +issue,68800,[1.10 regression][jit][cuda fuse] cat codegen wrong var name,https://github.com/pytorch/pytorch/issues/68800,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-23T04:02:53Z +issue,68759,`test_variant_consistency_jit` errors for `torch.nn.functional.prelu`,https://github.com/pytorch/pytorch/issues/68759,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-21T22:03:01Z +issue,71548,Support freezing of models containing ModuleContainerIndex,https://github.com/pytorch/pytorch/issues/71548,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-20T17:16:40Z +issue,59905,[JIT] batch operators in training/inference,https://github.com/pytorch/pytorch/issues/59905,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-20T03:01:00Z +issue,54612,Exceptions thrown in JIT interpreter always are translated into RuntimeError,https://github.com/pytorch/pytorch/issues/54612,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-19T21:08:06Z +issue,71470,torch.jit.script failed to compile nn.MultiheadAttention when specifying the kdim and vdim parameters.,https://github.com/pytorch/pytorch/issues/71470,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-19T11:26:47Z +issue,65983,"traced EmbeddingBag cannot be used with different batch size, functional.embedding_bag is not de",https://github.com/pytorch/pytorch/issues/65983,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-18T16:25:59Z +issue,71151,The implement of `containsTensorType(const TypePtr& t)` in jit pass `PropagateInputShapes` ignore some situation?,https://github.com/pytorch/pytorch/issues/71151,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-17T02:56:02Z +issue,71165,traced module fails on second execution,https://github.com/pytorch/pytorch/issues/71165,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-15T04:11:07Z +issue,71288,`memory_format` argument isn't supported in torchscript for tensor.is_contiguous(),https://github.com/pytorch/pytorch/issues/71288,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-13T23:51:07Z +issue,70951,A diagnostics mode to report when constraints aren't being met for optimal performance,https://github.com/pytorch/pytorch/issues/70951,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,categorize and decide next maintainer owner/action,2022-01-07T04:42:29Z +issue,52271,torchscript should work with type: ignore comments from mypy even if they are not at the end of the line,https://github.com/pytorch/pytorch/issues/52271,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-06T13:09:46Z +issue,68785,[elastic/distributed] API to retrieve consolidated worker information,https://github.com/pytorch/pytorch/issues/68785,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: r2p",categorize and decide next maintainer owner/action,2022-01-06T09:01:19Z +issue,70267,Inconsistent multi-node latency with NCCL and OpenMPI,https://github.com/pytorch/pytorch/issues/70267,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: nccl, module: openmp, oncall: distributed",categorize and decide next maintainer owner/action,2022-01-05T23:50:06Z +issue,70155,RuntimeError: Tensor must be CUDA and dense when calling all_gather_object even though there is no tensor in the object.,https://github.com/pytorch/pytorch/issues/70155,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2022-01-05T23:36:03Z +issue,70627,Torchscript Compiled functions don't support keyword-only arguments with defaults,https://github.com/pytorch/pytorch/issues/70627,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-04T19:45:53Z +issue,70573,JIT magic method which returns class instance fails.,https://github.com/pytorch/pytorch/issues/70573,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2022-01-03T09:52:35Z +issue,41492,Make torch.iinfo/torch.finfo torchscriptable,https://github.com/pytorch/pytorch/issues/41492,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-12-31T13:58:43Z +issue,65490,optimize_for_mobile() assert fail: missing op for prepacked::conv2d_clamp_prepack,https://github.com/pytorch/pytorch/issues/65490,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-12-29T16:58:16Z +issue,41353,[jit] support for torch.distributed primitives,https://github.com/pytorch/pytorch/issues/41353,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-12-24T06:43:40Z +issue,41416,[jit] support for generators and `yield`,https://github.com/pytorch/pytorch/issues/41416,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-12-23T15:31:54Z +issue,51673,[JIT/Futures] Future callbacks/then() APIs should throw the correct exception,https://github.com/pytorch/pytorch/issues/51673,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2021-12-23T01:19:29Z +issue,69608,Optimize For Inference - add warning/error upon serialization,https://github.com/pytorch/pytorch/issues/69608,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-12-22T20:42:45Z +issue,70054,[c10d] have a way to determine two ranks on the same hosts or not,https://github.com/pytorch/pytorch/issues/70054,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2021-12-21T20:10:38Z +issue,69938,"JIT is overriding variables of type List[int] to type Tuple[int, int] for an unknown reason during function calls",https://github.com/pytorch/pytorch/issues/69938,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-12-20T21:30:09Z +issue,69320,[JIT] Break up JIT Opinfo Tests,https://github.com/pytorch/pytorch/issues/69320,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-12-16T17:18:27Z +issue,69522,PyTorch 1.9.1 incorrect result for all-reduce,https://github.com/pytorch/pytorch/issues/69522,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-12-16T02:28:13Z +issue,52214,`hasattr` check not supported on objects other than `self`,https://github.com/pytorch/pytorch/issues/52214,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-12-14T16:08:19Z +issue,49782,TorchScript model's output different on Mobile and PC,https://github.com/pytorch/pytorch/issues/49782,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-12-05T13:49:13Z +issue,68917,"Bugs related to NCCL on RTX 6000, code freezes with no output when using DistributedDataParallel",https://github.com/pytorch/pytorch/issues/68917,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2021-11-30T18:20:08Z +issue,68862,find_unused_parameters of DDP,https://github.com/pytorch/pytorch/issues/68862,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2021-11-30T18:06:38Z +issue,68927,Thank you for the diagnostic script.,https://github.com/pytorch/pytorch/issues/68927,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2021-11-29T23:50:23Z +issue,63395,How to efficiently (without looping) get data from tensor predicted by a torchscript in C++?,https://github.com/pytorch/pytorch/issues/63395,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-28T03:54:22Z +issue,68536,Exporting a model with `_save_for_lite_interpreter()` fails with Segmentation fault,https://github.com/pytorch/pytorch/issues/68536,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-11-24T12:53:27Z +issue,56595,"TorchScript breaks (""internal assert failed"") on `torch.set_grad_enabled`",https://github.com/pytorch/pytorch/issues/56595,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-20T20:20:56Z +issue,57457,Valgrind Shows Potential Memory Leaks,https://github.com/pytorch/pytorch/issues/57457,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: jit",categorize and decide next maintainer owner/action,2021-11-18T15:40:22Z +issue,43237,[jit] remove shape/device specialization from tracing after ONNX done with its own shape inference,https://github.com/pytorch/pytorch/issues/43237,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-16T23:27:16Z +issue,68049,Request to revise the Pytorch tutorial.,https://github.com/pytorch/pytorch/issues/68049,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs, oncall: jit",categorize and decide next maintainer owner/action,2021-11-16T19:06:05Z +issue,17932,JIT torch.ones_like with dtype starts failing on master,https://github.com/pytorch/pytorch/issues/17932,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-14T21:51:32Z +issue,14455,[JIT] Tracing a script function/module where not all args are Tensors,https://github.com/pytorch/pytorch/issues/14455,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-14T21:51:30Z +issue,68297,[DDP] Debug mode should ensure reduction is finished in backward pass,https://github.com/pytorch/pytorch/issues/68297,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2021-11-12T23:18:06Z +issue,68104,DDP is not working with Pytorch Lightning,https://github.com/pytorch/pytorch/issues/68104,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-11-11T18:56:46Z +issue,68088,[distributed] Restructure our distributed collective tests,https://github.com/pytorch/pytorch/issues/68088,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2021-11-09T23:02:27Z +issue,68065,Incorrect return type annotation for _get_mobile_model_contained_types,https://github.com/pytorch/pytorch/issues/68065,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-09T16:43:56Z +issue,68064,In place `or` operator doesn't work in torchscript,https://github.com/pytorch/pytorch/issues/68064,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-09T16:25:43Z +issue,59551,"error when trying to call ""torch::jit::load""method, use Metal backend in PyTorch Mobile",https://github.com/pytorch/pytorch/issues/59551,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",categorize and decide next maintainer owner/action,2021-11-08T23:59:23Z +issue,61268,Unknown type name '__torch__.torch.classes.metal.Conv2dOpContext',https://github.com/pytorch/pytorch/issues/61268,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: ios, module: macos, oncall: mobile",categorize and decide next maintainer owner/action,2021-11-08T23:36:20Z +issue,67603,The minimum supported version of pytorch when using toch-mobile？,https://github.com/pytorch/pytorch/issues/67603,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-11-08T23:11:10Z +issue,60518,different gpus to train,https://github.com/pytorch/pytorch/issues/60518,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-11-08T02:57:52Z +issue,53451,Model loaded from full model state checkpoint does not run on multiple GPUs (device mismatch),https://github.com/pytorch/pytorch/issues/53451,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn, module: serialization, oncall: distributed",categorize and decide next maintainer owner/action,2021-11-05T14:23:32Z +issue,67889,[libtorch][torch.jit.script][input dynamic shape]Error in loading “. Pt” file on C + +: index out of range,https://github.com/pytorch/pytorch/issues/67889,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-05T02:31:42Z +issue,67838,`test_variant_consistency_jit_contiguous_cpu_float32` from `test_ops` fails,https://github.com/pytorch/pytorch/issues/67838,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-04T11:14:35Z +issue,54821,Decouple TorchScript compiler from operator library,https://github.com/pytorch/pytorch/issues/54821,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-03T18:07:00Z +issue,65814,INTERNAL ASSERT FAILED when reusing JIT compiled module,https://github.com/pytorch/pytorch/issues/65814,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-03T13:56:40Z +issue,67599,"Mobile builds need to make custom build of libtorch to run, let's just include it to make it easier",https://github.com/pytorch/pytorch/issues/67599,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-11-02T11:40:24Z +issue,67659,`test_variant_consistency_jit` failing for `max_unpool2d`,https://github.com/pytorch/pytorch/issues/67659,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-02T03:31:51Z +issue,58456,TSAN issues in torch::jit::ProfilingRecord,https://github.com/pytorch/pytorch/issues/58456,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-11-02T02:28:00Z +issue,65058,Add nnapi serialization for module components in Wav2Vec2Model,https://github.com/pytorch/pytorch/issues/65058,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-30T14:36:08Z +issue,65662,mobile deployment: optimized_for_mobile---wrap_cpp_module error,https://github.com/pytorch/pytorch/issues/65662,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-10-29T15:02:54Z +issue,67542,Segfault from ONNX exporting code where torch.jit.script and torch.no_grad are used together,https://github.com/pytorch/pytorch/issues/67542,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-29T10:42:06Z +issue,67473,"[Feature] Separate, Queryable Alias Parser",https://github.com/pytorch/pytorch/issues/67473,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-29T03:42:24Z +issue,65734,pytorch mobile: some quant models on android cause memory leak,https://github.com/pytorch/pytorch/issues/65734,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-28T22:25:44Z +issue,66545,Forward method slows down at some point if it is executed repeatedly on Android with NNAPI,https://github.com/pytorch/pytorch/issues/66545,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-28T22:24:21Z +issue,67474,[JIT][Feature] Runtime Verifier of Alias Annotations,https://github.com/pytorch/pytorch/issues/67474,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-28T20:51:55Z +issue,67458,OpInfo JIT tests fail for torch.nonzero,https://github.com/pytorch/pytorch/issues/67458,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-28T18:48:38Z +issue,67420,Avoid warnings when jitting pytorch modules,https://github.com/pytorch/pytorch/issues/67420,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-28T05:21:36Z +issue,67392,Missing doc for torch.jit functions,https://github.com/pytorch/pytorch/issues/67392,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: jit",categorize and decide next maintainer owner/action,2021-10-27T22:44:58Z +issue,66816,Jit fuser half precision support,https://github.com/pytorch/pytorch/issues/66816,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-27T19:42:09Z +issue,67160,"Include function name in ""This Python function is annotated to be ignored and cannot be run"" error message",https://github.com/pytorch/pytorch/issues/67160,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-26T15:34:22Z +issue,65824,Failing to compile /torch/csrc/jit/frontend/ir_emitter.cpp with pointer is null warning,https://github.com/pytorch/pytorch/issues/65824,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-25T23:14:39Z +issue,56703,[Pytorch Mobile] Error running build_pytorch_android.sh,https://github.com/pytorch/pytorch/issues/56703,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:30:23Z +issue,57271,RuntimeError after optimize_for_mobile,https://github.com/pytorch/pytorch/issues/57271,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:30:12Z +issue,60172,[Android] Upgrading to 1.9.0 causes NNAPI model loading to fail,https://github.com/pytorch/pytorch/issues/60172,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:27:27Z +issue,61693,"Pytorch Mobile on Android, LAPACK library not found in compilation",https://github.com/pytorch/pytorch/issues/61693,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:26:16Z +issue,61951,Vulkan: error -5 when trying to use it with Nvidia card,https://github.com/pytorch/pytorch/issues/61951,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, module: vulkan, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:24:50Z +issue,62165,"RuntimeError: xnn_status_success == run_status INTERNAL ASSERT FAILED at ""/pytorch/aten/src/ATen/native/xnnpack/Linear.cpp"":158",https://github.com/pytorch/pytorch/issues/62165,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, module: ios, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:24:25Z +issue,62676,pytorch_mobile custom build Module.forward null reference.,https://github.com/pytorch/pytorch/issues/62676,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:23:49Z +issue,64333,When compiling TAG 1.9.0 or master for android，An error occurred and the compilation failed！ABIS_LIST=arm64-v8a,https://github.com/pytorch/pytorch/issues/64333,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:21:35Z +issue,66768,Error: `copy_to_metal_ is implemented only for float dtype`,https://github.com/pytorch/pytorch/issues/66768,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",categorize and decide next maintainer owner/action,2021-10-25T22:17:59Z +issue,66627,Overhaul error handling in `TCPStore`,https://github.com/pytorch/pytorch/issues/66627,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-10-25T18:03:52Z +issue,67111,[doc] Can we update TorchScript union support?,https://github.com/pytorch/pytorch/issues/67111,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: jit",categorize and decide next maintainer owner/action,2021-10-24T19:58:49Z +issue,67099,JIT: inconsistency in LSTM.forward,https://github.com/pytorch/pytorch/issues/67099,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-22T16:38:50Z +issue,51941,RuntimeError: torch.dtype passed as int from JIT to ignored function,https://github.com/pytorch/pytorch/issues/51941,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-19T19:38:37Z +issue,66841,"[Torchscript] Comparison between list and int returns true [1] == (1, ) == 1",https://github.com/pytorch/pytorch/issues/66841,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-19T02:17:21Z +issue,51154,[JIT] Rough edges found while exploring expressions supported by TorchScript,https://github.com/pytorch/pytorch/issues/51154,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-18T18:01:28Z +issue,66791,jit tracer：Lost imported pre-model parameters,https://github.com/pytorch/pytorch/issues/66791,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-18T14:42:34Z +issue,51892,[cuda jit extentions] global shared destination conflicts with virtual envs,https://github.com/pytorch/pytorch/issues/51892,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-14T20:48:07Z +issue,57929,[JIT] provide a pass for alias reduction,https://github.com/pytorch/pytorch/issues/57929,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-14T20:25:10Z +issue,66651,[JIT] Profile optional tensor,https://github.com/pytorch/pytorch/issues/66651,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-14T20:23:51Z +issue,66624,Separate libtorch and non-libtorch specific files under `torch/csrc/distributed`,https://github.com/pytorch/pytorch/issues/66624,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",categorize and decide next maintainer owner/action,2021-10-14T14:58:41Z +issue,65907,Inplace error of BatchNorm layer in DistributedDataParallel module,https://github.com/pytorch/pytorch/issues/65907,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-10-12T06:09:27Z +issue,66150,JIT: empty_like doesn't support requires_grad,https://github.com/pytorch/pytorch/issues/66150,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-05T18:47:38Z +issue,66094,[JIT][Sym Shape Analysis] Extend analysis to extract output shape logic,https://github.com/pytorch/pytorch/issues/66094,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2021-10-05T01:03:21Z +issue,66065,torch.jit.script fails to cast explicit Optional parameter to bool,https://github.com/pytorch/pytorch/issues/66065,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-10-04T06:41:31Z +issue,65985,torch.cuda.power_usage (ESG 🌱),https://github.com/pytorch/pytorch/issues/65985,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: profiler",categorize and decide next maintainer owner/action,2021-10-01T17:48:03Z +issue,65918,C++ extensions don't correctly set USE_C10D_FOO preprocessor macros,https://github.com/pytorch/pytorch/issues/65918,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-09-30T12:29:28Z +issue,65915,"I am getting undefined symbol: _ZN5torch3jit17parseSchemaOrNameERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE error. This I am getting when I am trying to ""import torch from nemo.collections import nlp"". I am trying to use pytorch ngc container 21.05. I tried to import torch before the nemo extension. Please suggest how I can resolve this.",https://github.com/pytorch/pytorch/issues/65915,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-30T12:17:54Z +issue,63610,Enable some amount of CI tests on lazy_tensor_staging branch,https://github.com/pytorch/pytorch/issues/63610,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: lazy, oncall: jit",categorize and decide next maintainer owner/action,2021-09-29T06:21:09Z +issue,65776,`torch.jit.save` fails on Windows,https://github.com/pytorch/pytorch/issues/65776,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-28T20:42:09Z +issue,65630,Constants in torch.Jit.script,https://github.com/pytorch/pytorch/issues/65630,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-24T20:50:23Z +issue,65588,Make torch.utils.tensorboard.add_scalar accepts float global_step,https://github.com/pytorch/pytorch/issues/65588,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2021-09-24T15:39:49Z +issue,65611,"`torch.jit.trace` fails claiming ""forward method already defined""",https://github.com/pytorch/pytorch/issues/65611,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-24T14:33:06Z +issue,64462,torch.package: re-export docs suggestions don't work + other sharp edges,https://github.com/pytorch/pytorch/issues/64462,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2021-09-23T23:45:09Z +issue,64986,RandomSampler / DistributedSampler does not seem really random,https://github.com/pytorch/pytorch/issues/64986,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",categorize and decide next maintainer owner/action,2021-09-23T20:30:16Z +issue,65397,Vague linkage when using libcxxabi built torch in torch_glow,https://github.com/pytorch/pytorch/issues/65397,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-23T03:19:40Z +issue,65449,Make SummaryWriter.add_image supports PIL image directly,https://github.com/pytorch/pytorch/issues/65449,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2021-09-22T14:55:28Z +issue,65398,[JIT] JIT does not support non-persistent buffers,https://github.com/pytorch/pytorch/issues/65398,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-21T11:56:03Z +issue,64795,profiler crashes the program when trying to stop on different threads without context manager,https://github.com/pytorch/pytorch/issues/64795,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: multithreading, oncall: distributed, oncall: profiler",categorize and decide next maintainer owner/action,2021-09-20T18:01:05Z +issue,47565,Keyword-only function not allowed with TorchScript,https://github.com/pytorch/pytorch/issues/47565,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2021-09-20T16:27:03Z +issue,65297,[JIT] TorchScript not works as expected when ModuleDict has modules with different signatures,https://github.com/pytorch/pytorch/issues/65297,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-18T10:48:30Z +issue,65098,Bug: torch.jit.load cannot load from IOBuffer,https://github.com/pytorch/pytorch/issues/65098,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-17T13:58:29Z +issue,65052,Use hyphens to separate long CLI parameters instead of underscores,https://github.com/pytorch/pytorch/issues/65052,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-09-17T07:39:22Z +issue,65144,jit.trace's returned module returns a tuple instead of a named tuple,https://github.com/pytorch/pytorch/issues/65144,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-16T15:04:47Z +issue,63445,"getNcclVersion() in NCCLUtils.cpp handles ""2.10.3"" incorrectly",https://github.com/pytorch/pytorch/issues/63445,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: nccl, oncall: distributed",categorize and decide next maintainer owner/action,2021-09-14T23:11:36Z +issue,64799,Add support to profile all RPC worker threads that are run on a process,https://github.com/pytorch/pytorch/issues/64799,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-09-14T21:51:07Z +issue,44077,[jit] Better type refinement for class attributes which are class types,https://github.com/pytorch/pytorch/issues/44077,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-14T17:31:56Z +issue,18520,Jit fail with TracingCheckError with tracing model with layers created after init.,https://github.com/pytorch/pytorch/issues/18520,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-14T17:17:53Z +issue,49643,Tracing doesn't work with spectral normalization,https://github.com/pytorch/pytorch/issues/49643,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-14T17:14:01Z +issue,64896,pytorch1.9.0 can not use torch.jit.load() to load model which save in pytorch1.3,https://github.com/pytorch/pytorch/issues/64896,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-14T16:05:57Z +issue,64905,[docs] [feature request] Tutorial on implementing correctly distributed dataset evaluators,https://github.com/pytorch/pytorch/issues/64905,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-09-14T06:28:59Z +issue,64957,`INTERNAL ASSERT FAILED` after `torch.jit.freeze`,https://github.com/pytorch/pytorch/issues/64957,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-13T22:32:50Z +issue,64918,TorchScript error loading module with float literal default argument,https://github.com/pytorch/pytorch/issues/64918,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-13T10:36:55Z +issue,64488,It' so strange!!!,https://github.com/pytorch/pytorch/issues/64488,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, module: serialization, oncall: mobile",categorize and decide next maintainer owner/action,2021-09-10T20:31:32Z +issue,64775,[JIT] Canonicalize aten::rsub,https://github.com/pytorch/pytorch/issues/64775,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-09T20:03:58Z +issue,64490,Load torchscript model with custom operators (GNN) on Windows,https://github.com/pytorch/pytorch/issues/64490,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-09T09:41:08Z +issue,34713,torch.jit.script report error when using index to subscript nn.ModuleList,https://github.com/pytorch/pytorch/issues/34713,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-09T07:22:39Z +issue,60143,"torch.jit.save gives error - RuntimeError: Could not export Python function call 'NMSop'. Remove calls to Python functions before export. Did you forget to add @script or @script_method annotation? If this is a nn.ModuleList, add it to __constants__",https://github.com/pytorch/pytorch/issues/60143,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-09T03:06:59Z +issue,64150,Enhance DDP doc to have a complete example including data loading,https://github.com/pytorch/pytorch/issues/64150,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-09-08T15:26:06Z +issue,64628,[jit] cannot compile staticmethod of classes when they have a unscriptable __init__,https://github.com/pytorch/pytorch/issues/64628,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-08T06:46:32Z +issue,64487,Error in torchScript based on Pytorch tutorial,https://github.com/pytorch/pytorch/issues/64487,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-03T19:27:03Z +issue,63740,DDP Gradient reduction is not triggered after calling autograd.backward(),https://github.com/pytorch/pytorch/issues/63740,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-09-03T18:04:14Z +issue,57893,Android API to run model on Vulkan backend,https://github.com/pytorch/pytorch/issues/57893,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-09-02T07:33:22Z +issue,64130,Every process takes a tiny memory on each nvidia GPU card when using DDP.,https://github.com/pytorch/pytorch/issues/64130,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-09-01T18:57:44Z +issue,64258,Incubation of Graphical Representation in TF2 of HParams,https://github.com/pytorch/pytorch/issues/64258,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2021-09-01T18:54:24Z +issue,64334,How to add nan value judgment for variable t0_1 in fused_clamp kernel generated by torch/csrc/jit/tensorexpr/cuda_codegen.cpp.,https://github.com/pytorch/pytorch/issues/64334,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-09-01T02:53:24Z +issue,64260,Using Cloudpickle for pickling jitted functions,https://github.com/pytorch/pytorch/issues/64260,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-31T11:33:46Z +issue,63648,[JIT] Typing on math.ceil is inaccurate for Scalar input,https://github.com/pytorch/pytorch/issues/63648,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-31T11:13:09Z +issue,64247,How to optimize jit-script model performance (backend device is gpu),https://github.com/pytorch/pytorch/issues/64247,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-31T05:31:46Z +issue,64005,[LTC] Find a way to convert at::Generator into torch::jit::NamedValue,https://github.com/pytorch/pytorch/issues/64005,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: lazy, oncall: jit",categorize and decide next maintainer owner/action,2021-08-30T17:00:16Z +issue,58507,OpInfo JIT tests do not work with Tensor kwarg arguments,https://github.com/pytorch/pytorch/issues/58507,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: jit",categorize and decide next maintainer owner/action,2021-08-30T02:55:37Z +issue,42847,Get test_jit.py below 10k lines,https://github.com/pytorch/pytorch/issues/42847,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: jit",categorize and decide next maintainer owner/action,2021-08-30T02:44:31Z +issue,64119,Cleanup tests in rpc_test.py,https://github.com/pytorch/pytorch/issues/64119,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-30T02:39:22Z +issue,64096,String frontend doesn't support tuple unpacking,https://github.com/pytorch/pytorch/issues/64096,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-27T16:05:19Z +issue,64060,Consolidating NCCL version parsing functions,https://github.com/pytorch/pytorch/issues/64060,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-27T14:37:39Z +issue,62885,Type of first constructor parameter of ElasticDistributedSampler and DistributedSampler,https://github.com/pytorch/pytorch/issues/62885,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-27T09:37:35Z +issue,53421,[bug][JIT] Error when tracing model with partial hooks in PyTorch v1.8.0,https://github.com/pytorch/pytorch/issues/53421,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-26T16:09:53Z +issue,55749,Fail to introduce torch::jit::Module as a parameter of a cusomized operator.,https://github.com/pytorch/pytorch/issues/55749,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-26T16:09:48Z +issue,60089,torch.package: TypeError: 'NoneType' object is not iterable,https://github.com/pytorch/pytorch/issues/60089,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2021-08-26T16:08:29Z +issue,62542,3 Times memory cost when loading the model to torch.nn.parallel.DistributedDataParallel,https://github.com/pytorch/pytorch/issues/62542,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-25T20:35:09Z +issue,63832,[RFC] Add `torch.distributed.run` as a console script in pytorch's setup.py,https://github.com/pytorch/pytorch/issues/63832,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-25T19:44:57Z +issue,63455,distributed_error,https://github.com/pytorch/pytorch/issues/63455,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-08-25T17:09:30Z +issue,63405,torch.arange has issue tracing with bool input,https://github.com/pytorch/pytorch/issues/63405,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-20T17:05:37Z +issue,63403,Consolidate distributed benchmark folders,https://github.com/pytorch/pytorch/issues/63403,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-17T21:13:59Z +issue,62325,[jit]][script] handling of 'void' returns,https://github.com/pytorch/pytorch/issues/62325,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-17T14:37:05Z +issue,62775,[TorchScript] Support for registering builtin operators with control flow,https://github.com/pytorch/pytorch/issues/62775,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-16T19:10:12Z +issue,63247,Add a timeout argument to RPC shutdown(),https://github.com/pytorch/pytorch/issues/63247,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-13T18:50:29Z +issue,63124,torch.jit.trace quantized bigbird leads to 0INTERNAL ASSERT FAILED runtime error,https://github.com/pytorch/pytorch/issues/63124,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-13T17:17:06Z +issue,57559,[JIT] Investigate cloning/copying object before tracing it,https://github.com/pytorch/pytorch/issues/57559,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-12T21:08:39Z +issue,63171,[BE] Clean up DDP Single-Program Multi-Device Vestige in reducer.cpp,https://github.com/pytorch/pytorch/issues/63171,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-12T19:15:01Z +issue,63145,torch.jit.trace does not work if there is an autograd in the function,https://github.com/pytorch/pytorch/issues/63145,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-12T11:50:15Z +issue,54937,[JIT] export fast-transformer to TorchScript,https://github.com/pytorch/pytorch/issues/54937,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-12T06:22:51Z +issue,55742,[JIT] TorchScript to represent typed tensor annotations in IR graph,https://github.com/pytorch/pytorch/issues/55742,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-12T06:10:33Z +issue,62154,AttributeError: module 'torch.jit' has no attribute '_script_if_tracing',https://github.com/pytorch/pytorch/issues/62154,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-12T05:34:30Z +issue,62394,crash in libtorch jit,https://github.com/pytorch/pytorch/issues/62394,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-12T05:30:57Z +issue,54593,Multiple refinement with `isinstance` doesn't work,https://github.com/pytorch/pytorch/issues/54593,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-11T14:38:11Z +issue,42592,Lower performance when forwarding JIT model with libTorch 1.6 compared with 1.5,https://github.com/pytorch/pytorch/issues/42592,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-11T06:22:53Z +issue,53744,TorchScript divison by zero fails to error out,https://github.com/pytorch/pytorch/issues/53744,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-11T05:55:58Z +issue,56660,consistency jit tests gives error for gradient operator,https://github.com/pytorch/pytorch/issues/56660,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-11T05:14:48Z +issue,60953,Add support for Python's optimized mode to torchscript,https://github.com/pytorch/pytorch/issues/60953,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-11T04:55:34Z +issue,62854,"aten::pixel_shuffle(Tensor self, int upscale_factor) -> (Tensor): Argument self not provided.",https://github.com/pytorch/pytorch/issues/62854,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-05T23:28:27Z +issue,62283,c10d and discontiguous tensors with mismatch strides,https://github.com/pytorch/pytorch/issues/62283,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-04T13:51:08Z +issue,62314,dist.destroy_process_group() don't release master_port when there're at least 2 process groups,https://github.com/pytorch/pytorch/issues/62314,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-04T13:49:12Z +issue,62403,DDP,https://github.com/pytorch/pytorch/issues/62403,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2021-08-04T13:32:43Z +issue,8473,RuntimeError: /pytorch/torch/csrc/jit/tracer.h:117: getTracingState: Assertion `var_state == state` failed.,https://github.com/pytorch/pytorch/issues/8473,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-04T06:27:58Z +issue,62543,FAIL: test_threshold (test_jit_fuser.TestFuser),https://github.com/pytorch/pytorch/issues/62543,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-03T06:18:48Z +issue,62603,Unexpected `is_training` after using `train` in a JIT Module,https://github.com/pytorch/pytorch/issues/62603,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-02T20:29:08Z +issue,62556,[TorchScript] Inconsistency between Python and PyTorch when raising 0 to a negative power,https://github.com/pytorch/pytorch/issues/62556,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-08-02T05:40:55Z +issue,54728,JIT prim ops v.s. dispatcher,https://github.com/pytorch/pytorch/issues/54728,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: dispatch, oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2021-08-02T05:26:32Z +issue,54982,[RFC] A PyTorch Tensor Shape DSL For Symbolic Shape Inference,https://github.com/pytorch/pytorch/issues/54982,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-07-27T20:55:34Z +issue,62007,Add Future return from default process group collective communications,https://github.com/pytorch/pytorch/issues/62007,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2021-07-23T02:28:50Z +issue,61765,How to save tensors on mobile (lite interpreter)?,https://github.com/pytorch/pytorch/issues/61765,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-07-21T10:06:17Z +issue,51574,[Pipe] conversion to Sequential: various complex scenarios,https://github.com/pytorch/pytorch/issues/51574,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-07-13T19:16:46Z +issue,50734,[TorchScript] Document TorchScript IR,https://github.com/pytorch/pytorch/issues/50734,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-07-13T15:24:02Z +issue,61267,Incorrect trace with MKLDNN (adding scalar),https://github.com/pytorch/pytorch/issues/61267,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, oncall: jit",categorize and decide next maintainer owner/action,2021-07-12T20:07:38Z +issue,60847,How to release CPU memory cache in Libtorch JIT ?,https://github.com/pytorch/pytorch/issues/60847,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-07-07T03:01:06Z +issue,57894,"JIT ""optimizations"" take way too long",https://github.com/pytorch/pytorch/issues/57894,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-07-07T00:50:10Z +issue,60640,Building fails on a node with libibverbs installed with no Infiniband hardware present on RHEL8.,https://github.com/pytorch/pytorch/issues/60640,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorpipe, oncall: distributed",categorize and decide next maintainer owner/action,2021-07-02T07:51:14Z +issue,60440,DDP with cuda rpc failed with DistributedOptimizer Adagrad,https://github.com/pytorch/pytorch/issues/60440,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: distributed",categorize and decide next maintainer owner/action,2021-07-01T22:50:11Z +issue,56969,[JIT / NNC] Incomplete Ops Coverage,https://github.com/pytorch/pytorch/issues/56969,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-29T21:41:46Z +issue,60732,Potential memory leaks reported by static analysis tool,https://github.com/pytorch/pytorch/issues/60732,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-28T08:12:03Z +issue,59865,Custom build of c++ libtorch for Android program or dynamic so lib does not reduce program or lib size,https://github.com/pytorch/pytorch/issues/59865,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-06-25T11:47:24Z +issue,60359,Expose the .pt model file path during TorchScript custom class Serialization/Deserialization,https://github.com/pytorch/pytorch/issues/60359,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-22T01:40:34Z +issue,60189,[package] Tutorial not accessible,https://github.com/pytorch/pytorch/issues/60189,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,categorize and decide next maintainer owner/action,2021-06-19T10:02:03Z +issue,60062,[c10d] Make `broadcast_object_list` accept a device parameter,https://github.com/pytorch/pytorch/issues/60062,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",categorize and decide next maintainer owner/action,2021-06-16T01:42:44Z +issue,59441,distributed/test_jit_c10d.py fails with RuntimeError: Address already in use,https://github.com/pytorch/pytorch/issues/59441,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-06-14T23:54:07Z +issue,58907,"JIT fails when calling an exported method inside another one, if kwargs with defaults are not passed",https://github.com/pytorch/pytorch/issues/58907,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-14T10:05:06Z +issue,59823,print_regression should handle sharded test properly,https://github.com/pytorch/pytorch/issues/59823,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-06-11T21:58:47Z +issue,59806,Context manager for torch.jit.ignore is not covered by test coverage,https://github.com/pytorch/pytorch/issues/59806,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-10T18:28:57Z +issue,59652,pin_memory mutates its input but this is not reflected in JIT schema,https://github.com/pytorch/pytorch/issues/59652,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-08T19:57:51Z +issue,59552,test_batch_isend_irecv_nccl fails with NCCL 2.8.3,https://github.com/pytorch/pytorch/issues/59552,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-06-07T20:20:10Z +issue,59153,JIT compiling a ParameterList raises an unnecessary warning,https://github.com/pytorch/pytorch/issues/59153,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-07T19:04:15Z +issue,59528,AttributeError: module 'torch.jit' has no attribute '_script_if_tracing',https://github.com/pytorch/pytorch/issues/59528,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-07T16:56:06Z +issue,59114,[JIT] OpInfo / Autodiff tests dont seem to actually invoke symbolic backwards,https://github.com/pytorch/pytorch/issues/59114,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-06T17:27:38Z +issue,59257,Optimizer step stuck during distributed training,https://github.com/pytorch/pytorch/issues/59257,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: data parallel, oncall: distributed",categorize and decide next maintainer owner/action,2021-06-04T21:18:02Z +issue,35372,How to support single-process-multiple-devices in DistributedDataParallel other than CUDA device,https://github.com/pytorch/pytorch/issues/35372,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-06-04T13:57:34Z +issue,37442,Tensor.is_distributed not documented,https://github.com/pytorch/pytorch/issues/37442,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: distributed",categorize and decide next maintainer owner/action,2021-06-03T02:07:06Z +issue,57716,Unknown builtin op: torchvision::nms in LibTorch,https://github.com/pytorch/pytorch/issues/57716,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: serialization, module: vision, oncall: jit",categorize and decide next maintainer owner/action,2021-06-02T00:14:23Z +issue,59174,[jit] Error: Results of original model and exported/imported version of model differed,https://github.com/pytorch/pytorch/issues/59174,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-02T00:12:28Z +issue,42267,assertRaisesRegxWithHighlight isn't triggered correctly,https://github.com/pytorch/pytorch/issues/42267,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-06-01T21:50:51Z +issue,41242,"test_backward_deadlock fails with ""Directory not empty""",https://github.com/pytorch/pytorch/issues/41242,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: power, oncall: jit",categorize and decide next maintainer owner/action,2021-06-01T16:46:34Z +issue,58982,Can't re-init rpc with a different rank after rpc shutdown,https://github.com/pytorch/pytorch/issues/58982,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-06-01T06:24:09Z +issue,37387,The pytorch's graph is lack of common names for nodes,https://github.com/pytorch/pytorch/issues/37387,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2021-05-25T21:52:01Z +issue,58836,Latest nightly produces CUDA torchscript that cannot be run,https://github.com/pytorch/pytorch/issues/58836,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-24T21:04:03Z +issue,58857,Call non class methods from `torch::jit::Module` in C++,https://github.com/pytorch/pytorch/issues/58857,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-24T20:40:39Z +issue,57887,[Mobile] Valid scale ratio with invalid error,https://github.com/pytorch/pytorch/issues/57887,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-05-24T17:58:00Z +issue,58804,A Null pointer dereference bug,https://github.com/pytorch/pytorch/issues/58804,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-24T16:09:50Z +issue,58516,Unhelpful error message when type annotation is violated.,https://github.com/pytorch/pytorch/issues/58516,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-18T22:58:32Z +issue,57639,"After tracing, the device of the named buffer is still `cuda` although it was moved to `cpu `",https://github.com/pytorch/pytorch/issues/57639,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-18T16:56:06Z +issue,53472,After upgrade to 1.8.0 memory leak in libTorch after using torch::jit::load(),https://github.com/pytorch/pytorch/issues/53472,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",categorize and decide next maintainer owner/action,2021-05-18T14:50:15Z +issue,57739,object with __torch_functon__ does not work with jit function,https://github.com/pytorch/pytorch/issues/57739,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-11T18:03:57Z +issue,57793,"LibTorch inference script model is slower than PyTorch, and the speed is not stable. Why?",https://github.com/pytorch/pytorch/issues/57793,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-07T23:38:04Z +issue,57717,TorchScript JIT Compiler Can't Recognize Repeated Branches (Scripting),https://github.com/pytorch/pytorch/issues/57717,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-07T12:50:30Z +issue,57592,`jit.trace` fails to capture single-element tensor passed into op requiring scalar-type argument,https://github.com/pytorch/pytorch/issues/57592,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-05T05:59:28Z +issue,51931,[Pipe] Tied layers,https://github.com/pytorch/pytorch/issues/51931,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-05-05T00:30:15Z +issue,51932,[Pipe] Memory-Efficient Model Construction,https://github.com/pytorch/pytorch/issues/51932,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-05-05T00:30:09Z +issue,43765,JIT fails sanity checks during tracing torch.rand_like,https://github.com/pytorch/pytorch/issues/43765,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-04T21:27:59Z +issue,55999,[JIT][ProfilingExecutor] A reliable and consistent API/protocol to query & propagate profiling data,https://github.com/pytorch/pytorch/issues/55999,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-04T19:45:54Z +issue,57313,[jit] pickling custom classes may lead to invalid pickle files,https://github.com/pytorch/pytorch/issues/57313,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-04T17:12:33Z +issue,48066,[JIT] Saving and Loading JIT Model Does Not Preserve __dir__,https://github.com/pytorch/pytorch/issues/48066,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-04T17:02:42Z +issue,53904,[JIT] Support `torch.ones([])`,https://github.com/pytorch/pytorch/issues/53904,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-05-01T17:23:03Z +issue,57163,Segmentation fault when using add_graph from tensorboard in combination with try except in forward pass,https://github.com/pytorch/pytorch/issues/57163,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2021-04-29T16:36:39Z +issue,57035,`typing.List` need to be on the same line as the variable declaration?,https://github.com/pytorch/pytorch/issues/57035,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-27T16:51:17Z +issue,56006,Unable to get a Vulkan tensor using `to('vulkan')`,https://github.com/pytorch/pytorch/issues/56006,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: vulkan, oncall: mobile",categorize and decide next maintainer owner/action,2021-04-26T17:04:38Z +issue,56244,Cuda RPC error when using then(),https://github.com/pytorch/pytorch/issues/56244,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorpipe, oncall: distributed",categorize and decide next maintainer owner/action,2021-04-26T08:03:37Z +issue,56571,Python static variable access for TorchScript's custom C++ classes,https://github.com/pytorch/pytorch/issues/56571,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-22T16:30:29Z +issue,52655,Optimize `einsum` in TorchScript profile guided optimization,https://github.com/pytorch/pytorch/issues/52655,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance, oncall: jit",categorize and decide next maintainer owner/action,2021-04-20T21:14:25Z +issue,56096,[JIT] Merge/reconcile `__jit_unused_properties` and `__jit_ignored_attributes__`,https://github.com/pytorch/pytorch/issues/56096,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T17:35:25Z +issue,56159,Addition/subtraction/multiplication/division of bool variables are not supported in JIT,https://github.com/pytorch/pytorch/issues/56159,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T17:19:35Z +issue,43249,[JIT] Cannot iterate over nn.ModuleList in JITted code when accessed through one level of indirection,https://github.com/pytorch/pytorch/issues/43249,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:34:41Z +issue,49614,Improve error on missing/problematic Module member type annotation,https://github.com/pytorch/pytorch/issues/49614,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:32:35Z +issue,50231,"Tracing with autocast failed with error: ""Cannot insert a Tensor that requires grad as a constant""",https://github.com/pytorch/pytorch/issues/50231,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), oncall: jit",categorize and decide next maintainer owner/action,2021-04-20T16:32:24Z +issue,51018,len of ModuleList is incorrect for jitted models,https://github.com/pytorch/pytorch/issues/51018,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:32:13Z +issue,49261,[TorchScipt] Custom Objects C++ API,https://github.com/pytorch/pytorch/issues/49261,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:31:56Z +issue,49820,[JIT] Don't support math.exp and math.log in the Script,https://github.com/pytorch/pytorch/issues/49820,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:31:34Z +issue,51465,torch::jit::Module cannot be forward declared and used into a `unique_ptr`,https://github.com/pytorch/pytorch/issues/51465,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:30:49Z +issue,51137,[TorchScript Usability] Earlier type checking needed for interface type,https://github.com/pytorch/pytorch/issues/51137,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:29:28Z +issue,51942,Overwritten method has wrong self when JIT'ed,https://github.com/pytorch/pytorch/issues/51942,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:28:45Z +issue,51994,Allow Custom Classes to register a handler for .to operations,https://github.com/pytorch/pytorch/issues/51994,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, oncall: jit",categorize and decide next maintainer owner/action,2021-04-20T16:28:22Z +issue,52309,[JIT] __getitem__ on scripted ModuleDict returns unscripted module,https://github.com/pytorch/pytorch/issues/52309,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:28:02Z +issue,52920,[TorchScript Usability] More intuitive error message when passing wrong types to `torch.jit.script()`,https://github.com/pytorch/pytorch/issues/52920,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:27:11Z +issue,53344,[JIT] raise unittest.SkipTest and self.skipTest cannot be used to skip tests in JitTestCase,https://github.com/pytorch/pytorch/issues/53344,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-20T16:26:22Z +issue,56314,test_variant_consistency_jit fails for torch.tensordot with dtype float32 with error INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/56314,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: jit",categorize and decide next maintainer owner/action,2021-04-20T03:54:54Z +issue,55779,Request setNumThreads API for Java,https://github.com/pytorch/pytorch/issues/55779,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-04-19T17:38:19Z +issue,56007,NNAPI support for torch.cat([...]) when dim=-1,https://github.com/pytorch/pytorch/issues/56007,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-04-19T17:37:52Z +issue,56111,ProcessGroupGlooTest and other distributed tests don't capture stdout/stderr in test UI view,https://github.com/pytorch/pytorch/issues/56111,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-04-16T15:34:22Z +issue,51055,Native code assert or hang on process exit with a torch.jit.script model on multi-GPU,https://github.com/pytorch/pytorch/issues/51055,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-16T15:05:53Z +issue,34682,Pytorch not compatible with react native android,https://github.com/pytorch/pytorch/issues/34682,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-04-15T23:32:03Z +issue,48108,JIT should not force users to write ugly code,https://github.com/pytorch/pytorch/issues/48108,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-15T18:14:14Z +issue,56010,[tensorboard] Expected data-type of images in tensorboard SummaryWriter,https://github.com/pytorch/pytorch/issues/56010,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2021-04-15T10:36:09Z +issue,56024,Support for mkldnn + ddp,https://github.com/pytorch/pytorch/issues/56024,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, oncall: distributed",categorize and decide next maintainer owner/action,2021-04-14T15:46:05Z +issue,55541,`torch.jit.interface` does not understand dunder methods (e.g. `__call__`),https://github.com/pytorch/pytorch/issues/55541,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-13T17:17:39Z +issue,35215,Get hold of Backward graph from a C++ module.,https://github.com/pytorch/pytorch/issues/35215,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-13T08:12:03Z +issue,55782,YoloV4 Pytorch model inference fails on NNAPI Android,https://github.com/pytorch/pytorch/issues/55782,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-04-13T00:38:19Z +issue,44469,Error in the converting to pytorch mobile..,https://github.com/pytorch/pytorch/issues/44469,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-04-07T17:42:20Z +issue,55194,Tracing `len(tensor)` gets static constant value in result graph,https://github.com/pytorch/pytorch/issues/55194,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-04-02T17:36:10Z +issue,32901,QNNPACK linear doesn't preserve dimensions,https://github.com/pytorch/pytorch/issues/32901,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-04-01T00:13:19Z +issue,55072,[JIT][Testing] Replace use of `assertRaisesRegex` with `assertRaisesRegexWithHilight`,https://github.com/pytorch/pytorch/issues/55072,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-31T18:30:08Z +issue,54912,[autocast] DataParallel in a single process - possibly outdated docs,https://github.com/pytorch/pytorch/issues/54912,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-03-31T18:29:22Z +issue,54598,Type refinement with `isinstance` only works with generics,https://github.com/pytorch/pytorch/issues/54598,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-30T17:14:19Z +issue,54599,Better documentation for `torch.jit.isinstance`,https://github.com/pytorch/pytorch/issues/54599,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: jit",categorize and decide next maintainer owner/action,2021-03-30T17:13:27Z +issue,50933,[JIT] Type refinement failure in if-elif-else structure,https://github.com/pytorch/pytorch/issues/50933,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-30T15:55:58Z +issue,54625,"[torchscript] nn.Embedding(_weight), tensor.masked_fill_, and torch.autograd.grad causes wrong gradient",https://github.com/pytorch/pytorch/issues/54625,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-30T05:24:19Z +issue,45524,[JIT] JIT Autograd is saving more results than are necessary for bprop like the result of RELU when it is an intermediate,https://github.com/pytorch/pytorch/issues/45524,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-26T17:55:21Z +issue,13164,Tracing custom ops,https://github.com/pytorch/pytorch/issues/13164,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-26T10:15:40Z +issue,54348,[JIT] TorchScript does not correctly handle `torch.` namespace functions called with `kwargs`,https://github.com/pytorch/pytorch/issues/54348,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-23T21:06:14Z +issue,54435,[JIT] prim::ModuleContainerIndex doesnt work with freezing,https://github.com/pytorch/pytorch/issues/54435,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-23T17:38:25Z +issue,54365,Support `Generic` Type in TorchScript,https://github.com/pytorch/pytorch/issues/54365,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-23T17:16:40Z +issue,54146,Conversion of Pytorch model to Torchscript for model LaneATT(Lane Detection),https://github.com/pytorch/pytorch/issues/54146,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-23T04:08:24Z +issue,46944,Export TorchScript Classes as TorchScript code,https://github.com/pytorch/pytorch/issues/46944,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-18T01:30:33Z +issue,53921,[JIT] TorchScript infers incorrect type for `or` in non-boolean context,https://github.com/pytorch/pytorch/issues/53921,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-16T17:41:00Z +issue,53862,"After the source code is compiled, an error will be reported fvcore",https://github.com/pytorch/pytorch/issues/53862,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-16T17:27:20Z +issue,53878,[JIT] TorchScript errors out when assigning a value to an existing variable that's defined in outer block,https://github.com/pytorch/pytorch/issues/53878,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-16T17:26:22Z +issue,53877,[JIT] Bad interaction between if-else-style and assert-style type refinement,https://github.com/pytorch/pytorch/issues/53877,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-16T17:22:40Z +issue,53753,TorchScript cannot handle declared module attributes of Python types,https://github.com/pytorch/pytorch/issues/53753,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-16T17:20:14Z +issue,50971,Failing to trace Huggingface models using Trace API from Torchscript,https://github.com/pytorch/pytorch/issues/50971,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-15T12:13:17Z +issue,41512,Tensor.new_tensor is not supported,https://github.com/pytorch/pytorch/issues/41512,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-14T23:44:03Z +issue,47692,pyre not able to infer the type of torch.jit.script,https://github.com/pytorch/pytorch/issues/47692,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-12T21:27:40Z +issue,53824,[JIT] torch.jit.optimized_execution(True) greatly slows down some operations in PyTorch 1.8.0,https://github.com/pytorch/pytorch/issues/53824,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-12T05:50:45Z +issue,53264,How to load trained . torch from conversion to .mlmodel,https://github.com/pytorch/pytorch/issues/53264,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-03-12T05:28:21Z +issue,53867,[JIT][Static Runtime] Memory optimization for output tensors,https://github.com/pytorch/pytorch/issues/53867,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-12T02:17:33Z +issue,53414,[JIT] Metaprogram non-type based expression trees,https://github.com/pytorch/pytorch/issues/53414,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-09T18:26:34Z +issue,52589,[JIT-autodiff] batch_norm isn't differentiable,https://github.com/pytorch/pytorch/issues/52589,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-09T07:25:09Z +issue,53140,[JIT] Document __jit_ignored_attributes__,https://github.com/pytorch/pytorch/issues/53140,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-09T02:24:27Z +issue,52768,[nnc] Optimize tiny convolutions,https://github.com/pytorch/pytorch/issues/52768,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-07T03:25:44Z +issue,52772,[nnc] Optimize 1x1 convolutions,https://github.com/pytorch/pytorch/issues/52772,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-07T03:25:37Z +issue,52777,[nnc] Placeholder::load braced init list fails to deduce,https://github.com/pytorch/pytorch/issues/52777,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-07T03:25:25Z +issue,52821,[nnc] Compute expressions with > 4 axes is pretty tedious,https://github.com/pytorch/pytorch/issues/52821,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-07T03:24:20Z +issue,46995,max_pool2d always fails on native Android app,https://github.com/pytorch/pytorch/issues/46995,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-03-05T16:23:39Z +issue,52599,torch::jit::Node::isNondeterministic() does not include aten::feature_dropout or inline kinds,https://github.com/pytorch/pytorch/issues/52599,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-03T16:53:04Z +issue,53017,[TorchScript Usability] Optional type does not support `x != None`,https://github.com/pytorch/pytorch/issues/53017,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-02T18:11:08Z +issue,53022,[TorchScript Usability] Instance attribute annotation does not seem to take effect,https://github.com/pytorch/pytorch/issues/53022,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-02T18:10:18Z +issue,52970,Wrong link in readme for the android nightly version,https://github.com/pytorch/pytorch/issues/52970,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, module: docs, oncall: mobile",categorize and decide next maintainer owner/action,2021-03-01T16:26:37Z +issue,50444,[TorchScript] Rough edges discovered during type system rewrite,https://github.com/pytorch/pytorch/issues/50444,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-03-01T15:56:34Z +issue,52788,distributed/optim/test_zero_redundancy_optimizer uses unittest.main() instead of common_utils.run_tests(),https://github.com/pytorch/pytorch/issues/52788,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-02-25T17:06:54Z +issue,52795,Add format indexed argument support in JIT,https://github.com/pytorch/pytorch/issues/52795,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-25T02:07:16Z +issue,52720,[JIT] Finish migrating torch._C._jit_ bindings to torch._C._jit,https://github.com/pytorch/pytorch/issues/52720,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-24T03:02:05Z +issue,49935,[FX] intermediate types of empty lists/dicts not preserved during torch.fx tracing,https://github.com/pytorch/pytorch/issues/49935,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-23T05:51:09Z +issue,52607,Enable timed barrier in c10d,https://github.com/pytorch/pytorch/issues/52607,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",categorize and decide next maintainer owner/action,2021-02-22T18:44:55Z +issue,46946,Changing order of field assignments breaks TorchScript classes,https://github.com/pytorch/pytorch/issues/46946,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-17T23:07:36Z +issue,39690,[Quantization] Output tensor type is lost after serializing and loading back a quantized model,https://github.com/pytorch/pytorch/issues/39690,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-15T12:48:57Z +issue,51835,methods decorated with torch.jit._overload_method are not accessible,https://github.com/pytorch/pytorch/issues/51835,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-11T22:37:46Z +issue,51379,TorchScript class compilation failure with pybind11-2.6.2,https://github.com/pytorch/pytorch/issues/51379,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-11T18:34:25Z +issue,52147,Pointer passed where number is expected for PYTORCH_CUDA_FUSER_JIT_OPT_LEVEL leading to crash,https://github.com/pytorch/pytorch/issues/52147,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-11T17:44:14Z +issue,52144,Concatenation of Java tensors (Android for mobile),https://github.com/pytorch/pytorch/issues/52144,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-02-11T17:37:40Z +issue,51409,C++ load model error,https://github.com/pytorch/pytorch/issues/51409,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-09T22:23:24Z +issue,48703,Caffe2 upsample do not support float times output_size,https://github.com/pytorch/pytorch/issues/48703,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-02-08T23:34:44Z +issue,51808,[jit] Keyword-arg expansion for constants,https://github.com/pytorch/pytorch/issues/51808,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-08T17:32:28Z +issue,45465,torchscript model size increases when loading and saving it again,https://github.com/pytorch/pytorch/issues/45465,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-05T19:38:15Z +issue,40914,Inconsistent handling of torch.Size.__add__,https://github.com/pytorch/pytorch/issues/40914,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T21:20:37Z +issue,41513,Python object None check not supported,https://github.com/pytorch/pytorch/issues/41513,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T20:58:41Z +issue,41415,[jit] `set` builtin support,https://github.com/pytorch/pytorch/issues/41415,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T20:51:26Z +issue,41046,Human-readable names and operations for TorchScript model graphs,https://github.com/pytorch/pytorch/issues/41046,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T20:48:14Z +issue,43873,PackedSequence objects that are created within jit.script are not PackedSequence,https://github.com/pytorch/pytorch/issues/43873,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T20:33:23Z +issue,45199,JIT doesn't support rfloordiv with scalars correctly,https://github.com/pytorch/pytorch/issues/45199,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T17:48:05Z +issue,48549,JIT script has wrong signature for .norm,https://github.com/pytorch/pytorch/issues/48549,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T17:26:04Z +issue,48191,[JIT] Contained Module Attributes should be recursively compiled,https://github.com/pytorch/pytorch/issues/48191,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-04T01:24:40Z +issue,51136,[TorchScript Usability] error message when invoking __init__ of module types inside TorchScript,https://github.com/pytorch/pytorch/issues/51136,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-03T08:08:11Z +issue,51616,[RPC] rref.get_type() should have default return type of future,https://github.com/pytorch/pytorch/issues/51616,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",categorize and decide next maintainer owner/action,2021-02-03T05:31:37Z +issue,50834,Precision of scalars is lost in script mode for binary operations with tensors,https://github.com/pytorch/pytorch/issues/50834,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-02T18:06:53Z +issue,51563,[TorchScript Usability] Type-checker too strict for Optional type initialization pattern,https://github.com/pytorch/pytorch/issues/51563,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-02T15:29:28Z +issue,48408,TorchScript: Optional sized lists cannot be called with optional scalar values,https://github.com/pytorch/pytorch/issues/48408,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-02-01T17:58:27Z +issue,29092,[FR] script returns subclass of the original module class,https://github.com/pytorch/pytorch/issues/29092,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-28T00:40:30Z +issue,51140,[TorchScript Usability] Non-intuitive error messages for user errors,https://github.com/pytorch/pytorch/issues/51140,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-26T20:40:31Z +issue,48928,"When multiple GPUs run multiple processes, it is found that any process not running in GPU 0 will have some more memory (such as 200m) in GPU 0. What is the cause of this?（多个GPU跑多进程时候，发现只要不在0号GPU跑的进程都会在0号GPU多出一些内存(如200M)，请问这是什么情况导致的？）",https://github.com/pytorch/pytorch/issues/48928,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2021-01-21T06:46:15Z +issue,49650,Annotating submodule with correct type fails scripting,https://github.com/pytorch/pytorch/issues/49650,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-21T00:07:05Z +issue,47964,TorchScript is unable to distinguish int and ScalarType,https://github.com/pytorch/pytorch/issues/47964,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-21T00:04:42Z +issue,42673,"Unexpected behavior of ""to"" method inside a torch.jit.script decorated function",https://github.com/pytorch/pytorch/issues/42673,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-21T00:04:10Z +issue,50772,[TorchScript Usability] Incorrect default types,https://github.com/pytorch/pytorch/issues/50772,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-20T01:31:04Z +issue,50678,[PyTorch Mobile] NNAPI prototype installer no longer available,https://github.com/pytorch/pytorch/issues/50678,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2021-01-18T23:20:51Z +issue,50531,"PyTorch mobile perf recipes, fails when building android",https://github.com/pytorch/pytorch/issues/50531,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2021-01-15T03:00:18Z +issue,49372,[jit] Support `torch.Tensor.numel()` and `torch.Size.numel()` properly in tracing,https://github.com/pytorch/pytorch/issues/49372,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-13T01:42:28Z +issue,49569,Update jit/OVERVIEW.md,https://github.com/pytorch/pytorch/issues/49569,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-12T00:48:31Z +issue,49585,[JIT] Serialization forward compatibility tests,https://github.com/pytorch/pytorch/issues/49585,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-08T22:57:10Z +issue,50127,JIT does not correctly compile custom classes derived from torch.Tensor,https://github.com/pytorch/pytorch/issues/50127,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2021-01-06T05:21:23Z +issue,16602,Error while installing pytorch,https://github.com/pytorch/pytorch/issues/16602,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: build,categorize and decide next maintainer owner/action,2021-01-04T18:46:29Z +issue,49823,Why trace method used insted of script method in graph creation?,https://github.com/pytorch/pytorch/issues/49823,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2020-12-29T21:35:50Z +issue,8818,MPI causing job to hang --- unresponsive to external (termination) signals,https://github.com/pytorch/pytorch/issues/8818,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2020-12-29T18:21:19Z +issue,49536,Pytorch mobile Vulkan API bug during call to ReLU,https://github.com/pytorch/pytorch/issues/49536,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: java, oncall: mobile",categorize and decide next maintainer owner/action,2020-12-21T06:00:55Z +issue,43488,"c10::Error: Couldn’t find an operator for aten::dropout(Tensor input, float p, bool train)",https://github.com/pytorch/pytorch/issues/43488,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-12-20T09:16:16Z +issue,43261,OSError: could not find class definition when export torchscript module,https://github.com/pytorch/pytorch/issues/43261,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-12-18T07:27:34Z +issue,48869,Tracing fails sanity check,https://github.com/pytorch/pytorch/issues/48869,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-12-17T22:15:13Z +issue,49253,Differentiate between objects with a true `Tensor` type and objects with a default `Tensor` type,https://github.com/pytorch/pytorch/issues/49253,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-12-15T19:14:41Z +issue,49377,[JIT-DEBUG] Suppress GRAPH dump in profile-graph-executor,https://github.com/pytorch/pytorch/issues/49377,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-12-15T00:23:03Z +issue,40696,"Trace model with floor, Inconsistent performance",https://github.com/pytorch/pytorch/issues/40696,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-12-10T05:06:05Z +issue,48984,Scripting silently ignores methods of class derived from `NamedTuple`,https://github.com/pytorch/pytorch/issues/48984,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-12-10T02:22:45Z +issue,40906,will you support SRU on mobile?,https://github.com/pytorch/pytorch/issues/40906,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2020-12-03T05:14:19Z +issue,48710,RuntimeError: CUDNN_STATUS_BAD_PARAM when testing environment,https://github.com/pytorch/pytorch/issues/48710,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,categorize and decide next maintainer owner/action,2020-12-03T00:11:43Z +issue,48478,Callbacks of Futures shouldn't wait inline on another future,https://github.com/pytorch/pytorch/issues/48478,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",categorize and decide next maintainer owner/action,2020-12-02T01:22:57Z +issue,41419,[jit] Support NamedTuple in tracing,https://github.com/pytorch/pytorch/issues/41419,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2020-12-01T18:19:30Z +issue,48307,Provide torchscript support for staticmethods for classes derived from Enum,https://github.com/pytorch/pytorch/issues/48307,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-12-01T08:37:43Z +issue,48410,PyTorch Mobile speed_benchmark_torch crashes on mobile optimized model,https://github.com/pytorch/pytorch/issues/48410,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-11-26T01:14:43Z +issue,48045,"[NNC] Improve ""UNSUPPORTED DTYPE"" error messages",https://github.com/pytorch/pytorch/issues/48045,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",categorize and decide next maintainer owner/action,2020-11-17T15:43:17Z +issue,48094,[JIT][PE] Profiling Executor should re-profile `DifferentiableGraph`,https://github.com/pytorch/pytorch/issues/48094,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-11-17T13:37:45Z +issue,48090,[JIT] Constant folding on no-op `aten::mul`,https://github.com/pytorch/pytorch/issues/48090,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-11-17T12:47:11Z +issue,48089,[JIT] Optimization pass in profiling executor to fold away conditional branches,https://github.com/pytorch/pytorch/issues/48089,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-11-17T12:22:41Z +issue,47342,My demo app uses other package name will crash but org.pytorch.demo,https://github.com/pytorch/pytorch/issues/47342,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-11-13T03:29:25Z +issue,47883,`.modules()` is not callable in TorchScript when any submodule is a module interface type (i.e. a class decorated by `@torch.jit.interface`),https://github.com/pytorch/pytorch/issues/47883,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-11-13T01:39:08Z +issue,47335,[jit] support length inference for list comprehension,https://github.com/pytorch/pytorch/issues/47335,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-11-10T20:03:56Z +issue,47682,Common dunder methods fail when called directly,https://github.com/pytorch/pytorch/issues/47682,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-11-10T18:35:56Z +issue,47265,[Feature Request] tensorboard add_graph support for multiple method,https://github.com/pytorch/pytorch/issues/47265,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2020-11-04T16:06:18Z +issue,47177,LayerNorm on Android will cause very high cpu usage,https://github.com/pytorch/pytorch/issues/47177,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-11-03T05:26:35Z +issue,47010,[JIT] Tracing turns None Device Input into a Constant,https://github.com/pytorch/pytorch/issues/47010,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-11-02T23:26:38Z +issue,46979,[JIT] scripting torchaudio.transforms.MFCC() error,https://github.com/pytorch/pytorch/issues/46979,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-11-02T23:23:06Z +issue,46654,Converting object detector model to TorchScript,https://github.com/pytorch/pytorch/issues/46654,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-10-27T07:33:43Z +issue,19053,RNN module weights not compacted,https://github.com/pytorch/pytorch/issues/19053,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-10-22T03:02:26Z +issue,46505,Cannot script when a static method or cross module function calls custom op,https://github.com/pytorch/pytorch/issues/46505,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: torchbind, oncall: jit",categorize and decide next maintainer owner/action,2020-10-20T17:53:10Z +issue,46368,[JIT] Inconsistent behavior of torchScript parser in python and c++,https://github.com/pytorch/pytorch/issues/46368,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-10-20T17:31:33Z +issue,46574,Mismatch between the documentation and implementation of caffe2 PReLU operator,https://github.com/pytorch/pytorch/issues/46574,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs,categorize and decide next maintainer owner/action,2020-10-20T15:32:12Z +issue,43125,"Non-conforming variable identifiers in JIT code errors / printouts: e.g. ""Tensor 0""",https://github.com/pytorch/pytorch/issues/43125,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-10-08T02:10:35Z +issue,40500,Not able to launch tensorboard using pytorch,https://github.com/pytorch/pytorch/issues/40500,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2020-10-07T09:04:03Z +issue,45327,Tensorboard summary_iterator,https://github.com/pytorch/pytorch/issues/45327,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2020-10-05T16:54:58Z +issue,17126,Support callables in scripted functions,https://github.com/pytorch/pytorch/issues/17126,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-10-05T12:21:07Z +issue,45208,Adding @torch.no_grad() to forward () causes undefined value in torch.jit.script(),https://github.com/pytorch/pytorch/issues/45208,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-29T17:12:13Z +issue,43323,Set SummaryWriter step globally,https://github.com/pytorch/pytorch/issues/43323,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2020-09-24T13:52:43Z +issue,44608,building torchscript extension results in INVALID TYPE: Only int64_t and bool are supported as an integral argument type custom_class,https://github.com/pytorch/pytorch/issues/44608,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-23T04:26:11Z +issue,44768,[JIT] function decorated with @torch.no_grad() can not be exported when another function is called inside it.,https://github.com/pytorch/pytorch/issues/44768,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-21T22:12:30Z +issue,44992,[jit] cann't pass class objects across boundary of tracing/scripting,https://github.com/pytorch/pytorch/issues/44992,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-21T22:11:07Z +issue,45012,torch.jit saves nonpersistent buffers,https://github.com/pytorch/pytorch/issues/45012,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-21T21:57:50Z +issue,44800,torchscript continue training,https://github.com/pytorch/pytorch/issues/44800,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-21T21:16:56Z +issue,39240,[mobile] Running on armeabi-v7a is inconsistent with arm64-v8a,https://github.com/pytorch/pytorch/issues/39240,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-09-20T11:54:32Z +issue,44503,TestJitGeneratedModule.test_nn_CTCLoss_lengths_intlists fails if not skipped,https://github.com/pytorch/pytorch/issues/44503,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-16T14:17:39Z +issue,44446,[jit] class type optional type annotation not working properly,https://github.com/pytorch/pytorch/issues/44446,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-09-14T21:58:21Z +issue,44630,No pytorch_jni in StartLocal's zip,https://github.com/pytorch/pytorch/issues/44630,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-09-14T17:03:46Z +issue,43119,Tracing can leak CUDA memory when using tensor constants,https://github.com/pytorch/pytorch/issues/43119,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-25T21:27:00Z +issue,43378,[JIT] Using Any type variable inside can produce incorrect results due to type unification problems,https://github.com/pytorch/pytorch/issues/43378,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-24T22:21:35Z +issue,42947,[jit] std::exception_ptr can't be torchbind/convert to ivalue,https://github.com/pytorch/pytorch/issues/42947,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-24T20:35:50Z +issue,42948,[jit] TorchBind std::chrono::milliseconds,https://github.com/pytorch/pytorch/issues/42948,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-24T20:35:08Z +issue,41683,'torch::jit::script::ErrorReport' from 'torch::jit::load('modelpath') in Visual Studio 2017,https://github.com/pytorch/pytorch/issues/41683,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-22T06:22:36Z +issue,43409,Switch C10_EXPORT_CAFFE2_OP_TO_C10 to new operator registration API,https://github.com/pytorch/pytorch/issues/43409,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,categorize and decide next maintainer owner/action,2020-08-21T22:37:37Z +issue,43328,torch.utils.tensorboard.SummaryWriter.add_histogram() execution time explodes with each epoch.,https://github.com/pytorch/pytorch/issues/43328,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2020-08-20T17:05:11Z +issue,43134,[jit] need a better way to handle mix CPU/GPU (Inference/Training) for tracing,https://github.com/pytorch/pytorch/issues/43134,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-19T17:49:31Z +issue,41291,Return list of tuples from custom operator,https://github.com/pytorch/pytorch/issues/41291,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-10T21:40:53Z +issue,38744,[jit] `jit.annotate` and multiple compilation of functions with different types,https://github.com/pytorch/pytorch/issues/38744,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-08T21:30:39Z +issue,42170,Tracing a[b] records converting b to fixed device type of a,https://github.com/pytorch/pytorch/issues/42170,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-08-07T22:44:13Z +issue,42039,Segmentation fault (core dumped) when running optimize_for_mobile,https://github.com/pytorch/pytorch/issues/42039,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-08-03T14:48:41Z +issue,42355,nn::Sequential Link error in Android version,https://github.com/pytorch/pytorch/issues/42355,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",categorize and decide next maintainer owner/action,2020-08-01T08:18:51Z +issue,42379,TensorBoard Summaries Written with `SummaryWriter.add_scalars()` are not Purged,https://github.com/pytorch/pytorch/issues/42379,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2020-07-31T21:43:32Z +issue,42308,Not able to run quantized model on android,https://github.com/pytorch/pytorch/issues/42308,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-07-30T19:50:57Z +issue,42145,Support Python type() in TorchScript,https://github.com/pytorch/pytorch/issues/42145,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-07-28T18:56:04Z +issue,42081,Document optimizations enabled by JIT scripting and tracing (TorchScript),https://github.com/pytorch/pytorch/issues/42081,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-07-28T17:51:37Z +issue,39724,pickle_save on mobile (no longer works with 1.5.0 release),https://github.com/pytorch/pytorch/issues/39724,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-07-24T06:53:58Z +issue,41433,[JIT][to-backend] `selective_to_backend` infra,https://github.com/pytorch/pytorch/issues/41433,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-07-14T23:31:51Z +issue,41418,[jit] support `rpc_remote` and `rpc_sync`,https://github.com/pytorch/pytorch/issues/41418,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-07-14T18:41:18Z +issue,41141,Sum of two tuples becomes a list in torchscript,https://github.com/pytorch/pytorch/issues/41141,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-07-09T14:24:06Z +issue,39569,[JIT] OrderedDict doesn't support custom objects,https://github.com/pytorch/pytorch/issues/39569,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-06-05T19:04:50Z +issue,39539,[JIT] Print out mutation in IR Dumps,https://github.com/pytorch/pytorch/issues/39539,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-06-04T21:44:11Z +issue,38614,`SummaryWriter.add_graph` borks with simple example,https://github.com/pytorch/pytorch/issues/38614,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2020-06-04T17:19:57Z +issue,39309,[JIT] nn.ModuleList loses None objects inside it after scripting,https://github.com/pytorch/pytorch/issues/39309,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-06-04T15:29:30Z +issue,38312,Status of support for training on mobile,https://github.com/pytorch/pytorch/issues/38312,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-06-02T16:31:05Z +issue,39292,Tensorboard，graph in pytorch 1.4 is more complicated than pytorch 1.1？,https://github.com/pytorch/pytorch/issues/39292,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2020-06-01T01:33:45Z +issue,36211,[JIT] support self.named_buffers and self.named_parameters in TorchScript,https://github.com/pytorch/pytorch/issues/36211,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-05-31T07:17:24Z +issue,38273,Adding model metadata in TorchScript model file,https://github.com/pytorch/pytorch/issues/38273,needs triage,47,"triage signal present; old update age or low-urgency/stale label; component/oncall labels: module: serialization, oncall: jit",categorize and decide next maintainer owner/action,2020-05-20T23:56:56Z +issue,37740,torch Summary writer does not display torchvision.io.read_video output,https://github.com/pytorch/pytorch/issues/37740,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2020-05-04T22:01:25Z +issue,36926,JIT string ops and other miscellaneous ops probably shouldn't be in aten namespace,https://github.com/pytorch/pytorch/issues/36926,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-04-27T14:50:42Z +issue,37016,Guidance on implementing a new backend,https://github.com/pytorch/pytorch/issues/37016,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-04-22T18:34:20Z +issue,36603,TorchScript Support for Named Tensors,https://github.com/pytorch/pytorch/issues/36603,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-04-17T20:29:15Z +issue,33281,"Uninitialised value was created by a stack allocation, reported by valgrind",https://github.com/pytorch/pytorch/issues/33281,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-04-10T07:20:36Z +issue,36182,## 🐛 Bug: QNNPACK tests failing on master on Nexus 6,https://github.com/pytorch/pytorch/issues/36182,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-04-07T23:33:04Z +issue,35526,RuntimeError during converting Reformer model to TorchScript,https://github.com/pytorch/pytorch/issues/35526,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-04-03T21:16:59Z +issue,35606,"Could not find any similar ops to ""foo..."" in the `Libtorch`",https://github.com/pytorch/pytorch/issues/35606,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-04-01T18:39:32Z +issue,35095,"tensorboard add_graph 's ""operator_export_type""",https://github.com/pytorch/pytorch/issues/35095,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",categorize and decide next maintainer owner/action,2020-03-23T18:18:27Z +issue,33631,bitmapToFloat32Tensor() 1 channel Tensor [feature] [mobile],https://github.com/pytorch/pytorch/issues/33631,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: java, oncall: mobile",categorize and decide next maintainer owner/action,2020-03-22T17:41:31Z +issue,33124,QNNPACK: GNU aarch64 assembler does not support 4s on neon mov,https://github.com/pytorch/pytorch/issues/33124,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,categorize and decide next maintainer owner/action,2020-03-21T16:50:07Z +issue,34851,[jit] `__hash__` magic method is missing,https://github.com/pytorch/pytorch/issues/34851,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-03-17T18:25:21Z +issue,29125,_compared_saved_loaded doesn't work with torch.tensor constants,https://github.com/pytorch/pytorch/issues/29125,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-03-16T19:13:35Z +issue,33993,ModuleList doesn't support slicing in TorchScript,https://github.com/pytorch/pytorch/issues/33993,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-02-29T01:24:55Z +issue,29128,JIT should respect SKIP_PYTHON_BINDINGS and SKIP_PYTHON_BINDINGS_SIGNATURES,https://github.com/pytorch/pytorch/issues/29128,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-02-28T18:52:17Z +issue,28267,Problem with jit TorchScript while copying data between GRUs,https://github.com/pytorch/pytorch/issues/28267,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-02-28T18:52:15Z +issue,28191,[JIT] Allow user to provide aliasing information on input tensors and model parameters,https://github.com/pytorch/pytorch/issues/28191,needs triage,47,triage signal present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-02-28T18:52:15Z +issue,33810,Wrongly detected: Division of ints in TorchScript uses Python 3 true division semantics,https://github.com/pytorch/pytorch/issues/33810,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-02-26T11:33:19Z +issue,33285,iOS libtorch superpoint model bug,https://github.com/pytorch/pytorch/issues/33285,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: ios, module: macos, oncall: mobile",categorize and decide next maintainer owner/action,2020-02-22T18:41:35Z +issue,33110,InlineAutodiffSubgraphs in JIT inlines non-differentiable custom groups unexpectedly.,https://github.com/pytorch/pytorch/issues/33110,needs triage,47,"PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, oncall: jit",categorize and decide next maintainer owner/action,2020-02-14T20:00:30Z +issue,9595,[JIT] Unify python -> SugaredValue construction + fix inlining of graphs with Tuple-typed inputs,https://github.com/pytorch/pytorch/issues/9595,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2020-02-11T02:11:14Z +issue,30863,"[TensorBoard] Installed tensorboard 1.14.0, TestTensorBoardWriter.test_writer failed.",https://github.com/pytorch/pytorch/issues/30863,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2020-01-11T12:19:32Z +issue,31603,Typo in `torch.utils.tensorboard.add_image`,https://github.com/pytorch/pytorch/issues/31603,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2020-01-11T12:11:40Z +issue,7904,TracedModules don't support parameter sharing between modules,https://github.com/pytorch/pytorch/issues/7904,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-12-30T13:38:31Z +issue,30812,Tensorboard graph failing to nest modules,https://github.com/pytorch/pytorch/issues/30812,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,categorize and decide next maintainer owner/action,2019-12-17T10:04:38Z +issue,18312,[jit] Trace perform weirdly on BatchNorm,https://github.com/pytorch/pytorch/issues/18312,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-11-22T03:26:53Z +issue,15116,torch.save does not work if nn.Module has partial JIT.,https://github.com/pytorch/pytorch/issues/15116,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-11-21T19:08:31Z +issue,14270,Pytorch C++ API with cuda : Expected object of backend CPU but got backend CUDA for sequence element 1 in sequence argument at position #1 'tensors',https://github.com/pytorch/pytorch/issues/14270,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-10-14T07:51:09Z +issue,18975,Tracing and Scripting nn.Conv2d shows different op in the graph,https://github.com/pytorch/pytorch/issues/18975,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-07-29T16:50:57Z +issue,9797,Build torch as a submodule with static linking doesn't work (CAFFE2_PERF_WITH_AVX2 is not defined),https://github.com/pytorch/pytorch/issues/9797,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: static linking,categorize and decide next maintainer owner/action,2019-06-13T23:05:26Z +issue,21255,IsType() ASSERT FAILED [Detectron e2e_mask_rcnn_R-50-C4_1x.yaml],https://github.com/pytorch/pytorch/issues/21255,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: assert failure,categorize and decide next maintainer owner/action,2019-06-04T15:43:54Z +issue,16263,[JIT] Support C++ front end module and JIT interop,https://github.com/pytorch/pytorch/issues/16263,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-05-29T23:07:18Z +issue,15421,JIT is not compatible with data parallel,https://github.com/pytorch/pytorch/issues/15421,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-05-14T00:57:08Z +issue,6898,[Caffe2] cudnn versions compatibility issue.,https://github.com/pytorch/pytorch/issues/6898,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: cudnn,categorize and decide next maintainer owner/action,2019-04-09T16:17:14Z +issue,18376,Cleanup code in register_c10_ops.cpp,https://github.com/pytorch/pytorch/issues/18376,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-04-06T16:06:32Z +issue,18380,"C++ inference (CPU-only) stalls in Android, and crashes on Mac/Linux",https://github.com/pytorch/pytorch/issues/18380,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: android,categorize and decide next maintainer owner/action,2019-04-06T16:05:03Z +issue,16293,Unexpected behavior of jit.trace when PYTORCH_JIT=0,https://github.com/pytorch/pytorch/issues/16293,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-03-22T16:21:57Z +issue,15043,[JIT] jit.trace fails with custom GRUs and CUDA when the sequence is longer,https://github.com/pytorch/pytorch/issues/15043,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2019-03-11T21:09:23Z +issue,14750,[caffe2] ConvTranspose with group attribute,https://github.com/pytorch/pytorch/issues/14750,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: op-unification,categorize and decide next maintainer owner/action,2019-01-07T07:28:48Z +issue,14484,prim::ConstantChunk derivative formula doesn't handle undefined inputs,https://github.com/pytorch/pytorch/issues/14484,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-11-28T18:20:39Z +issue,14332,[jit][script] support slicing with tensor literals,https://github.com/pytorch/pytorch/issues/14332,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-11-23T06:27:05Z +issue,13477,[jit] restrict promotion of single-element arguments to lists,https://github.com/pytorch/pytorch/issues/13477,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-11-01T20:57:14Z +issue,13170,Allow traced modules to return dictionaries,https://github.com/pytorch/pytorch/issues/13170,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-10-29T17:38:28Z +issue,11967,Jit cannot trace autograd for certain operator,https://github.com/pytorch/pytorch/issues/11967,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-09-24T17:43:00Z +issue,11532,[JIT][tracer] Slicing shape is specialized to tensor rank,https://github.com/pytorch/pytorch/issues/11532,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-09-11T18:24:53Z +issue,8453,"[JIT] Generalize checkTrace() to also compare gradients of parameters, not just inputs",https://github.com/pytorch/pytorch/issues/8453,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-07-10T16:24:16Z +issue,8556,[JIT] Interleaved C++-Python execution loses inner Python stacks,https://github.com/pytorch/pytorch/issues/8556,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-06-15T15:35:11Z +issue,6465,[caffe2] ChannelShuffle example,https://github.com/pytorch/pytorch/issues/6465,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: module: docs,categorize and decide next maintainer owner/action,2018-04-10T18:31:10Z +issue,6010,Develop a strategy for writing leak tests for pytorch.,https://github.com/pytorch/pytorch/issues/6010,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-03-28T17:50:31Z +issue,6007,Clean up the extra copies that occur in the execution engine and other places,https://github.com/pytorch/pytorch/issues/6007,needs triage,47,PyTorch issue lacks triaged label; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,categorize and decide next maintainer owner/action,2018-03-28T17:50:22Z +pr,187404,[inductor][cutlass] Normalize reshaped external reads in EVT epilogue fusion,https://github.com/pytorch/pytorch/pull/187404,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-09T01:25:28Z +pr,188029,[dynamo] Add namespace for cell variables that collide with locals,https://github.com/pytorch/pytorch/pull/188029,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-09T01:03:53Z +pr,189306,"[dynamo, 3.15] Fix test_bytecode_utils",https://github.com/pytorch/pytorch/pull/189306,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-09T00:20:53Z +pr,189022,[dynamo] Support itertools.count keyword args and repr,https://github.com/pytorch/pytorch/pull/189022,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T23:49:13Z +pr,188176,[dynamo] fix test_perf.py tests under canonicalization,https://github.com/pytorch/pytorch/pull/188176,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T23:22:21Z +pr,189188,flex_gemm: support scalar captured epilogue args,https://github.com/pytorch/pytorch/pull/189188,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T23:12:52Z +pr,188468,flex_gemm: add local reduce analysis helpers,https://github.com/pytorch/pytorch/pull/188468,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T23:12:48Z +pr,180020,Add a debug tool to warn on NULL stream use,https://github.com/pytorch/pytorch/pull/180020,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T21:58:38Z +pr,189155,Add CUDA Support for Fused `_add_relu` With Half and BFloat16,https://github.com/pytorch/pytorch/pull/189155,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T21:37:10Z +pr,189274,[dynamo] Route set/dict-view/list dunder methods through tp_slots,https://github.com/pytorch/pytorch/pull/189274,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T21:27:00Z +pr,188802,[MPS] Migrate conv3d to metal kernels,https://github.com/pytorch/pytorch/pull/188802,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T21:26:44Z +pr,189199,[DTensor] Fix spurious cross-device FakeTensor mismatch in multi-threaded tests,https://github.com/pytorch/pytorch/pull/189199,ready for maintainer decision,45,"CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T21:18:46Z +pr,189303,"Reapply ""Dispatch index_put_ to fill_ for scalar values (#188267 / #189105)""",https://github.com/pytorch/pytorch/pull/189303,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T21:16:25Z +pr,188991,[NVGEMM] Support `cutlass.operators`,https://github.com/pytorch/pytorch/pull/188991,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T21:16:04Z +pr,189102,[TEST][Inductor][CUDA] Bump tolerances for `test_comprehensive_combinations_cuda_float16`,https://github.com/pytorch/pytorch/pull/189102,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T21:13:45Z +pr,189037,Preserve strides for CuTeDSL grouped MM fake tensors,https://github.com/pytorch/pytorch/pull/189037,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T21:13:02Z +pr,189100,[ROCm] Fix test_is_pinned_no_context for Python 3.14 forkserver default,https://github.com/pytorch/pytorch/pull/189100,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T21:11:41Z +pr,189148,Add maybe_unused label to strides when strides is unused,https://github.com/pytorch/pytorch/pull/189148,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T21:00:41Z +pr,189182,make the return more specified,https://github.com/pytorch/pytorch/pull/189182,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T20:55:38Z +pr,189053,[dynamo] Support integer arithmetic on a sourceless id()/hash() FakeIdVariable,https://github.com/pytorch/pytorch/pull/189053,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:53:25Z +pr,189052,[dynamo] Model deque iterators with mutation-during-iteration detection,https://github.com/pytorch/pytorch/pull/189052,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:53:24Z +pr,188814,"Fix typos in comments and docstrings across ao, distributed, masked, nn, and sparse modules",https://github.com/pytorch/pytorch/pull/188814,ready for maintainer decision,45,"CI lane labels: ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:53:13Z +pr,188870,"Fix typos in comments, docstrings, and error messages",https://github.com/pytorch/pytorch/pull/188870,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:52:54Z +pr,180230,[TEST][CUDA][TF32] Update TF32 test decorators to use new fp32 precision path,https://github.com/pytorch/pytorch/pull/180230,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T20:48:25Z +pr,189160,Use SafePyObject in PyAnomalyMetadata and PySavedVariableHooks,https://github.com/pytorch/pytorch/pull/189160,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T20:39:09Z +pr,189123,cuda graphs: accept None && float graph-partition inputs,https://github.com/pytorch/pytorch/pull/189123,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:36:26Z +pr,189146,[MPS] softmax: non-last-dim and huge-row native Metal kernels (perf specializations),https://github.com/pytorch/pytorch/pull/189146,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T20:22:40Z +pr,188973,"Reapply ""Avoid redundant scalar-to-tensor conversion in `real`/`imag`setters on complex tensors (#188603)",https://github.com/pytorch/pytorch/pull/188973,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T20:20:39Z +pr,189162,[inductor] Use non-NaN-propagating fmax in softmax persistent reduction,https://github.com/pytorch/pytorch/pull/189162,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:19:52Z +pr,188681,"[Inductor] Use CUDA toolkit's libdevice by default, fallback to Triton's bundled if not found",https://github.com/pytorch/pytorch/pull/188681,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:19:31Z +pr,189251,[Dynamo] Store backend tags in the registry,https://github.com/pytorch/pytorch/pull/189251,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T20:18:59Z +pr,189163,Enable expandable_segments on Windows (port NVML driver API as prerequisite),https://github.com/pytorch/pytorch/pull/189163,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T20:01:21Z +pr,185206,Make minifier_tests more device agnostic,https://github.com/pytorch/pytorch/pull/185206,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T18:09:41Z +pr,189260,PERF: improve cuda/xpu Stream.__hash__,https://github.com/pytorch/pytorch/pull/189260,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T17:38:31Z +pr,185917,[Test] Refactor test_autocast.py: make behavior/cache tests device-agnostic,https://github.com/pytorch/pytorch/pull/185917,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T17:33:19Z +pr,188366,[functorch] Inline codegen'd epilogue fns directly into the runtime wrapper,https://github.com/pytorch/pytorch/pull/188366,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T16:57:44Z +pr,188367,[functorch] Extract _compile_and_exec_source into a generic codegen.py,https://github.com/pytorch/pytorch/pull/188367,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T16:57:43Z +pr,188376,[functorch] Add source_emit: reconstruct live values as standalone source,https://github.com/pytorch/pytorch/pull/188376,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T16:57:40Z +pr,182024,[docs] Fix typo in data page,https://github.com/pytorch/pytorch/pull/182024,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T14:17:29Z +pr,187457,[MPS] Migrate log_softmax to native Metal kernels,https://github.com/pytorch/pytorch/pull/187457,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T09:53:32Z +pr,184678,[5/N][Test] Migrate TestMeta device-agnostic skips to `@skipOps`,https://github.com/pytorch/pytorch/pull/184678,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T09:31:55Z +pr,187456,[MPS] Migrate softmax and _softmax_backward to native Metal kernels (core),https://github.com/pytorch/pytorch/pull/187456,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T09:10:26Z +pr,186345,[Test] Make test_comparison_utils.py device-generic,https://github.com/pytorch/pytorch/pull/186345,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T08:40:27Z +pr,182987,[Inductor] Parallel batched precompile in combo sequential autotune,https://github.com/pytorch/pytorch/pull/182987,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T08:18:47Z +pr,180209,XPU oneDNN: thread-local LRU cache for FP matmul primitives + post_ops deduplication,https://github.com/pytorch/pytorch/pull/180209,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T08:18:40Z +pr,181109,tools/stats: add script to generate test-times.json from local JUnit XML reports,https://github.com/pytorch/pytorch/pull/181109,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T04:49:48Z +pr,182377,[DTensor] SymmetricMemory-backed DTensors,https://github.com/pytorch/pytorch/pull/182377,ready for maintainer decision,45,"CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-08T04:44:20Z +pr,186273,[TEST][CUDA][TF32] Re-enable TF32 setter check test,https://github.com/pytorch/pytorch/pull/186273,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T03:39:16Z +pr,189011,include activation epilogue in GemmAndBiasParams::Signature(),https://github.com/pytorch/pytorch/pull/189011,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-08T02:58:16Z +pr,189184,[inductor][fb] Whole-graph batch_addmm->bmm fusion for independent same-shape addmm (#189184),https://github.com/pytorch/pytorch/pull/189184,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T23:17:51Z +pr,176830,feat: add polyfills for setitem/delitem op,https://github.com/pytorch/pytorch/pull/176830,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T22:49:45Z +pr,189177,[inductor] Drop unused cache_info from CacheCompiledArtifact._validate_and_unpack,https://github.com/pytorch/pytorch/pull/189177,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T21:08:24Z +pr,188999,Add RISC-V CPU pause hint to cpu_atomic_add_float spin loop,https://github.com/pytorch/pytorch/pull/188999,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T20:50:48Z +pr,182785,[Inductor] Fix FX wrapper codegen for symbolic floor/ceil offsets in ReinterpretView (#182785),https://github.com/pytorch/pytorch/pull/182785,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T20:12:15Z +pr,188250,Remove dead USE_ROCM block in CUDADeviceAssertionHost,https://github.com/pytorch/pytorch/pull/188250,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T20:00:42Z +pr,186245,[1/N] Optimize PyTorch CUDA build with LLVM BOLT,https://github.com/pytorch/pytorch/pull/186245,ready for maintainer decision,45,"CI lane labels: ciflow/binaries, ciflow/docker, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T19:56:04Z +pr,188946,[Bugfix][FSDP] Capture out-of-place redistribute() return value in optimizer state reconstruction,https://github.com/pytorch/pytorch/pull/188946,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T18:20:48Z +pr,188816,"[Metal][ops] remove nonzero INT_MAX limit, widen to 64-bit (#149325)",https://github.com/pytorch/pytorch/pull/188816,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T18:14:26Z +pr,188434,[dcp] Free staged dict on caller thread via weakref.finalize in async_save (#188434),https://github.com/pytorch/pytorch/pull/188434,ready for maintainer decision,45,"CI lane labels: ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T17:56:04Z +pr,187901,[TEST] Prevent PLATFORM_SUPPORTS_* from skipping tests on privateuse1 backends,https://github.com/pytorch/pytorch/pull/187901,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T17:45:17Z +pr,189089,[fmha_fwd][rocm] Extend CK ck_tile FMHA seqlen-pad guard to the GQA swap path,https://github.com/pytorch/pytorch/pull/189089,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T17:42:50Z +pr,189108,[Inductor] Bound exact index expressions for int32 indexing eligibility,https://github.com/pytorch/pytorch/pull/189108,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T17:42:13Z +pr,184718,[Test] Make test_testing.py device-agnostic for out-of-tree accelerators,https://github.com/pytorch/pytorch/pull/184718,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T17:17:47Z +pr,189028,[HOP] Moved branch clamping to user surface for switch,https://github.com/pytorch/pytorch/pull/189028,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T16:41:56Z +pr,188951,Add ROCm Docker release images,https://github.com/pytorch/pytorch/pull/188951,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T13:51:06Z +pr,184742,[Test] Make test_expanded_weights.py device-agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/184742,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T13:41:45Z +pr,189143,[Test] Make test_native_mha.py testcases device-agnostic,https://github.com/pytorch/pytorch/pull/189143,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T12:33:39Z +pr,189140,[Test] Update device type tests in TestVarlenAttention for device-generic,https://github.com/pytorch/pytorch/pull/189140,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T11:39:31Z +pr,188939,[profiler][cupti] Environment + per-kernel cycle GPU counters in the JSON export,https://github.com/pytorch/pytorch/pull/188939,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T05:51:28Z +pr,187898,"[profiler][cupti] Perfetto-native (.pftrace) export, full chrome-args parity",https://github.com/pytorch/pytorch/pull/187898,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T05:51:14Z +pr,188867,Add USE_KLEIDIAI_SME option to skip SME/SME2 kernels on CPUs without SME,https://github.com/pytorch/pytorch/pull/188867,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T05:50:25Z +pr,185486,[OpenReg] Add minimal implementation for inductor integration in openreg,https://github.com/pytorch/pytorch/pull/185486,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-07T05:41:40Z +pr,189101,[inductor] Fix data-dependent GuardOnDataDependentSymNode in get_fill_order,https://github.com/pytorch/pytorch/pull/189101,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-07T02:59:08Z +pr,188903,[HOP] Adding input mutation support for switch (2/2),https://github.com/pytorch/pytorch/pull/188903,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-06T23:06:37Z +pr,182687,Support generator arg in dirichlet and beta torch.distributions API,https://github.com/pytorch/pytorch/pull/182687,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T21:59:31Z +pr,188846,"Fix torch.pow docstring: scalar params are Number, not float",https://github.com/pytorch/pytorch/pull/188846,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T20:41:24Z +pr,138068,Add TORCH_CHECK_INDEX in convert_indices_from_coo_to_csr_cpu,https://github.com/pytorch/pytorch/pull/138068,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T20:13:54Z +pr,126348,[AOTAutograd] tweak min-cut partitioner to avoid saving softmax output,https://github.com/pytorch/pytorch/pull/126348,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T20:11:52Z +pr,188976,[HOP] Inductor lowering for switch (1/2),https://github.com/pytorch/pytorch/pull/188976,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-06T19:21:25Z +pr,188947,[Bugfix][DTensor] Fix wrong shard offsets for uneven DTensor shards in TP-FSDP metadata,https://github.com/pytorch/pytorch/pull/188947,ready for maintainer decision,45,"CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-06T19:01:04Z +pr,188885,[DataLoader] Clean up stale Py2 comments and pickle-error warning,https://github.com/pytorch/pytorch/pull/188885,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T18:55:40Z +pr,188874,Enable XPUGraph native recording mode,https://github.com/pytorch/pytorch/pull/188874,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T18:47:12Z +pr,188804,[MPS] Add solo large-tensor allocator + torch.mps.set_allocator_settings(),https://github.com/pytorch/pytorch/pull/188804,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T18:45:20Z +pr,188944,[NCCL] try to remove the redundant copy in functional collectives,https://github.com/pytorch/pytorch/pull/188944,ready for maintainer decision,45,CI lane labels: ciflow/torchtitan; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T16:59:48Z +pr,188761,MSVC compiler flag change for JIT cpp extensions on Windows,https://github.com/pytorch/pytorch/pull/188761,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T16:39:28Z +pr,188823,[Inductor][Host-side TMA] Cache host-side TMA descriptors across launches,https://github.com/pytorch/pytorch/pull/188823,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-06T15:48:40Z +pr,183755,[MPS] Enable bf16/fp16 ConvTranspose3D by removing the gating,https://github.com/pytorch/pytorch/pull/183755,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T14:32:58Z +pr,185881,[Test] Refactor test/test_reductions.py,https://github.com/pytorch/pytorch/pull/185881,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T13:43:34Z +pr,188156,[MPS] reduce: share the two-pass value_reduction kernel for sum and value,https://github.com/pytorch/pytorch/pull/188156,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T13:39:46Z +pr,182747,Resolve PYTHON_SIX_SOURCE_DIR from installed `six` to skip pypi download,https://github.com/pytorch/pytorch/pull/182747,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T09:49:59Z +pr,187317,[pr-review skill] Add device portability checks for shared accelerator code,https://github.com/pytorch/pytorch/pull/187317,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T09:39:58Z +pr,188977,Fix typos in docstrings and comments across torch modules,https://github.com/pytorch/pytorch/pull/188977,ready for maintainer decision,45,"CI lane labels: ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-06T07:50:58Z +pr,182582,Add addmm quantization graph passes for PT2 readiness benchmarks (#182582),https://github.com/pytorch/pytorch/pull/182582,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-06T01:09:42Z +pr,182159,Fix List isSubtypeExt,https://github.com/pytorch/pytorch/pull/182159,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-05T22:49:44Z +pr,179485,Add torchbind custom class inheritance support (#179485),https://github.com/pytorch/pytorch/pull/179485,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-05T20:49:08Z +pr,185788,Support 64-bit indexing in channels_last bilinear upsampling,https://github.com/pytorch/pytorch/pull/185788,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-05T18:30:10Z +pr,188500,Allow pre-split pipeline microbatches,https://github.com/pytorch/pytorch/pull/188500,ready for maintainer decision,45,"CI lane labels: ciflow/torchtitan, ciflow/trunk; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-05T17:24:13Z +pr,181622,[Inductor][NVGEMM] Refactor rendering,https://github.com/pytorch/pytorch/pull/181622,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-04T21:41:47Z +pr,181902,[Inductor][NVGEMM] Remove duplicate nvgemm_max_profiling config,https://github.com/pytorch/pytorch/pull/181902,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-04T21:37:22Z +pr,181903,[Inductor][NVGEMM] Avoid double iteration over kernel cache during choice enumeration,https://github.com/pytorch/pytorch/pull/181903,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-04T21:37:06Z +pr,181910,[Inductor][NVGEMM] Link tracking issue in supplement configs comment,https://github.com/pytorch/pytorch/pull/181910,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-04T21:32:46Z +pr,182061,Document low-memory state_dict loading with meta device,https://github.com/pytorch/pytorch/pull/182061,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-04T18:50:04Z +pr,182053,Partition load_state_dict child keys once,https://github.com/pytorch/pytorch/pull/182053,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-04T18:50:02Z +pr,181862,collect_env: add uv support,https://github.com/pytorch/pytorch/pull/181862,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-04T12:16:34Z +pr,175083,[pytree] add another simplified pytree module `torch.pytree`,https://github.com/pytorch/pytorch/pull/175083,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-04T12:10:37Z +pr,182340,Bump Windows CPU trunk test shards from 4 to 5,https://github.com/pytorch/pytorch/pull/182340,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-04T01:04:22Z +pr,187556,[MPS] Migrate smooth_l1_loss and huber_loss to native Metal kernels,https://github.com/pytorch/pytorch/pull/187556,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-03T22:25:26Z +pr,182426,Enable FP32 emulation via BF16 in cuSOLVER for qr and lu,https://github.com/pytorch/pytorch/pull/182426,ready for maintainer decision,45,CI lane labels: ciflow/b200; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-03T13:51:18Z +pr,179212,feat(pipelining): add configurable auto-partitioning algorithm for pipeline parallelism,https://github.com/pytorch/pytorch/pull/179212,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-03T13:20:31Z +pr,188660,[Inductor] Enable mix-order reduction for wide reductions + tests,https://github.com/pytorch/pytorch/pull/188660,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-03T11:52:52Z +pr,187557,[MPS] Migrate binary_cross_entropy to native Metal kernels,https://github.com/pytorch/pytorch/pull/187557,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-03T10:54:23Z +pr,188760,Guard user stream compile tests with xfailIfNoAcceleratorTriton,https://github.com/pytorch/pytorch/pull/188760,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-03T10:11:34Z +pr,181849,Enable FP32 emulation via BF16 in cuSOLVER,https://github.com/pytorch/pytorch/pull/181849,ready for maintainer decision,45,CI lane labels: ciflow/b200; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-03T07:39:16Z +pr,188379,[HOP] None input handling for scan,https://github.com/pytorch/pytorch/pull/188379,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-03T07:14:35Z +pr,182919,Add forward-only cuDNN FP8 per-tensor attention,https://github.com/pytorch/pytorch/pull/182919,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-03T01:01:36Z +pr,188126,[inductor] Enable cudagraph partition inside invoke_subgraph subgraphs,https://github.com/pytorch/pytorch/pull/188126,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-02T22:20:37Z +pr,188613,[MPS] copy for contiguous same-dtype using a compute kernel,https://github.com/pytorch/pytorch/pull/188613,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-02T16:07:21Z +pr,188315,[HOP] Autograd implementation for switch,https://github.com/pytorch/pytorch/pull/188315,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-02T10:36:24Z +pr,188348,[HOP] Fixes for zero-length scan dim,https://github.com/pytorch/pytorch/pull/188348,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-02T07:59:29Z +pr,188555,Fix rocm lazy context initialization in test_set_device_0,https://github.com/pytorch/pytorch/pull/188555,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-02T00:57:00Z +pr,188606,Relax complex tensor consistency tolerances for CUDA binary nonlinear ops,https://github.com/pytorch/pytorch/pull/188606,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-02T00:56:01Z +pr,183716,Validate dim in cummax/cummin meta for empty tensors,https://github.com/pytorch/pytorch/pull/183716,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-01T17:12:31Z +pr,188333,[MPS] prod: migrate to the shared value_reduction kernel + shape specializations,https://github.com/pytorch/pytorch/pull/188333,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-01T16:44:40Z +pr,151218,Implement MKLGenerator,https://github.com/pytorch/pytorch/pull/151218,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-01T12:19:18Z +pr,188329,Optimize all-gather copy-in with a flat parameter buffer,https://github.com/pytorch/pytorch/pull/188329,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-07-01T09:36:32Z +pr,186603,Exposing MiMalloc options and stats,https://github.com/pytorch/pytorch/pull/186603,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-01T08:34:53Z +pr,188626,build-almalinux-images: migrate to OSDC via remote BuildKit,https://github.com/pytorch/pytorch/pull/188626,ready for maintainer decision,45,CI lane labels: ciflow/docker; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-01T03:44:03Z +pr,188625,binary-docker-build: add opt-in remote BuildKit (OSDC) mode,https://github.com/pytorch/pytorch/pull/188625,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-07-01T02:22:23Z +pr,188530,optim: remove redundant Adafactor assert,https://github.com/pytorch/pytorch/pull/188530,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-30T22:49:40Z +pr,188402,Fix matrix_exp crash on non-contiguous (channels-last) inputs,https://github.com/pytorch/pytorch/pull/188402,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-30T18:30:39Z +pr,186617,Added profiler record for THPVariable_dealloc,https://github.com/pytorch/pytorch/pull/186617,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-30T13:54:47Z +pr,188407,[Windows] Add triton_windows commit pin for Windows nightly builds,https://github.com/pytorch/pytorch/pull/188407,ready for maintainer decision,45,"CI lane labels: ciflow/docker, ciflow/inductor; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-29T22:39:43Z +pr,188364,Allow from_blob / TensorMaker to attach BackendMeta at construction,https://github.com/pytorch/pytorch/pull/188364,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T21:14:50Z +pr,179020,Update third-party/rocm_composable_kernel to f1746955,https://github.com/pytorch/pytorch/pull/179020,ready for maintainer decision,45,CI lane labels: ciflow/rocm-mi300; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T20:13:31Z +pr,188164,Fix THPDTypeInfo_compare reading off the end of non-finfo/iinfo objects,https://github.com/pytorch/pytorch/pull/188164,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T19:12:40Z +pr,188433,linear_cross_entropy: test and document bias with probability targets,https://github.com/pytorch/pytorch/pull/188433,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T18:51:42Z +pr,182030,[ROCm] Make AOTriton bundling optional via `BUILD_AOTRITON_INTO_WHEEL` flag,https://github.com/pytorch/pytorch/pull/182030,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T17:34:08Z +pr,182040,Download magma & sccache install scripts over https instead of aws s3,https://github.com/pytorch/pytorch/pull/182040,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T17:23:04Z +pr,188293,Allow tuple subclasses for run_backward inputs,https://github.com/pytorch/pytorch/pull/188293,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T16:27:54Z +pr,181996,[torch.package] Register package modules in sys.modules for Python 3.12 dataclass compat,https://github.com/pytorch/pytorch/pull/181996,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-29T10:04:28Z +pr,187999,Fix uninitialized memory warnings in CPU linear algebra kernels,https://github.com/pytorch/pytorch/pull/187999,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-28T20:29:44Z +pr,188031,Fix 64-bit indexing in spatial softmax backward,https://github.com/pytorch/pytorch/pull/188031,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-28T20:23:09Z +pr,181330,[Profiler] Add a skip_build_tree metadata field to Kineto events to m…,https://github.com/pytorch/pytorch/pull/181330,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-27T22:50:18Z +pr,176584,[RNG] Decouple Philox subsequence from hardware topology for cross-platform consistency,https://github.com/pytorch/pytorch/pull/176584,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-27T17:50:48Z +pr,188191,Add sm_121a (NVIDIA GB10/DGX Spark) to default CUDA arch list,https://github.com/pytorch/pytorch/pull/188191,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-27T17:12:14Z +pr,186881,nn/utils: skip no-op .to(first_device) in _get_total_norm,https://github.com/pytorch/pytorch/pull/186881,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-27T10:32:51Z +pr,188252,Fix ROCm round-robin counter reset in initSingleStream,https://github.com/pytorch/pytorch/pull/188252,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-27T01:02:30Z +pr,181030,[MPS] Migrate SDPA from MPSGraph to Metal,https://github.com/pytorch/pytorch/pull/181030,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-26T21:53:45Z +pr,176317,[MPS] Leveraging unified memory memcpy instead of blit encoding for tensor device transfers,https://github.com/pytorch/pytorch/pull/176317,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-26T20:09:48Z +pr,188263,(fix) NNPACK PeachPy air gapped build,https://github.com/pytorch/pytorch/pull/188263,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-26T19:26:34Z +pr,188125,[AI Codemod][PerfAICT-General] perf: Special case non-symbolic set_sizes_contiguous,https://github.com/pytorch/pytorch/pull/188125,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-26T16:45:01Z +pr,181285,"[TEST][ARM] Remove Pointer Authentication Code from address, fall back to `dladdr_lookup` where possible",https://github.com/pytorch/pytorch/pull/181285,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-26T16:17:38Z +pr,188045,[MPS] better memory allocator,https://github.com/pytorch/pytorch/pull/188045,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-26T12:09:36Z +pr,180758,Remove dead TORCH_HIP_VERSION guards,https://github.com/pytorch/pytorch/pull/180758,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-26T06:09:24Z +pr,186145,"Spelling fix: Change ""structeur"" to ""structure"" in Lumen CLI VLLM test",https://github.com/pytorch/pytorch/pull/186145,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-25T22:07:38Z +pr,178037,[MPS] Raise clear error when MPS is used in forked subprocess,https://github.com/pytorch/pytorch/pull/178037,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-25T16:24:00Z +pr,187813,Generalize memory viz and cycle detection,https://github.com/pytorch/pytorch/pull/187813,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-24T19:32:34Z +pr,188089,[forward AD] Remove obsolete __debug__ guard when loading JVP decompositions,https://github.com/pytorch/pytorch/pull/188089,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-24T18:51:53Z +pr,186624,Fix CLANGTIDY errors in c10/cuda,https://github.com/pytorch/pytorch/pull/186624,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-24T18:46:28Z +pr,187713,Fix B008 flake8-bugbear violations (function calls in default arguments),https://github.com/pytorch/pytorch/pull/187713,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-24T18:32:01Z +pr,187687,fix: aten embed bwd thread sync before proceeding to next batch,https://github.com/pytorch/pytorch/pull/187687,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-24T03:00:15Z +pr,180584,Test,https://github.com/pytorch/pytorch/pull/180584,ready for maintainer decision,45,"CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-23T22:02:02Z +pr,178069,[docs] Add examples to reduce_scatter documentation,https://github.com/pytorch/pytorch/pull/178069,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T15:36:42Z +pr,178068,[docs] Add comprehensive introduction to PyTorch Elastic documentation,https://github.com/pytorch/pytorch/pull/178068,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T15:36:40Z +pr,187881,Fix margin docstring rendering in HingeEmbeddingLoss,https://github.com/pytorch/pytorch/pull/187881,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:26:10Z +pr,187816,Clean up mentions of deprecated nvidia-docker in README files,https://github.com/pytorch/pytorch/pull/187816,ready for maintainer decision,45,CI lane labels: ciflow/docker; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:23:49Z +pr,187847,fix(nn.utils.prune): don't crash adding a method to an empty PruningContainer,https://github.com/pytorch/pytorch/pull/187847,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:22:49Z +pr,187796,Add native MPS SVD path,https://github.com/pytorch/pytorch/pull/187796,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:21:30Z +pr,187800,Verify Magma archive before extraction,https://github.com/pytorch/pytorch/pull/187800,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:21:22Z +pr,187742,Fix calculate_out_dimension using stride[0] for the width dimension,https://github.com/pytorch/pytorch/pull/187742,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:21:17Z +pr,187799,Fix ROCm libdrm readlink buffer bounds,https://github.com/pytorch/pytorch/pull/187799,ready for maintainer decision,45,CI lane labels: ciflow/docker; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:20:51Z +pr,187769,[docs] Clarify Softplus overflow behavior for reduced-precision dtypes,https://github.com/pytorch/pytorch/pull/187769,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T14:00:44Z +pr,187717,[CUDA] solve_triangular: fix silently wrong batched results and slowness in large-RHS workaround,https://github.com/pytorch/pytorch/pull/187717,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T13:58:43Z +pr,187921,Support freeing blocks with stream uses during graph capture,https://github.com/pytorch/pytorch/pull/187921,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T13:35:14Z +pr,187694,Unpack Python autograd saved tensors in reverse order,https://github.com/pytorch/pytorch/pull/187694,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-23T13:12:02Z +pr,184398,Enable 3.15 in CI,https://github.com/pytorch/pytorch/pull/184398,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-22T23:27:15Z +pr,185446,Adding a test case using fx.wrap to test whether TLX user-defined kernel is visible in the inductor output (#185301),https://github.com/pytorch/pytorch/pull/185446,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-22T20:20:57Z +pr,178373,docs(nn): clarify return type for MaxPool1d/2d/3d when return_indices=True,https://github.com/pytorch/pytorch/pull/178373,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-22T15:24:46Z +pr,178568,Add support for activation checkpointing with nested data structures,https://github.com/pytorch/pytorch/pull/178568,ready for maintainer decision,45,CI lane labels: ciflow/torchtitan; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-22T11:27:58Z +pr,175512,[kineto] Generate unique group_trace_id instead of reusing trace_id (#175512),https://github.com/pytorch/pytorch/pull/175512,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-22T08:34:02Z +pr,162454,[xpu][feature] Integrate OneDNN SDPA training forward/backward into XPU OVERRIDEABLE Backend,https://github.com/pytorch/pytorch/pull/162454,ready for maintainer decision,45,CI lane labels: ciflow/binaries_wheel; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-22T06:57:10Z +pr,180171,Fix empty changed-files handling in Filepath heuristic,https://github.com/pytorch/pytorch/pull/180171,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-21T21:58:34Z +pr,175425,Improve error message for CPU-CUDA copy during graph capture,https://github.com/pytorch/pytorch/pull/175425,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-21T15:14:30Z +pr,180788,"Use C++20 concepts/requires in Tensor, TensorBase, and IValue",https://github.com/pytorch/pytorch/pull/180788,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-21T06:27:52Z +pr,180988,[fbandroid][buckconfig removal] removal of last files referencing tools/build_defs/android/build_mode_defs.bzl,https://github.com/pytorch/pytorch/pull/180988,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-20T17:57:38Z +pr,187333,[CUDA] linalg: overlap looped cuSOLVER LU (getrf) across auxiliary streams,https://github.com/pytorch/pytorch/pull/187333,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-20T15:58:08Z +pr,183811,[docs] Document 10 public APIs in torch.nn.parallel submodules,https://github.com/pytorch/pytorch/pull/183811,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-20T06:31:10Z +pr,180686,test: relax error tolerances for complex64 tests with tan and tanh,https://github.com/pytorch/pytorch/pull/180686,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-19T20:55:26Z +pr,180187,add OneCycleLR degenerate phase cases,https://github.com/pytorch/pytorch/pull/180187,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-19T17:18:03Z +pr,175706,add C++20 bit operations and contains to C10,https://github.com/pytorch/pytorch/pull/175706,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-18T15:46:18Z +pr,185930,[Elastic][Distributed] Flush dirty rendezvous state immediately,https://github.com/pytorch/pytorch/pull/185930,ready for maintainer decision,45,CI lane labels: ciflow/torchtitan; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-18T14:33:39Z +pr,178088,Use more structure binding,https://github.com/pytorch/pytorch/pull/178088,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-18T08:15:24Z +pr,180527,Add thread_time_ns and idle_pct to dynamo_timed chromium events (#180527),https://github.com/pytorch/pytorch/pull/180527,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-17T21:25:15Z +pr,187566,[docs] Improve autograd anomaly mode documentation,https://github.com/pytorch/pytorch/pull/187566,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-17T17:17:48Z +pr,179747,Add rotary positional embeddings support to nn.MultiheadAttention,https://github.com/pytorch/pytorch/pull/179747,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-17T03:48:39Z +pr,187410,[distributed] Use node.target instead of node.name in apply_manual_re…,https://github.com/pytorch/pytorch/pull/187410,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-16T21:31:07Z +pr,187245,update backend parametrization for varlen attn,https://github.com/pytorch/pytorch/pull/187245,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-16T21:26:56Z +pr,179700,[torch][inductor] Add RAIIAtenRecordFunctionHandle profiling to remaining uncovered kernel paths (#179700),https://github.com/pytorch/pytorch/pull/179700,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-16T20:31:36Z +pr,185790,Fix torch.nn.functional.gelu returning NaN for infinite inputs,https://github.com/pytorch/pytorch/pull/185790,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-16T09:23:20Z +pr,179698,[torch][inductor] Fix numel variable scoping with enable_kernel_profile (#179698),https://github.com/pytorch/pytorch/pull/179698,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-16T03:48:36Z +pr,187192,Protect GraphPassRegistry map with shared_mutex against concurrent access (#187192),https://github.com/pytorch/pytorch/pull/187192,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-15T16:49:09Z +pr,175476,[inductor] Hoist output buffer allocations into prior CUDAGraph partition,https://github.com/pytorch/pytorch/pull/175476,ready for maintainer decision,45,"CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-15T14:56:19Z +pr,173182,Updated foreach_norm to accept dim and keepdim arguments.,https://github.com/pytorch/pytorch/pull/173182,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-15T12:32:54Z +pr,177147,[FSDP2] Remove redundant event wait for all_gather_stream,https://github.com/pytorch/pytorch/pull/177147,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-15T03:33:33Z +pr,176936,Add driver api guard to cuda utils,https://github.com/pytorch/pytorch/pull/176936,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-14T20:03:45Z +pr,174605,make Store timeout_ atomic for thread safety.,https://github.com/pytorch/pytorch/pull/174605,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-14T20:03:40Z +pr,179594,[CUDA] Use #pragma unroll in Normaization.cuh,https://github.com/pytorch/pytorch/pull/179594,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-14T16:59:33Z +pr,179399,Improve GELU documentation formula,https://github.com/pytorch/pytorch/pull/179399,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-14T16:59:33Z +pr,179000,Add type conversion test for IValue: implements TODO(gmagogsfm),https://github.com/pytorch/pytorch/pull/179000,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-14T16:59:26Z +pr,179372,[Inductor] Fix view_as_complex stride requirement in Inductor backward by forcing contiguous layout,https://github.com/pytorch/pytorch/pull/179372,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-14T16:03:23Z +pr,187126,[c10] Remove deprecated CUDAAllocatorConfig forwarders,https://github.com/pytorch/pytorch/pull/187126,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-13T02:18:03Z +pr,187097,[BE] Fix B007 unused loop variables,https://github.com/pytorch/pytorch/pull/187097,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-12T17:12:17Z +pr,187146,test all named dims test with dynamic sepcs using automatic conversion,https://github.com/pytorch/pytorch/pull/187146,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-12T07:28:34Z +pr,187135,[xpu] Remove deprecated at::xpu CachingHostAllocator forwarders,https://github.com/pytorch/pytorch/pull/187135,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-12T05:51:32Z +pr,186046,[MPS] Remove dead xfail/tolerance keys from MPS test infra,https://github.com/pytorch/pytorch/pull/186046,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-12T05:51:16Z +pr,187006,Avoid duplicate sm_103 family-specific CUDA codegen,https://github.com/pytorch/pytorch/pull/187006,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-11T17:40:48Z +pr,181292,[inductor] Autotune cache stability and perf improvements (#181292),https://github.com/pytorch/pytorch/pull/181292,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-11T16:46:20Z +pr,178924,[inductor] Fix FP8 dtype detection in _get_estimated_runtime,https://github.com/pytorch/pytorch/pull/178924,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-11T16:34:57Z +pr,186924,Also search $MKLROOT/lib in Linux for mkl wheel and install_mkl.sh,https://github.com/pytorch/pytorch/pull/186924,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-11T10:08:12Z +pr,186232,Add opt-in cross-stream segment reclaim to CUDA allocator,https://github.com/pytorch/pytorch/pull/186232,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-11T08:32:39Z +pr,184190,[fx]: Add SymNode helper to get raw expr,https://github.com/pytorch/pytorch/pull/184190,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-11T08:23:21Z +pr,183343,dynamo: trace torch.cuda.Event.record / Event.wait under fx proxy mode,https://github.com/pytorch/pytorch/pull/183343,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/torchtitan; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-10T19:15:57Z +pr,186887,Update test_mm_sparse_first_NT/NN input generation,https://github.com/pytorch/pytorch/pull/186887,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-10T18:05:18Z +pr,178545,[MPS] Add native ops fast path for SDPA on long sequences (~2x speedup),https://github.com/pytorch/pytorch/pull/178545,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-10T11:45:28Z +pr,183852,ci: declare workflow-level `contents: read` on 8 workflows,https://github.com/pytorch/pytorch/pull/183852,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-10T08:09:37Z +pr,178960,profiler: fix USE_KINETO=OFF build failure (unconditional ActivityType.h include),https://github.com/pytorch/pytorch/pull/178960,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-10T05:03:49Z +pr,185507,Pre-compute igates for XPU LSTM to reduce per-step kernel launches,https://github.com/pytorch/pytorch/pull/185507,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-09T14:02:36Z +pr,184585,[MPS] Wire SDPBackend dispatch for scaled_dot_product_attention,https://github.com/pytorch/pytorch/pull/184585,ready for maintainer decision,45,"CI lane labels: ciflow/inductor, ciflow/mps; open non-draft PR has no obvious automated blocker",review diff and decide acceptance path,2026-06-09T01:51:15Z +pr,184559,[Test]Refactor test/test_reductions.py to be device generic,https://github.com/pytorch/pytorch/pull/184559,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-08T17:31:19Z +pr,186568,Add expected-failure test for ConstantLR factor=0,https://github.com/pytorch/pytorch/pull/186568,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-08T14:08:52Z +pr,186307,Add contains() to ska::flat_hash_map/set,https://github.com/pytorch/pytorch/pull/186307,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-08T14:06:30Z +pr,186565,Support parametrized modules in functional_call,https://github.com/pytorch/pytorch/pull/186565,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-08T13:51:20Z +pr,186559,Remove deprecated pytree_to_str and str_to_pytree functions,https://github.com/pytorch/pytorch/pull/186559,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-08T13:41:45Z +pr,186473,[accelerator] Support multiple private backends,https://github.com/pytorch/pytorch/pull/186473,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-08T13:36:52Z +pr,186239,[ROCm][CI] Add linux.rocm.mi350.docker-cache to docker-cache-rocm,https://github.com/pytorch/pytorch/pull/186239,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-08T13:34:18Z +pr,185901,[Test] Refactor test_schema_check.py to be device-agnostic,https://github.com/pytorch/pytorch/pull/185901,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-05T12:51:21Z +pr,185648,[Test] Make sigmoid_add extension test device-agnostic in test_cpp_extensions_aot.py,https://github.com/pytorch/pytorch/pull/185648,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-05T07:47:34Z +pr,185184,[Test][distributed] Refactor some distributed test helpers to be device-agnostic,https://github.com/pytorch/pytorch/pull/185184,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-04T06:55:13Z +pr,185883,Simplify ostream formatting with fmt::join and fmt::print,https://github.com/pytorch/pytorch/pull/185883,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-03T22:45:38Z +pr,185993,Fix SM121 compute capability corruption in CMake GPU arch detection,https://github.com/pytorch/pytorch/pull/185993,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-03T22:38:53Z +pr,185906,add type hint to seed argument in torch.random.manual_seed,https://github.com/pytorch/pytorch/pull/185906,ready for maintainer decision,45,CI lane labels: ciflow/torchtitan; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-03T22:18:32Z +pr,186045,Merge identical elif branches and inline single-use locals in torchgen,https://github.com/pytorch/pytorch/pull/186045,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-03T22:01:20Z +pr,186110,Remove test_transformers from windows blocklist,https://github.com/pytorch/pytorch/pull/186110,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-03T22:01:09Z +pr,184548,[Test]Refactor test/test_sort_and_select.py to be device generic,https://github.com/pytorch/pytorch/pull/184548,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-03T09:35:44Z +pr,183855,[c10d] Surface operational metadata in WorkNCCL operator<<,https://github.com/pytorch/pytorch/pull/183855,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-03T04:34:32Z +pr,178590,Correct type hints for ConvNd subclasses,https://github.com/pytorch/pytorch/pull/178590,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-02T09:19:15Z +pr,185758,Tidy MPSGuardImpl: name capability bitmask and fix clang-tidy nits,https://github.com/pytorch/pytorch/pull/185758,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-02T03:10:57Z +pr,171313,Support torch.accelerator.Graph on CUDA,https://github.com/pytorch/pytorch/pull/171313,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-02T02:02:32Z +pr,181833,Add boolean evaluation for LazyVal wrapping str,https://github.com/pytorch/pytorch/pull/181833,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-01T17:59:36Z +pr,180954,Add value property to LazyVal,https://github.com/pytorch/pytorch/pull/180954,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-01T17:59:35Z +pr,185743,Mark CUDA kernel helpers static,https://github.com/pytorch/pytorch/pull/185743,ready for maintainer decision,45,CI lane labels: ciflow/trunk; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-06-01T16:44:09Z +pr,184859,Deprecate c10::ssize as alias for std::ssize,https://github.com/pytorch/pytorch/pull/184859,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-30T03:13:49Z +pr,185125,Remove unused caffe2/utils/string_utils,https://github.com/pytorch/pytorch/pull/185125,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-30T01:29:45Z +pr,184142,Fix LPPool signed fractional inputs,https://github.com/pytorch/pytorch/pull/184142,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-29T17:19:31Z +pr,182289,[MPS] Pinned memory - Add pinned memory allocator,https://github.com/pytorch/pytorch/pull/182289,ready for maintainer decision,45,CI lane labels: ciflow/mps; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-29T12:36:07Z +pr,183870,[stats] Handle alternate test-reports path in backfill_test_jsons_while_running,https://github.com/pytorch/pytorch/pull/183870,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-29T05:58:43Z +pr,184426,Support getting CUDA patch version,https://github.com/pytorch/pytorch/pull/184426,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-28T15:24:58Z +pr,152159,Add dynamo config to HOP-ify context managers,https://github.com/pytorch/pytorch/pull/152159,ready for maintainer decision,45,CI lane labels: ciflow/inductor; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-28T12:46:52Z +pr,183738,Multi-warp TMA scatter_add via PerWarpTmaPipeline,https://github.com/pytorch/pytorch/pull/183738,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-27T12:57:44Z +pr,183501,[PyTorch][Pinned Memory] Fix stuck event blocking in CachingHostAllocator,https://github.com/pytorch/pytorch/pull/183501,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-27T06:42:14Z +pr,183739,TMA scatter_add: 2D-tile multi-row path for small-N shapes,https://github.com/pytorch/pytorch/pull/183739,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-26T16:55:12Z +pr,185137,[fx][_library] Honor wrapped op's schema.is_mutable in is_impure's auto_functionalized HOP branch (#185137),https://github.com/pytorch/pytorch/pull/185137,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-26T15:09:04Z +pr,184866,[distributed] Handle empty inputs and kwargs in _to_kwargs,https://github.com/pytorch/pytorch/pull/184866,ready for maintainer decision,45,CI lane labels: ciflow/torchtitan; open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-26T13:37:56Z +pr,183752,[BE]: Revert back to upstream uv action to solve rate-limiting,https://github.com/pytorch/pytorch/pull/183752,ready for maintainer decision,45,open non-draft PR has no obvious automated blocker,review diff and decide acceptance path,2026-05-26T04:42:21Z +pr,178082,Fix ninja workdir collisions in BuildExtension,https://github.com/pytorch/pytorch/pull/178082,waiting on contributor,45,review decision requests changes; PR closes or references open issue numbers; PR has not updated for at least 45 days,wait for contributor update,2026-05-21T03:55:36Z +pr,179719,Skip div-to-mul-reciprocal when division_rounding is enabled,https://github.com/pytorch/pytorch/pull/179719,waiting on contributor,45,"review decision requests changes; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-05-15T13:39:14Z +pr,172003,Add __getitems__ to TensorDataset for faster DataLoader iteration,https://github.com/pytorch/pytorch/pull/172003,waiting on contributor,45,review decision requests changes; PR closes or references open issue numbers; PR has not updated for at least 45 days,wait for contributor update,2026-05-14T21:34:37Z +pr,173072,s390x: rework ZVECTOR data into 16-byte-sized vectors,https://github.com/pytorch/pytorch/pull/173072,waiting on contributor,45,"review decision requests changes; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/s390, ciflow/torchtitan",wait for contributor update,2026-05-11T23:32:03Z +pr,181096,inductor/flex_attention: fix fake block-mask generator for autotuner,https://github.com/pytorch/pytorch/pull/181096,waiting on contributor,45,"review decision requests changes; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-11T21:25:32Z +pr,172991,Fix torch.compile for jagged nested tensors created inside compiled functions,https://github.com/pytorch/pytorch/pull/172991,waiting on contributor,45,review decision requests changes; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for contributor update,2026-04-16T13:38:42Z +issue,150199,`asarray`: device does not propagate from input to output after `set_default_device`,https://github.com/pytorch/pytorch/issues/150199,has linked PR,44,"open PR references this issue; component/oncall labels: module: python array api, module: python frontend",review linked PR state before touching issue,2026-07-08T21:44:27Z +issue,141287,MPS operator coverage tracking issue (2.6+ version),https://github.com/pytorch/pytorch/issues/141287,has linked PR,44,open PR references this issue; component/oncall labels: module: mps,review linked PR state before touching issue,2026-07-08T20:26:39Z +issue,189271,[inductor] KeyError in Scheduler.compute_ancestors: self-edge from realizing a reshaped in-place-mutated buffer,https://github.com/pytorch/pytorch/issues/189271,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: dynamic shapes, module: inductor, module: pt2-dispatcher",review linked PR state before touching issue,2026-07-08T15:25:09Z +issue,129131,"Torch compile initialises CUDA context, even compiling CPU functions",https://github.com/pytorch/pytorch/issues/129131,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-07-08T13:42:20Z +issue,187811,DISABLED test_combo_kernel_dynamic_scale_rblock (__main__.ComboKernelTests),https://github.com/pytorch/pytorch/issues/187811,has linked PR,44,open PR references this issue; component/oncall labels: module: xpu,review linked PR state before touching issue,2026-07-08T02:07:23Z +issue,187810,DISABLED test_combo_kernel_dynamic_scale_rblock (__main__.ComboKernelTestsPerSubkernelBlocks),https://github.com/pytorch/pytorch/issues/187810,has linked PR,44,open PR references this issue; component/oncall labels: module: xpu,review linked PR state before touching issue,2026-07-08T02:07:08Z +issue,186796,weakrefs don't when tensor swap is used,https://github.com/pytorch/pytorch/issues/186796,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: guards, oncall: pt2",review linked PR state before touching issue,2026-07-07T20:58:51Z +issue,167725,`torch._inductor.utils.fresh_cache` fails due to persistant process pool,https://github.com/pytorch/pytorch/issues/167725,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-07-07T18:19:40Z +issue,189157,GHA stats artifact downloader does not skip mismatched workflow run attempts,https://github.com/pytorch/pytorch/issues/189157,has linked PR,44,open PR references this issue; component/oncall labels: module: ci,review linked PR state before touching issue,2026-07-07T14:58:03Z +issue,189133,DISABLED test_vmap (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/189133,has linked PR,44,"open PR references this issue; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-07-07T13:25:26Z +issue,186867,DISABLED test_group_norm_backward_large_input_float16_trigger_32bit_overflow_True_mps_float16 (__main__.TestConsistencyMPS),https://github.com/pytorch/pytorch/issues/186867,has linked PR,44,"open PR references this issue; component/oncall labels: module: flaky-tests, module: macos, module: mps",review linked PR state before touching issue,2026-07-07T07:52:44Z +issue,187093,[inductor] baddbmm with broadcast bias: autotune never considers the bmm + pointwise-bias decomposition (1.3-1.45x left on the table at MoE expert-GEMM shapes),https://github.com/pytorch/pytorch/issues/187093,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2026-07-07T05:17:10Z +issue,189118,tools/testing: upload test-jsons artifact from the generated JSON zip,https://github.com/pytorch/pytorch/issues/189118,has linked PR,44,"open PR references this issue; component/oncall labels: module: ci, module: tests",review linked PR state before touching issue,2026-07-07T02:18:00Z +issue,186605,DISABLED test_group_norm_backward_large_input_float16_trigger_32bit_overflow_False_mps_float16 (__main__.TestConsistencyMPS),https://github.com/pytorch/pytorch/issues/186605,has linked PR,44,"open PR references this issue; component/oncall labels: module: flaky-tests, module: macos, module: mps",review linked PR state before touching issue,2026-07-07T01:15:52Z +issue,188773,DISABLED test_cow_inputs_accepted_by_override (__main__.TestBmmOuterProduct),https://github.com/pytorch/pytorch/issues/188773,has linked PR,44,open PR references this issue; component/oncall labels: module: xpu,review linked PR state before touching issue,2026-07-07T01:04:53Z +issue,189106,`get_signature_for_torch_op` reports incorrect `return_annotation` for `tuple`s,https://github.com/pytorch/pytorch/issues/189106,has linked PR,44,open PR references this issue; component/oncall labels: module: fx,review linked PR state before touching issue,2026-07-06T22:42:29Z +issue,188680,[test] OpInfo gap: allclose-based compile-vs-eager checks are blind to signed-zero and signed-infinity divergences,https://github.com/pytorch/pytorch/issues/188680,has linked PR,44,"open PR references this issue; component/oncall labels: module: tests, oncall: pt2",review linked PR state before touching issue,2026-07-06T16:41:39Z +issue,160744,MPS copy_ produces incorrect output when input has more than one element,https://github.com/pytorch/pytorch/issues/160744,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: mps",review linked PR state before touching issue,2026-07-06T16:27:22Z +issue,188422,Some test_sdpa_can_compile_dynamic in transformers fail (with torch.compile(dynamic=True)) with torch 2.13 RC but pass with torch 2.12,https://github.com/pytorch/pytorch/issues/188422,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2026-07-06T12:40:03Z +issue,186348,torch.mm slower than pointwise mul+sum when K and N are small (under 8),https://github.com/pytorch/pytorch/issues/186348,has linked PR,44,"open PR references this issue; component/oncall labels: module: decompositions, module: inductor, module: linear algebra, module: performance",review linked PR state before touching issue,2026-07-06T05:51:22Z +issue,188718,torch.compile (Inductor) fails with KeyError: 'complex64' on view_as_complex of a valid non-contiguous CUDA tensor,https://github.com/pytorch/pytorch/issues/188718,has linked PR,44,"open PR references this issue; component/oncall labels: module: complex, module: cuda, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-07-06T05:07:14Z +issue,188989,[FSDP] summon_full_params(offload_to_cpu=True) fails with CPU-offloaded FSDP model,https://github.com/pytorch/pytorch/issues/188989,has linked PR,44,"open PR references this issue; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-07-06T02:43:21Z +issue,161132,The Inductor test suite test_torchinductor_opinfo was incorrectly disabled on all devices except CUDA.,https://github.com/pytorch/pytorch/issues/161132,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: tests, oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2026-07-05T20:41:16Z +issue,182277,[CUDA][CUDACachingAllocator] size overflow in caching_allocator_alloc raises SystemError for size >= 2**63,https://github.com/pytorch/pytorch/issues/182277,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: edge cases, module: error checking",review linked PR state before touching issue,2026-07-05T08:31:09Z +issue,173232,error C2872: 'std': ambiguous symbol when compiling `csrc/dynamo/compiled_autograd.h`,https://github.com/pytorch/pytorch/issues/173232,has linked PR,44,"open PR references this issue; component/oncall labels: module: compiled autograd, module: windows, oncall: pt2",review linked PR state before touching issue,2026-07-05T03:28:25Z +issue,184575,"`F.normalize` returns ~1e12 gradient at zero-vector input instead of NaN, while the forward pass already returns NaN",https://github.com/pytorch/pytorch/issues/184575,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: edge cases, module: norms and normalization",review linked PR state before touching issue,2026-07-04T08:42:22Z +issue,187759,"torch.linalg.svdvals swallows NaN (returns finite values), disagreeing with torch.linalg.svd; matrix_rank then reports full rank for a NaN matrix",https://github.com/pytorch/pytorch/issues/187759,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: linear algebra, module: nans and infs",review linked PR state before touching issue,2026-07-04T08:42:06Z +issue,188891,"F.linear with a 1-D weight and bias is inconsistent: eager fails for 2-D input, torch.func transforms (grad/jvp/vmap) fail for all input ranks",https://github.com/pytorch/pytorch/issues/188891,has linked PR,44,"open PR references this issue; component/oncall labels: module: functorch, module: nn",review linked PR state before touching issue,2026-07-03T19:01:49Z +issue,188723,ShapeEnv lacks a metadata-tree transfer API that preserves foreign unbacked SymInt relationships,https://github.com/pytorch/pytorch/issues/188723,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: faketensor, module: fx, module: pt2-dispatcher",review linked PR state before touching issue,2026-07-03T16:45:20Z +issue,188882,Empty autograd section in backend YAML is rejected,https://github.com/pytorch/pytorch/issues/188882,has linked PR,44,"open PR references this issue; component/oncall labels: module: backend, module: codegen",review linked PR state before touching issue,2026-07-03T11:20:21Z +issue,188536,Graph-break reconstruction loses identity of graph-created torch.Event subclasses; record() and wait() operate on different event objects,https://github.com/pytorch/pytorch/issues/188536,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-07-03T09:04:51Z +issue,167223,[feature request] Re-enable `__torch_function__` dispatch when invoking autograd hooks for Tensor subclasses,https://github.com/pytorch/pytorch/issues/167223,has linked PR,44,"open PR references this issue; component/oncall labels: module: __torch_function__, module: autograd, module: docs",review linked PR state before touching issue,2026-07-03T07:36:06Z +issue,188866,[dynamo] TVM backend fails to import with all pip-installable TVM versions (relay removed in TVM 0.20),https://github.com/pytorch/pytorch/issues/188866,has linked PR,44,"open PR references this issue; component/oncall labels: module: backend, module: dependency bug, module: dynamo, module: third_party",review linked PR state before touching issue,2026-07-03T03:44:18Z +issue,171905,Use typing_extensions.TypeAliasType for better reexport of `__module__`,https://github.com/pytorch/pytorch/issues/171905,has linked PR,44,"open PR references this issue; component/oncall labels: module: lint, module: typing",review linked PR state before touching issue,2026-07-02T18:55:05Z +issue,188796,[opaque object] Callable value Opaque Object on a tensor subclass is not accessible for method calls in torch.compile and leads to graph-break,https://github.com/pytorch/pytorch/issues/188796,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: graph breaks, module: library, module: pt2-dispatcher",review linked PR state before touching issue,2026-07-02T12:39:18Z +issue,188757,Windows CI: test_overrides.py class TestTorchFunctionRedispatchOpsCUDA fails consistently (same as #182869 on Linux),https://github.com/pytorch/pytorch/issues/188757,has linked PR,44,"open PR references this issue; component/oncall labels: module: __torch_function__, module: ci, module: cuda, module: tests",review linked PR state before touching issue,2026-07-02T10:52:33Z +issue,188790,Inconsistent device check in test_staging.py: guard uses torch.cuda.is_available() while block body uses torch.accelerator.is_available(),https://github.com/pytorch/pytorch/issues/188790,has linked PR,44,"open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",review linked PR state before touching issue,2026-07-02T09:20:09Z +issue,187806,"F.gelu(approximate='none'): 1+erf catastrophic cancellation — up to 57% relative error in fp32 around x≈−5, zeros the negative tail",https://github.com/pytorch/pytorch/issues/187806,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: nn, module: numerical-stability",review linked PR state before touching issue,2026-07-02T03:17:16Z +issue,188538,torch.fx _format_target emits invalid or semantically incorrect Python for valid module names,https://github.com/pytorch/pytorch/issues/188538,has linked PR,44,"open PR references this issue; component/oncall labels: module: fx, oncall: pt2",review linked PR state before touching issue,2026-07-02T00:57:59Z +issue,188711,DISABLED test_main_loop_scaling_shape0_use_fast_accum_True_scaling_block_sizes2_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188711,has linked PR,44,open PR references this issue; component/oncall labels: module: cuda,review linked PR state before touching issue,2026-07-02T00:34:17Z +issue,188728,[MPS] F.linear does not respect the PYTORCH_MPS_PREFER_METAL flag,https://github.com/pytorch/pytorch/issues/188728,has linked PR,44,open PR references this issue; component/oncall labels: module: mps,review linked PR state before touching issue,2026-07-01T20:25:44Z +issue,184002,[DTensor][Compile] Compile fails when DTensor function returns an aliased output,https://github.com/pytorch/pytorch/issues/184002,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: dtensor, module: pt2-dispatcher, oncall: distributed",review linked PR state before touching issue,2026-07-01T19:31:16Z +issue,188635,torch.compile misses fusion for producer -> view/transpose/contiguous layout clone in attention-style paths,https://github.com/pytorch/pytorch/issues/188635,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2026-07-01T02:52:01Z +issue,188477,[CI][B200] smoke_b200 test_nv_universal_gemm fails: cutlass.cute.arch has no attribute 'ProxyKind' (unpinned cutlass_api vs pinned cutlass-dsl),https://github.com/pytorch/pytorch/issues/188477,has linked PR,44,open PR references this issue; component/oncall labels: module: ci,review linked PR state before touching issue,2026-07-01T01:55:25Z +issue,188544,[opaque object] Value Opaque Object errors with instance static method.,https://github.com/pytorch/pytorch/issues/188544,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: guards, oncall: pt2",review linked PR state before touching issue,2026-06-30T12:00:58Z +issue,188398,[Docs] torch.utils.mobile_optimizer link broken in 2.7 API documentation,https://github.com/pytorch/pytorch/issues/188398,has linked PR,44,open PR references this issue; component/oncall labels: module: docs,review linked PR state before touching issue,2026-06-30T11:43:01Z +issue,188325,torch._dynamo: EventVariable.python_type() returns torch.Event instead of type(self.value),https://github.com/pytorch/pytorch/issues/188325,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-06-30T11:30:36Z +issue,187107,"[MPS] linalg.lu_factor not implemented, causing RunTimeErrors on torch.linalg.det and torch.linalg.solve",https://github.com/pytorch/pytorch/issues/187107,has linked PR,44,"open PR references this issue; component/oncall labels: module: complex, module: linear algebra, module: mps",review linked PR state before touching issue,2026-06-29T19:49:18Z +issue,188405,"torch.compile reconstructs a graph-created torch.Stream/torch.Event subclass as the base class, losing the subclass type across a graph break",https://github.com/pytorch/pytorch/issues/188405,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-06-29T19:22:50Z +issue,188226,[inductor] `ValueRangeError: Invalid ranges [0:-61]` during index propagation,https://github.com/pytorch/pytorch/issues/188226,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: pooling, oncall: pt2",review linked PR state before touching issue,2026-06-29T18:20:26Z +issue,188166,load_state_dict(assign=True) mutates the caller's state_dict._metadata,https://github.com/pytorch/pytorch/issues/188166,has linked PR,44,open PR references this issue; component/oncall labels: module: nn,review linked PR state before touching issue,2026-06-29T18:19:45Z +issue,188213,[inductor] `LoweringException: AssertionError` in `scatter_reduce_` lowering,https://github.com/pytorch/pytorch/issues/188213,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: scatter & gather ops, oncall: pt2",review linked PR state before touching issue,2026-06-29T17:38:24Z +issue,188048,"[Inductor][CUDA] torch.div(rounding_mode=""floor"") mismatches eager for infinite inputs",https://github.com/pytorch/pytorch/issues/188048,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: cuda, module: edge cases, module: inductor",review linked PR state before touching issue,2026-06-29T17:10:11Z +issue,188128,`asarray` does not work for boolean dtype when using buffer protocol,https://github.com/pytorch/pytorch/issues/188128,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: python array api, module: python frontend, module: tensor creation",review linked PR state before touching issue,2026-06-29T17:07:00Z +issue,188423,test_sdpa_can_compile_dynamic in transformers fails (with torch.compile(dynamic=True)) with torch 2.13 RC but pass with torch 2.12,https://github.com/pytorch/pytorch/issues/188423,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, module: sdpa, oncall: pt2",review linked PR state before touching issue,2026-06-29T14:48:32Z +issue,188131,DISABLED test_fallback_kernel_with_symexpr_output_xpu (__main__.AOTInductorTestABICompatibleGpu),https://github.com/pytorch/pytorch/issues/188131,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: xpu, oncall: pt2",review linked PR state before touching issue,2026-06-29T08:54:01Z +issue,188390,"[DTensor][Compile][AOTAutograd cache] DTensor stable hashing generates identical AOTAutograd hash keys across ranks, causing false cache hits.",https://github.com/pytorch/pytorch/issues/188390,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: dtensor, module: pt2-dispatcher, oncall: distributed",review linked PR state before touching issue,2026-06-29T06:19:29Z +issue,186107,torch.scan gaps,https://github.com/pytorch/pytorch/issues/186107,has linked PR,44,"open PR references this issue; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-06-28T22:53:48Z +issue,188370,[Meta][PrivateUse1]: Data-dependent ATen ops need eager shape-oracle support for PU1 backends.,https://github.com/pytorch/pytorch/issues/188370,has linked PR,44,"open PR references this issue; component/oncall labels: module: backend, module: meta tensors, module: privateuse1",review linked PR state before touching issue,2026-06-28T17:14:35Z +issue,188338,[ROCm]: CUDAGraph safe Convolutional Kernel e.g. `F.conv_transpose1d`,https://github.com/pytorch/pytorch/issues/188338,has linked PR,44,"open PR references this issue; component/oncall labels: module: convolution, module: cuda graphs, module: rocm",review linked PR state before touching issue,2026-06-27T15:36:44Z +issue,188245,Broken tutorial URL,https://github.com/pytorch/pytorch/issues/188245,has linked PR,44,open PR references this issue; component/oncall labels: module: docs,review linked PR state before touching issue,2026-06-26T15:58:35Z +issue,187953,"torch.compile: out= variants of CompositeImplicitAutograd linalg ops (linalg.norm, cholesky, ...) recompile on every input shape",https://github.com/pytorch/pytorch/issues/187953,has linked PR,44,"open PR references this issue; component/oncall labels: module: compile-time, module: dynamic shapes, module: faketensor, module: guards",review linked PR state before touching issue,2026-06-26T14:00:33Z +issue,144580,`torch._foreach_mul` does not support autograd,https://github.com/pytorch/pytorch/issues/144580,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: mta",review linked PR state before touching issue,2026-06-26T06:37:37Z +issue,188230,[inductor] `index_propagation` `ValueError: The argument 'False' is not comparable`,https://github.com/pytorch/pytorch/issues/188230,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-26T01:25:42Z +issue,188225,[inductor] `index_propagation` `ValueError: The argument 'nan' is not comparable`,https://github.com/pytorch/pytorch/issues/188225,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-26T01:21:30Z +issue,188217,[inductor] `TypeError: reduce() of empty iterable with no initial value` during lowering,https://github.com/pytorch/pytorch/issues/188217,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-26T00:51:46Z +issue,188182,torch.fft has no sharding strategy,https://github.com/pytorch/pytorch/issues/188182,has linked PR,44,"open PR references this issue; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-25T23:20:26Z +issue,188150,torch.compile reduce-overhead: CUDAGraphs recompiles on every batch with dynamic padding (HF training loop),https://github.com/pytorch/pytorch/issues/188150,has linked PR,44,"open PR references this issue; component/oncall labels: module: compile ux, module: cuda graphs, module: dynamic shapes, module: performance",review linked PR state before touching issue,2026-06-25T13:00:29Z +issue,145208,Some FlexAttention learned bias bugs/limitations,https://github.com/pytorch/pytorch/issues/145208,has linked PR,44,"open PR references this issue; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-06-25T09:01:08Z +issue,188088,torch.clamp rejects mixed Tensor and Number min/max bounds although Tensor+Tensor and Number+Number both work,https://github.com/pytorch/pytorch/issues/188088,has linked PR,44,open PR references this issue; component/oncall labels: module: python frontend,review linked PR state before touching issue,2026-06-25T07:11:36Z +issue,186544,[Windows] Debug build fails: LNK1104 cannot open file 'python310_d.lib',https://github.com/pytorch/pytorch/issues/186544,has linked PR,44,"open PR references this issue; component/oncall labels: module: build, module: debug-build, module: pybind, module: windows",review linked PR state before touching issue,2026-06-25T02:29:40Z +issue,137096,Expressions involving Min/Max don't simplify with value ranges information,https://github.com/pytorch/pytorch/issues/137096,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2026-06-24T21:53:29Z +issue,188052,Allow PrivateUse1 backends to opt out of CompositeExplicitAutograd decomposition,https://github.com/pytorch/pytorch/issues/188052,has linked PR,44,"open PR references this issue; component/oncall labels: module: dispatch, module: privateuse1",review linked PR state before touching issue,2026-06-24T18:10:51Z +issue,185814,"[Inductor] Mix-Order Reduction: XBLOCK Derivation Ignores N in RMSNorm Backward, Causing Bandwidth Cliff",https://github.com/pytorch/pytorch/issues/185814,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-23T17:31:02Z +issue,185589,"[inductor] Severe numerical inconsistency: diag_embed + erf + normalize(p=-1) under torch.compile(dynamic=True) — max diff up to 565,248",https://github.com/pytorch/pytorch/issues/185589,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: dynamic shapes, module: inductor, module: numerical-stability",review linked PR state before touching issue,2026-06-23T17:30:41Z +issue,185581,`torch.compile` fails at FakeTensor validation for ConvTranspose2d,https://github.com/pytorch/pytorch/issues/185581,has linked PR,44,open PR references this issue,review linked PR state before touching issue,2026-06-23T17:30:15Z +issue,185481,[inductor] argmax over comparison mask is accepted and used for indexing while eager raises,https://github.com/pytorch/pytorch/issues/185481,has linked PR,44,"open PR references this issue; component/oncall labels: module: error checking, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-23T17:29:04Z +issue,185382,`torch.compile`: Inductor silently produces wrong results for `addcdiv_`/`addcmul_` when `.item()` graph break occurs in a loop,https://github.com/pytorch/pytorch/issues/185382,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-23T17:28:05Z +issue,187945,"`replace_pattern` and `replace_pattern_with_filters` do not propagate `meta[""val""]` to replacement nodes, causing downstream filters to silently reject matches",https://github.com/pytorch/pytorch/issues/187945,has linked PR,44,"open PR references this issue; component/oncall labels: module: fx, oncall: pt2",review linked PR state before touching issue,2026-06-23T14:00:38Z +issue,116465,torch.cartesian_prod inconsitent return shape for one input tensor,https://github.com/pytorch/pytorch/issues/116465,has linked PR,44,open PR references this issue; component/oncall labels: module: numpy,review linked PR state before touching issue,2026-06-23T09:36:28Z +issue,185248,`torch.compile` silently accepts negative standard deviations in `torch.normal` that eager mode rejects,https://github.com/pytorch/pytorch/issues/185248,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: dynamo, module: random, oncall: pt2",review linked PR state before touching issue,2026-06-23T08:11:05Z +issue,185246,torch.compile silently accepts invalid Bernoulli probabilities that eager mode rejects,https://github.com/pytorch/pytorch/issues/185246,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: dynamo, module: error checking, module: random",review linked PR state before touching issue,2026-06-23T08:07:37Z +issue,187332,[PT2] Output mismatch between eager mode and torch.compile for torch.special.i0 with inf inputs on CUDA,https://github.com/pytorch/pytorch/issues/187332,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: cuda, module: decompositions, module: edge cases",review linked PR state before touching issue,2026-06-23T08:03:36Z +issue,185696,[Inductor] torch.atan2 yields incorrect result for x=0.0 with float16 on CUDA (loses signed zero),https://github.com/pytorch/pytorch/issues/185696,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: cuda, module: edge cases, module: half",review linked PR state before touching issue,2026-06-23T08:03:35Z +issue,146938,"ModuleDict has an incomplete (or wrong) typing, producing type errors when accessed with attributes",https://github.com/pytorch/pytorch/issues/146938,has linked PR,44,"open PR references this issue; component/oncall labels: module: nn, module: typing",review linked PR state before touching issue,2026-06-20T02:07:56Z +issue,187178,torch.nn.functional.triplet_margin_loss returns NaN for p < 1 (p-norm overflow),https://github.com/pytorch/pytorch/issues/187178,has linked PR,44,"open PR references this issue; component/oncall labels: module: edge cases, module: nn",review linked PR state before touching issue,2026-06-19T06:35:32Z +issue,187177,"torch.nn.LocalResponseNorm returns +/-Inf when k=0 and alpha=0 (zero denominator, no validation)",https://github.com/pytorch/pytorch/issues/187177,has linked PR,44,"open PR references this issue; component/oncall labels: module: error checking, module: nn, module: norms and normalization",review linked PR state before touching issue,2026-06-19T06:35:23Z +issue,180807,[inductor][static launcher] Missing runtime dtype validation for pointer-typed Triton args can silently produce incorrect results,https://github.com/pytorch/pytorch/issues/180807,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-18T14:57:42Z +issue,141595,FX graph cache does not do provenance tracking for inductor guards,https://github.com/pytorch/pytorch/issues/141595,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-18T04:32:22Z +issue,136264,Compatibility with numpy 2 __array__ interface,https://github.com/pytorch/pytorch/issues/136264,has linked PR,44,open PR references this issue; component/oncall labels: module: numpy,review linked PR state before touching issue,2026-06-17T19:56:18Z +issue,187593,[Dynamo] Avoid guarding on Unused variables,https://github.com/pytorch/pytorch/issues/187593,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: guards, oncall: pt2",review linked PR state before touching issue,2026-06-17T18:55:37Z +issue,64947,Quantile is limited to 16 million elements and have poor performance.,https://github.com/pytorch/pytorch/issues/64947,has linked PR,44,"open PR references this issue; component/oncall labels: module: numpy, module: performance, module: sorting and selection",review linked PR state before touching issue,2026-06-17T13:36:29Z +issue,187296,torch.heaviside returns 0 for NaN input instead of propagating NaN,https://github.com/pytorch/pytorch/issues/187296,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: nans and infs, module: numpy",review linked PR state before touching issue,2026-06-17T11:54:29Z +issue,187547,Dynamo graph-breaks on raw unbacked SymInt BlockMask lengths in FlexAttention,https://github.com/pytorch/pytorch/issues/187547,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-06-17T04:14:29Z +issue,187336,"[Inductor] torch.compile loses signbit in torch.special.log_ndtr, causing discrepancies in sign-sensitive ops (e.g., copysign, atan2)",https://github.com/pytorch/pytorch/issues/187336,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: inductor, module: special",review linked PR state before touching issue,2026-06-17T02:34:08Z +issue,187451,RFC: Polyhedral Optimization Pass for PyTorch Inductor,https://github.com/pytorch/pytorch/issues/187451,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2026-06-16T17:19:59Z +issue,187338,torch.gcd returns an incorrect negative result when one input contains the minimum value of torch.int8 (-128),https://github.com/pytorch/pytorch/issues/187338,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases",review linked PR state before touching issue,2026-06-16T04:39:07Z +issue,187389,Missing reference CPU implementation of sparse `triangular_solve` (Mac OS / ARM and other non-MKL supported targets),https://github.com/pytorch/pytorch/issues/187389,has linked PR,44,"open PR references this issue; component/oncall labels: module: arm, module: cpu, module: linear algebra, module: macos",review linked PR state before touching issue,2026-06-15T22:22:34Z +issue,187295,torch.sign returns 0 for NaN inputs instead of propagating NaN,https://github.com/pytorch/pytorch/issues/187295,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: nans and infs, module: testing",review linked PR state before touching issue,2026-06-15T17:23:10Z +issue,186824,torch.distributions.Cauchy.icdf(0) returns large finite value with wrong sign in float32; Gumbel.icdf(0/1) also finite instead of ±inf,https://github.com/pytorch/pytorch/issues/186824,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: distributions, module: edge cases",review linked PR state before touching issue,2026-06-15T17:02:27Z +issue,187299,torch.nn.functional.unfold — RuntimeError instead of ValueError for zero stride,https://github.com/pytorch/pytorch/issues/187299,has linked PR,44,"open PR references this issue; component/oncall labels: module: edge cases, module: error checking, module: nn",review linked PR state before touching issue,2026-06-15T07:00:17Z +issue,187233,torch.nn.functional.max_unpool3d — internal MPS RuntimeError for complex64 input,https://github.com/pytorch/pytorch/issues/187233,has linked PR,44,"open PR references this issue; component/oncall labels: module: complex, module: error checking, module: mps, module: pooling",review linked PR state before touching issue,2026-06-15T05:14:04Z +issue,187298,convert_sync_batchnorm silently drops the activation / extra behavior of `_BatchNorm` subclasses (e.g. timm `BatchNormAct2d`) - please emit a warning,https://github.com/pytorch/pytorch/issues/187298,has linked PR,44,"open PR references this issue; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-15T03:07:13Z +issue,187294,torch.special.zeta returns NaN for q=+Inf instead of 0,https://github.com/pytorch/pytorch/issues/187294,has linked PR,44,"open PR references this issue; component/oncall labels: module: edge cases, module: nans and infs, module: special",review linked PR state before touching issue,2026-06-15T02:43:51Z +issue,176298,"[CPU] Binary ops (add, sub, mul) not implemented for uint16/uint32/uint64",https://github.com/pytorch/pytorch/issues/176298,has linked PR,44,"open PR references this issue; component/oncall labels: module: cpu, module: unsigned int",review linked PR state before touching issue,2026-06-14T20:24:59Z +issue,170367,DTensor redistributes on foreach_X operators are inefficient,https://github.com/pytorch/pytorch/issues/170367,has linked PR,44,"open PR references this issue; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-13T16:37:25Z +issue,170648,fsdp2 support grad_dtype,https://github.com/pytorch/pytorch/issues/170648,has linked PR,44,"open PR references this issue; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-12T17:09:39Z +issue,187170,[inductor] test_comprehensive_addbmm_cpu_float16 fails due to overly strict float32 reference comparison,https://github.com/pytorch/pytorch/issues/187170,has linked PR,44,"open PR references this issue; component/oncall labels: module: half, module: inductor, module: tests, oncall: pt2",review linked PR state before touching issue,2026-06-12T11:36:45Z +issue,187155,[FSDP] Cache post-forward DeviceMesh to deduplicate NCCL communicators,https://github.com/pytorch/pytorch/issues/187155,has linked PR,44,"open PR references this issue; component/oncall labels: module: fsdp, module: nccl, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-12T07:43:22Z +issue,187008,"nn.Module.to() fails on a tensor-subclass parameter held by an EMA / external weakref (forced swap_tensors path, no compile)",https://github.com/pytorch/pytorch/issues/187008,has linked PR,44,open PR references this issue; component/oncall labels: module: nn,review linked PR state before touching issue,2026-06-11T22:15:48Z +issue,187089,Support BFloat16 in ATen CPU tensor creation (tensor_cpu),https://github.com/pytorch/pytorch/issues/187089,has linked PR,44,"open PR references this issue; component/oncall labels: module: bfloat16, module: cpu, module: tensor creation",review linked PR state before touching issue,2026-06-11T22:05:58Z +issue,187041,[opaque object] Bug with storing in global dict and consuming in TensorSubclass,https://github.com/pytorch/pytorch/issues/187041,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: library, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-06-11T13:22:00Z +issue,73009,Improve loading for nn.modules.lazy.LazyModuleMixin,https://github.com/pytorch/pytorch/issues/73009,has linked PR,44,open PR references this issue; component/oncall labels: module: nn,review linked PR state before touching issue,2026-06-11T10:02:00Z +issue,186148,"`torch.matmul` misses `mm` fold for `[S, 1, H] @ [H, K]` when the size-1 batch dim has a non-contiguous stride",https://github.com/pytorch/pytorch/issues/186148,has linked PR,44,"open PR references this issue; component/oncall labels: module: linear algebra, module: performance",review linked PR state before touching issue,2026-06-11T08:43:58Z +issue,185396,[Inductor][CUDA] Internal KeyError: 'complex64' during Triton signature generation for graph with complex intermediates,https://github.com/pytorch/pytorch/issues/185396,has linked PR,44,"open PR references this issue; component/oncall labels: module: compile ux, module: complex, module: cuda, module: inductor",review linked PR state before touching issue,2026-06-10T21:30:53Z +issue,153485,Double backward error in Pytorch PP,https://github.com/pytorch/pytorch/issues/153485,has linked PR,44,"open PR references this issue; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-10T17:13:54Z +issue,184195,aoti_torch_cuda__scaled_dot_product_fused_attention_overrideable not implemented,https://github.com/pytorch/pytorch/issues/184195,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotinductor, module: privateuse1, module: sdpa, oncall: export",review linked PR state before touching issue,2026-06-10T17:02:07Z +issue,186875,[Inductor] fmod/remainder with Python float scalar diverge from eager for bf16/fp16 on CPU (same class as #185517),https://github.com/pytorch/pytorch/issues/186875,has linked PR,44,"open PR references this issue; component/oncall labels: oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2026-06-10T14:51:50Z +issue,186799,"`torch.compile` silently produces wrong gradient for `atan2(x, y).amin(dim)` on float32",https://github.com/pytorch/pytorch/issues/186799,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2026-06-10T14:14:32Z +issue,186901,Inconsistent behavior of torch.sign on NaN input values,https://github.com/pytorch/pytorch/issues/186901,has linked PR,44,"open PR references this issue; component/oncall labels: module: edge cases, module: nans and infs",review linked PR state before touching issue,2026-06-10T12:21:47Z +issue,146306,torch.compile failed on single-node multigpu setting.,https://github.com/pytorch/pytorch/issues/146306,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-09T17:12:59Z +issue,147075,[inductor] Performance Degradation and Hang in `torch.diff`,https://github.com/pytorch/pytorch/issues/147075,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2026-06-09T17:12:58Z +issue,151027,Add size/strides/alignment assertions for cpp_wrapper,https://github.com/pytorch/pytorch/issues/151027,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotinductor, module: inductor, oncall: export, oncall: pt2",review linked PR state before touching issue,2026-06-09T17:12:56Z +issue,154301,cudagraph copies in 'max-autotune' can add overhead on small model,https://github.com/pytorch/pytorch/issues/154301,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-09T17:12:55Z +issue,183259,InductorError on in-place multidimensional dynamic slicing with Tensor-derived slice bounds,https://github.com/pytorch/pytorch/issues/183259,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-09T17:12:54Z +issue,123339,"Inconsistency: BCELoss asserts target in [0, 1], BCEWithLogitsLoss does not",https://github.com/pytorch/pytorch/issues/123339,has linked PR,44,open PR references this issue; component/oncall labels: module: loss,review linked PR state before touching issue,2026-06-09T09:31:35Z +issue,185497,[pt2] RuntimeError: variable modified by inplace operation during backward in compiled mode (succeeds in eager),https://github.com/pytorch/pytorch/issues/185497,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-06-08T17:11:15Z +issue,186556,torch.ldexp returns incorrect gradient w.r.t. mantissa for negative exponents,https://github.com/pytorch/pytorch/issues/186556,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: correctness (silent), module: derivatives, module: windows",review linked PR state before touching issue,2026-06-08T17:07:37Z +issue,186176,"AOTInductor serializes entire underlying storage for each constant tensor, causing severe bloat for shared-storage weights",https://github.com/pytorch/pytorch/issues/186176,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotinductor, module: serialization, oncall: export, oncall: pt2",review linked PR state before touching issue,2026-06-08T16:11:57Z +issue,186572,[Inductor] Checkpointed reductions can choose different configs between forward and recompute,https://github.com/pytorch/pytorch/issues/186572,has linked PR,44,"open PR references this issue; component/oncall labels: module: activation checkpointing, module: correctness (silent), module: inductor, module: reductions",review linked PR state before touching issue,2026-06-08T15:00:27Z +issue,185509,"[inductor] Silent incorrect gradients with interpolate, conv2d, max_pool2d and torch.where",https://github.com/pytorch/pytorch/issues/185509,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-08T14:38:35Z +issue,185510,[Dynamo] TypeError: cannot determine truth value of Relational during stride sorting with dynamic=True,https://github.com/pytorch/pytorch/issues/185510,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-08T14:29:06Z +issue,185647,Unexpected storage from MSE loss output,https://github.com/pytorch/pytorch/issues/185647,has linked PR,44,"open PR references this issue; component/oncall labels: module: loss, module: memory usage",review linked PR state before touching issue,2026-06-08T13:31:54Z +issue,89459,Higher order derivatives of sinc explode,https://github.com/pytorch/pytorch/issues/89459,has linked PR,44,open PR references this issue; component/oncall labels: module: derivatives,review linked PR state before touching issue,2026-06-07T01:04:24Z +issue,178096,torch.compile(dynamic=True) on CUDA: large eager vs compiled mismatch for BatchNorm2d + Conv2d,https://github.com/pytorch/pytorch/issues/178096,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: pt2 accuracy, oncall: pt2",review linked PR state before touching issue,2026-06-06T08:44:11Z +issue,171249,[Bug] softplus returns inf for large beta while other frameworks return finite value,https://github.com/pytorch/pytorch/issues/171249,has linked PR,44,"open PR references this issue; component/oncall labels: module: edge cases, module: nans and infs, module: nn",review linked PR state before touching issue,2026-06-06T04:14:15Z +issue,160057,Error raised when running torch.compile under FakeTensorMode context,https://github.com/pytorch/pytorch/issues/160057,has linked PR,44,open PR references this issue; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2026-06-05T18:20:44Z +issue,140497,C++ standard library functions not found when compiling c10d debug handlers,https://github.com/pytorch/pytorch/issues/140497,has linked PR,44,"open PR references this issue; component/oncall labels: oncall: distributed, oncall: distributed infra",review linked PR state before touching issue,2026-06-05T17:01:35Z +issue,186374,[CUDA] `TestCudaAllocator.test_garbage_collect_expandable` failing on Orin,https://github.com/pytorch/pytorch/issues/186374,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: cudacachingallocator, module: jetson, module: tests",review linked PR state before touching issue,2026-06-05T14:35:57Z +issue,186028,"[dynamic] torch.linalg.cond(p=""fro"") fails with symbolic sizes while dynamic=False succeeds",https://github.com/pytorch/pytorch/issues/186028,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: faketensor, module: linear algebra, module: pt2-dispatcher",review linked PR state before touching issue,2026-06-05T12:17:46Z +issue,186354,Dynamo graph break in native bmm_outer_product eager router due to _is_cow_tensor / call_boxed,https://github.com/pytorch/pytorch/issues/186354,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: graph breaks, oncall: pt2",review linked PR state before touching issue,2026-06-05T10:01:51Z +issue,186140,fix qualname for OpOverload/OpOverloadPacket,https://github.com/pytorch/pytorch/issues/186140,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-06-04T18:50:45Z +issue,185888,[Inductor] as_strided_ metadata mutation is not reflected on original tensor,https://github.com/pytorch/pytorch/issues/185888,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: meta tensors, oncall: pt2",review linked PR state before touching issue,2026-06-04T17:01:40Z +issue,185975,torch.accelerator.default_stream API is missing,https://github.com/pytorch/pytorch/issues/185975,has linked PR,44,"open PR references this issue; component/oncall labels: module: accelerator, module: privateuse1",review linked PR state before touching issue,2026-06-04T16:18:26Z +issue,185885,[Inductor] torch.index_add accepts negative index while eager raises out-of-bounds error,https://github.com/pytorch/pytorch/issues/185885,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-06-04T15:03:14Z +issue,186136,The added test TestCompileWorker.test_shutdown_terminates_sidecar_worker_pool (and the corresponding version in TestCompileWorkerWithTimer) are flaky internally due to hitting timeout.,https://github.com/pytorch/pytorch/issues/186136,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: multiprocessing, module: tests, oncall: pt2",review linked PR state before touching issue,2026-06-04T04:28:27Z +issue,184783,`set_multithreading_enabled` still spawns autograd threads,https://github.com/pytorch/pytorch/issues/184783,has linked PR,44,open PR references this issue; component/oncall labels: module: autograd,review linked PR state before touching issue,2026-06-03T19:37:56Z +issue,185480,[Inductor][CUDA][bf16] `F.celu` returns `+0.0` where eager returns a negative bf16 subnormal,https://github.com/pytorch/pytorch/issues/185480,has linked PR,44,"open PR references this issue; component/oncall labels: module: bfloat16, module: correctness (silent), module: cuda, module: edge cases",review linked PR state before touching issue,2026-06-02T21:01:32Z +issue,185470,[Inductor] `F.threshold` bf16 miscompilation: scalar threshold not cast to tensor dtype,https://github.com/pytorch/pytorch/issues/185470,has linked PR,44,"open PR references this issue; component/oncall labels: module: bfloat16, module: correctness (silent), module: decompositions, oncall: pt2",review linked PR state before touching issue,2026-06-02T21:01:14Z +issue,185955,Integrate with ONNX 1.22.0 release branch,https://github.com/pytorch/pytorch/issues/185955,has linked PR,44,open PR references this issue; component/oncall labels: module: onnx,review linked PR state before touching issue,2026-06-02T20:20:11Z +issue,185464,torch.compile produces wrong results for F.multilabel_margin_loss when target contains -1,https://github.com/pytorch/pytorch/issues/185464,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: decompositions, module: loss, oncall: pt2",review linked PR state before touching issue,2026-06-02T18:07:50Z +issue,185185,torch.compile(fullgraph=True) fails with PendingUnbackedSymbolNotFound when tensor-derived scalar values are used as slice bounds,https://github.com/pytorch/pytorch/issues/185185,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2026-06-02T17:25:52Z +issue,185331,"[pipelining] Pipelines reuse recv buffer tensors and directly pass them to the stage, causing backward hooks on stage boundary activations to persist between steps",https://github.com/pytorch/pytorch/issues/185331,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: pipelining, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-02T17:23:06Z +issue,185780,[pipelining] Add downstream tests to test pipelining integration with popular HF transformers architectures,https://github.com/pytorch/pytorch/issues/185780,has linked PR,44,"open PR references this issue; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-06-01T19:46:25Z +issue,184031,torch.nn.functional.gumbel_softmax returns NaN when tau=0 (division by zero),https://github.com/pytorch/pytorch/issues/184031,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: nn",review linked PR state before touching issue,2026-06-01T17:08:53Z +issue,185006,ConstantLR accepts factor=0 and crashes with ZeroDivisionError after total_iters steps,https://github.com/pytorch/pytorch/issues/185006,has linked PR,44,"open PR references this issue; component/oncall labels: module: error checking, module: lrscheduler, module: optimizer",review linked PR state before touching issue,2026-05-30T11:23:50Z +issue,185640,"`make_fx` generates broken input codegen for a function with a `(tuple, dict)` positional signature",https://github.com/pytorch/pytorch/issues/185640,has linked PR,44,"open PR references this issue; component/oncall labels: module: fx, module: proxytensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-29T16:32:38Z +issue,185543,[inductor] Gradient mismatch between eager and inductor for torch.quantile on tied values,https://github.com/pytorch/pytorch/issues/185543,has linked PR,44,"open PR references this issue; component/oncall labels: module: decompositions, module: derivatives, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-29T14:44:20Z +issue,185630,[PT2] RuntimeError with garbage shape in apply_view_meta_sequence when using torch.compile(dynamic=True),https://github.com/pytorch/pytorch/issues/185630,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: dynamic shapes, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-29T13:39:33Z +issue,184219,serialization_id is not deterministic across machines,https://github.com/pytorch/pytorch/issues/184219,has linked PR,44,open PR references this issue; component/oncall labels: module: serialization,review linked PR state before touching issue,2026-05-29T12:50:29Z +issue,185575,"[inductor] Upsample(mode='area') fails under torch.compile(dynamic=True) — ""cannot determine truth value of Relational"" in _adaptive_avg_pool2d lowering",https://github.com/pytorch/pytorch/issues/185575,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, module: interpolation, module: pooling",review linked PR state before touching issue,2026-05-29T01:20:06Z +issue,185533,"torch.compile(mode=""max-autotune"") fails with NotImplementedError: SliceView in get_stride()",https://github.com/pytorch/pytorch/issues/185533,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-28T20:53:58Z +issue,180346,[DTensor] clip_grad_norm_ fails when per-parameter norms come from different DeviceMesh objects,https://github.com/pytorch/pytorch/issues/180346,has linked PR,44,"open PR references this issue; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms",review linked PR state before touching issue,2026-05-28T14:12:40Z +issue,185216,`torch.ops.symm_mem.one_shot_all_reduce` causes `CUDA_ERROR_ILLEGAL_ADDRESS`,https://github.com/pytorch/pytorch/issues/185216,has linked PR,44,"open PR references this issue; component/oncall labels: module: symm_mem, oncall: distributed",review linked PR state before touching issue,2026-05-27T23:07:31Z +issue,185296,[AOTAutograd] [DTensor] subclass output SymInt arity mismatch causes partitioner assertion,https://github.com/pytorch/pytorch/issues/185296,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: dtensor, module: pt2-dispatcher, oncall: distributed",review linked PR state before touching issue,2026-05-26T23:32:28Z +issue,176770,"torch.compile error in unit tests, but test passes when ran individually",https://github.com/pytorch/pytorch/issues/176770,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-26T17:55:03Z +issue,185046,[Bug] `torch.library._register_fake` crashes when any unrelated `LazyLoader` module in `sys.modules` fails to import,https://github.com/pytorch/pytorch/issues/185046,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: library, module: pt2-dispatcher, module: vllm",review linked PR state before touching issue,2026-05-26T17:51:45Z +issue,185200,[C++] Print tensors without storage,https://github.com/pytorch/pytorch/issues/185200,has linked PR,44,"open PR references this issue; component/oncall labels: module: cpp, module: printing",review linked PR state before touching issue,2026-05-26T15:08:45Z +issue,185141,"[Dynamo] torchvision.io functions raise ""RuntimeError when making fake tensor call"" instead of graph breaking",https://github.com/pytorch/pytorch/issues/185141,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-05-26T14:36:57Z +issue,102948,[onnx] aten::cumprod cannot be exported to ONNX,https://github.com/pytorch/pytorch/issues/102948,has linked PR,44,open PR references this issue; component/oncall labels: module: onnx,review linked PR state before touching issue,2026-05-26T14:02:18Z +issue,184842,`torch.autograd.functional.jacobian` raises `NotImplementedError` at execution time for `strategy='forward-mode'` with `vectorize=False` instead of raising `ValueError` at argument validation,https://github.com/pytorch/pytorch/issues/184842,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: error checking, module: forward ad",review linked PR state before touching issue,2026-05-26T13:42:39Z +issue,183957,enum custom op input is a little weird with torch.library.define,https://github.com/pytorch/pytorch/issues/183957,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: library, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-26T13:32:46Z +issue,184572,`torch.maximum` / `torch.clamp_min` gradients at tie point violate PyTorch's documented subgradient convention and are inconsistent with `F.relu`,https://github.com/pytorch/pytorch/issues/184572,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: edge cases",review linked PR state before touching issue,2026-05-26T13:20:37Z +issue,141184,Dynamo graph break on RNN/LSTM doesn't resume the function correctly.,https://github.com/pytorch/pytorch/issues/141184,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-05-25T12:57:08Z +issue,185023,register_buffer-created attributes and Parameter.data incorrectly typed as Tensor | Module,https://github.com/pytorch/pytorch/issues/185023,has linked PR,44,"open PR references this issue; component/oncall labels: module: nn, module: typing",review linked PR state before touching issue,2026-05-24T12:49:58Z +issue,185022,ModuleList not recognized as Iterable by type checkers,https://github.com/pytorch/pytorch/issues/185022,has linked PR,44,"open PR references this issue; component/oncall labels: module: nn, module: typing",review linked PR state before touching issue,2026-05-23T22:55:49Z +issue,184890,vmap of autograd.grad fails on the output of model run on AMP,https://github.com/pytorch/pytorch/issues/184890,has linked PR,44,"open PR references this issue; component/oncall labels: module: amp (automated mixed precision), module: autograd, module: functorch, module: norms and normalization",review linked PR state before touching issue,2026-05-22T18:04:59Z +issue,183971,`ldd libtorch_nvshmem.so` in 2.12 libtorch results in `libnvshmem_host.so.3 => not found`,https://github.com/pytorch/pytorch/issues/183971,has linked PR,44,"open PR references this issue; component/oncall labels: module: cpp, module: cuda",review linked PR state before touching issue,2026-05-22T15:40:59Z +issue,184845,"`torch.nn.functional.pad` raises `NotImplementedError` for valid `(ndim, pad_size)` combinations in non-constant modes — constraint not documented in the API",https://github.com/pytorch/pytorch/issues/184845,has linked PR,44,"open PR references this issue; component/oncall labels: module: docs, module: error checking, module: nn, module: padding",review linked PR state before touching issue,2026-05-22T03:39:32Z +issue,184642,`Module.to()` has no overload for `.to(memory_format=torch.channels_last)`,https://github.com/pytorch/pytorch/issues/184642,has linked PR,44,"open PR references this issue; component/oncall labels: module: memory format, module: nn, module: typing",review linked PR state before touching issue,2026-05-21T02:44:32Z +issue,161113,[UED][Falcon-7B x vLLM] selective_scan_fwd custom op runtime error,https://github.com/pytorch/pytorch/issues/161113,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, module: vllm, oncall: pt2",review linked PR state before touching issue,2026-05-19T22:58:20Z +issue,183988,[inductor] `torch.compile` crashes on `x[torch.randperm(n)]` — pattern matcher missing `slice_shape` input,https://github.com/pytorch/pytorch/issues/183988,has linked PR,44,"open PR references this issue; component/oncall labels: module: advanced indexing, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:09:43Z +issue,183899,torch.compile fails on addmm with beta=0 and non-broadcastable bias shape,https://github.com/pytorch/pytorch/issues/183899,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: linear algebra, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:09:30Z +issue,183887,torch.compile fails on int64 addmm/addmm_ comparison with fake tensor RuntimeError,https://github.com/pytorch/pytorch/issues/183887,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:09:16Z +issue,183249,`torch.compile` produces wrong results for `fractional_max_pool2d` when `_random_samples` is not provided,https://github.com/pytorch/pytorch/issues/183249,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: inductor, module: pooling, module: random",review linked PR state before touching issue,2026-05-19T18:08:11Z +issue,180823,"[Inductor] RuntimeError: weakref to UntypedStorage when compiling transpose(-2, -1) under inference_mode",https://github.com/pytorch/pytorch/issues/180823,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:07:29Z +issue,179368,torch.compile + view_as_complex: RuntimeError: Tensor must have a last dimension with stride 1,https://github.com/pytorch/pytorch/issues/179368,has linked PR,44,"open PR references this issue; component/oncall labels: module: complex, module: inductor, module: memory format, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:07:20Z +issue,176929,Addcmul CPU bitwise numerics not matching eager w/ torch.compile,https://github.com/pytorch/pytorch/issues/176929,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:06:50Z +issue,172942,cuda graphs dont play nice with numpy numbers,https://github.com/pytorch/pytorch/issues/172942,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:06:29Z +issue,166604,[inductor][pattern_matcher] RuntimeError in joint graph passes when compiling simple `randperm` + advanced indexing pipeline,https://github.com/pytorch/pytorch/issues/166604,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:05:52Z +issue,160124,FX graph segments for multiple triton kernels from reduction node are broken,https://github.com/pytorch/pytorch/issues/160124,has linked PR,44,"open PR references this issue; component/oncall labels: module: fx, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:05:31Z +issue,154146,inductor unbacked codegen results in undefined inputs,https://github.com/pytorch/pytorch/issues/154146,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:05:04Z +issue,148751,[Inductor] Torch inductor pattern match breaks topological sort after replace the pattern,https://github.com/pytorch/pytorch/issues/148751,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-19T18:04:52Z +issue,138301,max_block['X'] not synced with max XBLOCK size (error shows up in `create_block_mask`),https://github.com/pytorch/pytorch/issues/138301,has linked PR,44,open PR references this issue; component/oncall labels: module: inductor,review linked PR state before touching issue,2026-05-19T18:04:19Z +issue,174472,[Inductor][CSE] `split_with_sizes` not deduplicated on B200 with FP8 quantization,https://github.com/pytorch/pytorch/issues/174472,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: vllm, oncall: pt2",review linked PR state before touching issue,2026-05-19T15:22:23Z +issue,170798,nn.functional.pad silently changes the padding value if input dtype is int64 and padding value is larger than 2^53,https://github.com/pytorch/pytorch/issues/170798,has linked PR,44,"open PR references this issue; component/oncall labels: module: 64-bit, module: correctness (silent), module: edge cases, module: nn",review linked PR state before touching issue,2026-05-19T15:17:56Z +issue,170737,Expect to open the is_privateuseone branch in the LSTM interface of PyTorch's RNN.cpp to develop corresponding LSTM functionality on NPU.,https://github.com/pytorch/pytorch/issues/170737,has linked PR,44,open PR references this issue; component/oncall labels: module: privateuse1,review linked PR state before touching issue,2026-05-19T14:25:29Z +issue,184282,LazyBatchNorm2d complex64 input has inconsistent CPU and CUDA errors,https://github.com/pytorch/pytorch/issues/184282,has linked PR,44,"open PR references this issue; component/oncall labels: module: complex, module: cudnn, module: error checking, module: nn",review linked PR state before touching issue,2026-05-19T07:04:21Z +issue,151649,Inductor pattern matcher replaces aten.reshape with aten.view in pattern,https://github.com/pytorch/pytorch/issues/151649,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: vllm, oncall: pt2",review linked PR state before touching issue,2026-05-18T21:43:54Z +issue,133254,Shared memory out of resource when using flex attention,https://github.com/pytorch/pytorch/issues/133254,has linked PR,44,"open PR references this issue; component/oncall labels: module: flex attention, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-18T21:43:02Z +issue,179272,`torch.compile` returns different output stride/contiguity from eager for upsample_nearest3d on non-contiguous input,https://github.com/pytorch/pytorch/issues/179272,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: interpolation, oncall: pt2",review linked PR state before touching issue,2026-05-18T14:11:07Z +issue,178040,`torch.compile` raises RuntimeError on valid `torch.addmm` with shape mismatch where eager succeeds,https://github.com/pytorch/pytorch/issues/178040,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: linear algebra, oncall: pt2",review linked PR state before touching issue,2026-05-18T06:42:07Z +issue,175833,Duplicate SDPA Input Constraints,https://github.com/pytorch/pytorch/issues/175833,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: performance, module: sdpa, oncall: pt2",review linked PR state before touching issue,2026-05-18T06:40:25Z +issue,182940,Eager fake-tensor execution on symbolic torch.export metadata can specialize shapes and add guards outside tracing,https://github.com/pytorch/pytorch/issues/182940,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-18T04:59:34Z +issue,180156,GPU two-pass variance: intermediate squared values overflow float32. CPU Welford online algorithm avoids overflow.,https://github.com/pytorch/pytorch/issues/180156,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: edge cases, module: numerical-stability",review linked PR state before touching issue,2026-05-18T03:48:37Z +issue,182403,local_map deferred inlining can expose index_put_ and force default_partition fallback,https://github.com/pytorch/pytorch/issues/182403,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: functionalization, module: higher order operators, module: performance",review linked PR state before touching issue,2026-05-17T09:10:15Z +issue,184084,Expose public API for clearing cuBLAS workspaces (currently only private torch._C._cuda_clearCublasWorkspaces),https://github.com/pytorch/pytorch/issues/184084,has linked PR,44,"open PR references this issue; component/oncall labels: module: cublas, module: cuda, module: cudacachingallocator",review linked PR state before touching issue,2026-05-17T08:45:01Z +issue,85852,`torch.mm` produces wrong result on cpu when using in-place computation,https://github.com/pytorch/pytorch/issues/85852,has linked PR,44,open PR references this issue; component/oncall labels: module: docs,review linked PR state before touching issue,2026-05-17T03:28:46Z +issue,183901,torch.compile produces int32-overflowed results for int64 arange multiplication on CUDA,https://github.com/pytorch/pytorch/issues/183901,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: cuda, module: inductor, module: int overflow",review linked PR state before touching issue,2026-05-17T02:51:52Z +issue,183120,torch.compile Inductor produces wrong output for BatchNorm2d + ELU + GroupNorm + log clamp chain,https://github.com/pytorch/pytorch/issues/183120,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: inductor, module: norms and normalization, oncall: pt2",review linked PR state before touching issue,2026-05-17T02:50:57Z +issue,180651,"[Inductor][flex attention] flex_attention autotuner uses misleading block-mask, may causing overhead and slower configs",https://github.com/pytorch/pytorch/issues/180651,has linked PR,44,"open PR references this issue; component/oncall labels: module: flex attention, module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2026-05-17T02:47:52Z +issue,184037,torch.nn.LPPool2d returns NaN for non-integer norm_type on signed input (x**p instead of |x|**p),https://github.com/pytorch/pytorch/issues/184037,has linked PR,44,open PR references this issue; component/oncall labels: module: nn,review linked PR state before touching issue,2026-05-16T22:45:13Z +issue,146896,Assertion error in Flex Attention backward pass when indexing a parameter,https://github.com/pytorch/pytorch/issues/146896,has linked PR,44,"open PR references this issue; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-15T02:11:33Z +issue,149767,`flex_attention` slower than manual attention implementation,https://github.com/pytorch/pytorch/issues/149767,has linked PR,44,"open PR references this issue; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-15T02:03:49Z +issue,149586,UserWarning: Dynamo does not know how to trace the builtin `None.pybind11_object.__new__.`,https://github.com/pytorch/pytorch/issues/149586,has linked PR,44,"open PR references this issue; component/oncall labels: module: compiled autograd, module: dynamo, module: flex attention, module: higher order operators",review linked PR state before touching issue,2026-05-14T17:57:32Z +issue,147135,"Don't print fw_metadata in ""Found a graph input that requires gradients""",https://github.com/pytorch/pytorch/issues/147135,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-14T17:41:54Z +issue,148695,dynamo fakification errors with opaquetensorimpl,https://github.com/pytorch/pytorch/issues/148695,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-14T17:41:40Z +issue,182928,[Inductor] Filter extremely bad triton configs by device information for triton heuristics,https://github.com/pytorch/pytorch/issues/182928,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2026-05-14T06:55:57Z +issue,128160,[dynamo] Dynamo traces through __torch_dispatch__ on custom tensor subclasses,https://github.com/pytorch/pytorch/issues/128160,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-14T04:06:16Z +issue,145899,torch.compile + torch.autograd.grad: grad can be implicitly created only for scalar outputs,https://github.com/pytorch/pytorch/issues/145899,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: dynamo, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-13T23:36:12Z +issue,183621,[MPS] For accumulating ops allow the correctness tolerance to scale within the envelope of the reference implementations drift,https://github.com/pytorch/pytorch/issues/183621,has linked PR,44,"open PR references this issue; component/oncall labels: module: half, module: mps, module: tests",review linked PR state before touching issue,2026-05-13T21:42:14Z +issue,144376,torch.compile post_accumulate_grad_hook ordering is wrong for tiebreakers,https://github.com/pytorch/pytorch/issues/144376,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-13T04:02:15Z +issue,183396,`torch.compile` fails on `nn.CTCLoss` with missing fake impl for `aten._use_cudnn_ctc_loss.Tensor`,https://github.com/pytorch/pytorch/issues/183396,has linked PR,44,"open PR references this issue; component/oncall labels: module: cudnn, module: faketensor, module: loss, module: pt2-dispatcher",review linked PR state before touching issue,2026-05-12T01:44:24Z +issue,137635,[MPS] `torch.linspace` produces different results that CPU implementation for integral types,https://github.com/pytorch/pytorch/issues/137635,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: mps",review linked PR state before touching issue,2026-05-11T22:52:39Z +issue,183266,[NestedTensor] torch.compile fails when accessing _max_seqlen with GuardOnDataDependentSymNode,https://github.com/pytorch/pytorch/issues/183266,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: dynamo, module: nestedtensor, oncall: pt2",review linked PR state before touching issue,2026-05-11T11:59:15Z +issue,181806,CUDA torch.signbit returns false for negative float16 NaN,https://github.com/pytorch/pytorch/issues/181806,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: cuda, module: edge cases, module: half",review linked PR state before touching issue,2026-05-11T05:19:13Z +issue,183015,`torch.compile(torch.vmap(torch.func.hessian(f)))` crashes with TorchRuntimeError,https://github.com/pytorch/pytorch/issues/183015,has linked PR,44,"open PR references this issue; component/oncall labels: module: functorch, oncall: pt2",review linked PR state before touching issue,2026-05-09T00:58:31Z +issue,175608,torch.compile fails to trace try-except semantic and throw `RuntimeError`,https://github.com/pytorch/pytorch/issues/175608,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-05-08T17:10:56Z +issue,154306,graph recording observed an input tensor deallocate during graph recording that did not occur during replay,https://github.com/pytorch/pytorch/issues/154306,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-08T08:27:57Z +issue,161765,Add torch compile Dynamo support for protobuf,https://github.com/pytorch/pytorch/issues/161765,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-05-07T20:26:49Z +issue,182706,Dynamo is leaking internal state during optimizer tracing,https://github.com/pytorch/pytorch/issues/182706,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: pt2 optimizer, oncall: pt2",review linked PR state before touching issue,2026-05-06T20:47:50Z +issue,116203,Refactor precision handling in test/test_optim.py,https://github.com/pytorch/pytorch/issues/116203,has linked PR,44,open PR references this issue; component/oncall labels: module: optimizer,review linked PR state before touching issue,2026-05-06T18:47:51Z +issue,77176,Improve the overall design of MPSGraphCache,https://github.com/pytorch/pytorch/issues/77176,has linked PR,44,"open PR references this issue; component/oncall labels: module: backend, module: mps",review linked PR state before touching issue,2026-05-06T10:35:11Z +issue,181565,[C++ API][Packaging] nlohmann/json.hpp is exposed again by public headers in PyTorch 2.10 wheels,https://github.com/pytorch/pytorch/issues/181565,has linked PR,44,"open PR references this issue; component/oncall labels: module: binaries, module: cpp",review linked PR state before touching issue,2026-05-06T03:00:51Z +issue,182430,Inductor generates redundant upcast_to_fp32,https://github.com/pytorch/pytorch/issues/182430,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-05-05T13:31:23Z +issue,180397,[MPS] Add graph capture/replay API to eliminate per-op CPU dispatch overhead,https://github.com/pytorch/pytorch/issues/180397,has linked PR,44,"open PR references this issue; component/oncall labels: module: mps, module: performance",review linked PR state before touching issue,2026-05-05T08:30:25Z +issue,182318,Dynamo codegen leaks torch function enabled state via _EnableTorchFunction destructor,https://github.com/pytorch/pytorch/issues/182318,has linked PR,44,"open PR references this issue; component/oncall labels: module: __torch_function__, module: correctness (silent), module: dynamo, module: graph breaks",review linked PR state before touching issue,2026-05-05T01:33:14Z +issue,167568,Autoparallel's local_map doesn't functionalize triton kernels properly,https://github.com/pytorch/pytorch/issues/167568,has linked PR,44,"open PR references this issue; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-04T23:16:19Z +issue,174017,Custom autograd Functions are not aware of partial backward pass execution,https://github.com/pytorch/pytorch/issues/174017,has linked PR,44,open PR references this issue; component/oncall labels: module: autograd,review linked PR state before touching issue,2026-05-04T22:40:16Z +issue,96409,"Initialization on `meta` device failing for models containing `nn.utils.weight_norm`, with `NotImplementedError: Could not run 'aten::_weight_norm_interface' with arguments from the 'Meta' backend.`",https://github.com/pytorch/pytorch/issues/96409,has linked PR,44,open PR references this issue; component/oncall labels: module: meta tensors,review linked PR state before touching issue,2026-05-04T19:59:17Z +issue,182036,torch.while_loop()'s body_fn is not allowed to return nothing,https://github.com/pytorch/pytorch/issues/182036,has linked PR,44,"open PR references this issue; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-05-04T16:32:35Z +issue,181805,torch.fmax with uint8 out produces inconsistent overflow casting on CPU and CUDA,https://github.com/pytorch/pytorch/issues/181805,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: python frontend",review linked PR state before touching issue,2026-05-03T22:46:12Z +issue,181807,torch.linspace with integer dtype produces inconsistent results on CPU and CUDA,https://github.com/pytorch/pytorch/issues/181807,has linked PR,44,"open PR references this issue; component/oncall labels: module: cpu, module: cuda, module: edge cases, module: tensor creation",review linked PR state before touching issue,2026-05-01T14:49:59Z +issue,181576,torch.meshgrid raises RuntimeError if no arrays passed,https://github.com/pytorch/pytorch/issues/181576,has linked PR,44,"open PR references this issue; component/oncall labels: module: error checking, module: python array api",review linked PR state before touching issue,2026-05-01T04:35:33Z +issue,181891,torch.cond()'s true_fn and false_fn are not allowed to return None,https://github.com/pytorch/pytorch/issues/181891,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda graphs, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-04-29T20:01:43Z +issue,181347,"[Feedback about torch.multinomial][Docs] Inconsistency in torch.multinomial: Note says ""lower than"" but example allows ""equal to"" for num_samples without replacement",https://github.com/pytorch/pytorch/issues/181347,has linked PR,44,open PR references this issue; component/oncall labels: module: docs,review linked PR state before touching issue,2026-04-29T16:09:01Z +issue,180833,[MPS] Replace isin MPSGraph implementation with native Metal kernel and binary-search path for large inputs,https://github.com/pytorch/pytorch/issues/180833,has linked PR,44,"open PR references this issue; component/oncall labels: module: mps, module: performance",review linked PR state before touching issue,2026-04-29T11:03:58Z +issue,181177,[aot_autograd] RuntimeError: derivative for aten::_scaled_dot_product_flash_attention_for_cpu_backward is not implemented,https://github.com/pytorch/pytorch/issues/181177,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: cpu, module: functorch, module: pt2-dispatcher",review linked PR state before touching issue,2026-04-28T14:53:56Z +issue,181725,[MPS] nn.MultiheadAttention is ~9x slower than direct F.scaled_dot_product_attention with bit-identical output (B=1),https://github.com/pytorch/pytorch/issues/181725,has linked PR,44,"open PR references this issue; component/oncall labels: module: mps, module: nn, module: performance",review linked PR state before touching issue,2026-04-28T07:34:26Z +issue,181304,torch.compile fails with FakeTensor Assertion Error in torch/_subclasses/fake_tensor.py,https://github.com/pytorch/pytorch/issues/181304,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: faketensor, module: functorch, module: pt2-dispatcher",review linked PR state before touching issue,2026-04-27T14:45:44Z +issue,163836,[OpenReg] Using InlineEvent (or its base) for Carrying More Event Metadata in DeviceGuardImplInterface APIs,https://github.com/pytorch/pytorch/issues/163836,has linked PR,44,"open PR references this issue; component/oncall labels: module: openreg, module: privateuse1",review linked PR state before touching issue,2026-04-27T11:19:23Z +issue,181374,"`torch.empty((1,1), device=""cpu"", pin_memory=True).device` gives back MPS device on mac",https://github.com/pytorch/pytorch/issues/181374,has linked PR,44,"open PR references this issue; component/oncall labels: module: macos, module: mps",review linked PR state before touching issue,2026-04-27T09:10:40Z +issue,180377,[Silent Bug] torch.linalg.qr allows the same tensor for both Q and R in the 'out' argument when the input is square,https://github.com/pytorch/pytorch/issues/180377,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: error checking, module: linear algebra",review linked PR state before touching issue,2026-04-27T08:33:50Z +issue,131765,[typing] Add missing `__all__` to modules.,https://github.com/pytorch/pytorch/issues/131765,has linked PR,44,"open PR references this issue; component/oncall labels: module: optimizer, module: typing",review linked PR state before touching issue,2026-04-25T03:28:51Z +issue,181474,Add a document for reusing the testcases in PyTorch.,https://github.com/pytorch/pytorch/issues/181474,has linked PR,44,"open PR references this issue; component/oncall labels: module: docs, module: openreg, module: tests",review linked PR state before touching issue,2026-04-25T03:26:38Z +issue,146508,Something very wrong with float16 CPU implementation,https://github.com/pytorch/pytorch/issues/146508,has linked PR,44,"open PR references this issue; component/oncall labels: module: cpu, module: half, module: performance",review linked PR state before touching issue,2026-04-24T01:49:37Z +issue,145093,Inductor aten.clone lowering ignores Conjugate and Negative dispatch keys,https://github.com/pytorch/pytorch/issues/145093,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: inductor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-04-23T21:17:36Z +issue,166209,Add helper functions to massage common 3D+ params into 2D for Muon,https://github.com/pytorch/pytorch/issues/166209,has linked PR,44,open PR references this issue; component/oncall labels: module: optimizer,review linked PR state before touching issue,2026-04-23T12:53:10Z +issue,180957,Missing CSE in TorchInductor,https://github.com/pytorch/pytorch/issues/180957,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2026-04-21T03:52:54Z +issue,180467,remove TORCHDYNAMO_DYNAMIC_SHAPES=1 from docs,https://github.com/pytorch/pytorch/issues/180467,has linked PR,44,"open PR references this issue; component/oncall labels: module: docs, module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2026-04-20T20:41:08Z +issue,180855,"`""index_copy_cpu""` not implemented for `'Float8_e5m2'`",https://github.com/pytorch/pytorch/issues/180855,has linked PR,44,"open PR references this issue; component/oncall labels: module: advanced indexing, module: cpu, module: floatx (formerly float8)",review linked PR state before touching issue,2026-04-20T16:48:43Z +issue,180762,torch.utils.cpp_extension.load should support CUDA device linking (-dlink) similar to CUDAExtension,https://github.com/pytorch/pytorch/issues/180762,has linked PR,44,"open PR references this issue; component/oncall labels: module: cpp-extensions, module: cuda",review linked PR state before touching issue,2026-04-19T06:49:41Z +issue,180649,torch.linalg._powsum docstring advertises out= but it is not implemented,https://github.com/pytorch/pytorch/issues/180649,has linked PR,44,"open PR references this issue; component/oncall labels: module: docs, module: linear algebra",review linked PR state before touching issue,2026-04-17T06:29:57Z +issue,180624,[RFC] Decouple `torch.utils.cpp_extension` from setuptools and align with the scikit-build-core build backend,https://github.com/pytorch/pytorch/issues/180624,has linked PR,44,"open PR references this issue; component/oncall labels: module: build, module: cpp-extensions",review linked PR state before touching issue,2026-04-16T23:16:35Z +issue,180447,Poor BF16 torch.compile + Freezing perf on AArch64 CPUs,https://github.com/pytorch/pytorch/issues/180447,has linked PR,44,"open PR references this issue; component/oncall labels: module: arm, module: bfloat16, module: cpu, module: inductor",review linked PR state before touching issue,2026-04-16T08:01:30Z +issue,165650,"`torch.remainder` with integer divisor 0 should raise, but returns large values in CUDA",https://github.com/pytorch/pytorch/issues/165650,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: error checking, module: nans and infs",review linked PR state before touching issue,2026-04-15T08:30:35Z +issue,179608,[MPS] avg_pool1d produces negative values from non-negative input when zeros follow large floats,https://github.com/pytorch/pytorch/issues/179608,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: edge cases, module: mps, module: pooling",review linked PR state before touching issue,2026-04-13T17:28:05Z +issue,180119,Feedback about torch.Tensor.index_add_,https://github.com/pytorch/pytorch/issues/180119,has linked PR,44,open PR references this issue; component/oncall labels: module: docs,review linked PR state before touching issue,2026-04-11T06:53:33Z +issue,112044,DDE in pack_padded_sequence/pad_packed_sequence decomposition.,https://github.com/pytorch/pytorch/issues/112044,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: fx, module: rnn, oncall: pt2",review linked PR state before touching issue,2026-04-09T19:37:07Z +issue,80025,`torch.special.gammainc` backward pass with respect to the first argument,https://github.com/pytorch/pytorch/issues/80025,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: distributions",review linked PR state before touching issue,2026-04-09T19:02:46Z +issue,175211,CUDA/ROCm/Accelerator testing should replace get_device_capability() with feature queries,https://github.com/pytorch/pytorch/issues/175211,has linked PR,44,"open PR references this issue; component/oncall labels: module: accelerator, module: cuda, module: rocm, module: tests",review linked PR state before touching issue,2026-04-08T20:04:40Z +issue,178379,"[bug]`torch.linalg.solve_triangular(A, B, *, out=B)`: wrong result with `B.is_neg() == True`",https://github.com/pytorch/pytorch/issues/178379,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: cuda, module: linear algebra",review linked PR state before touching issue,2026-04-07T13:15:17Z +issue,179128,FSDP2 HSDP: AllReduceState holds FP32 reduce buffers across all layers during backward,https://github.com/pytorch/pytorch/issues/179128,has linked PR,44,"open PR references this issue; component/oncall labels: module: fsdp, oncall: distributed",review linked PR state before touching issue,2026-04-07T03:05:25Z +issue,179259,test_torch.py does not run any tests on MPS despite number of `skipIfMPS` decorators,https://github.com/pytorch/pytorch/issues/179259,has linked PR,44,"open PR references this issue; component/oncall labels: module: mps, module: tests",review linked PR state before touching issue,2026-04-04T02:31:35Z +issue,178977,`device_id` not honored when creating `process_group` with `$TORCH_DISTRIBUTED_DEBUG=DETAIL`,https://github.com/pytorch/pytorch/issues/178977,has linked PR,44,"open PR references this issue; component/oncall labels: module: c10d, oncall: distributed",review linked PR state before touching issue,2026-04-03T21:20:12Z +issue,169477,torch.compile on @torch.inferece_mode() cause RuntimeError: ,https://github.com/pytorch/pytorch/issues/169477,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-04-03T18:45:02Z +issue,175467,[DTensor] torch.export.export() fails with RuntimeError: wrapped_call_impl - DTensorSpec not registered as pytree constant,https://github.com/pytorch/pytorch/issues/175467,has linked PR,44,"open PR references this issue; component/oncall labels: module: dtensor, oncall: distributed",review linked PR state before touching issue,2026-04-02T18:25:44Z +issue,178677,`torch.compile` silently succeeds on `TransformerEncoder` with all-masked `src_key_padding_mask` where eager raises RuntimeError,https://github.com/pytorch/pytorch/issues/178677,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: dynamo, module: edge cases, module: nestedtensor",review linked PR state before touching issue,2026-03-31T14:41:26Z +issue,108798,Meta kernel not raising,https://github.com/pytorch/pytorch/issues/108798,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-30T18:16:23Z +issue,176211,TypeError in guard when working with list of tensors,https://github.com/pytorch/pytorch/issues/176211,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: guards, oncall: pt2",review linked PR state before touching issue,2026-03-29T10:27:51Z +issue,178128,conv_transpose3d meta execution produces output for invalid bias shape instead of matching eager validation,https://github.com/pytorch/pytorch/issues/178128,has linked PR,44,"open PR references this issue; component/oncall labels: module: convolution, module: error checking, module: meta tensors",review linked PR state before touching issue,2026-03-27T06:45:32Z +issue,178127,torch.conv_transpose2d meta execution fails to validate invalid output_padding/stride/dilation combination,https://github.com/pytorch/pytorch/issues/178127,has linked PR,44,"open PR references this issue; component/oncall labels: module: convolution, module: error checking, module: meta tensors, oncall: pt2",review linked PR state before touching issue,2026-03-27T06:45:28Z +issue,178125,conv_transpose1d meta implementation allows invalid output_padding that real CUDA execution rejects,https://github.com/pytorch/pytorch/issues/178125,has linked PR,44,"open PR references this issue; component/oncall labels: module: convolution, module: error checking, module: meta tensors",review linked PR state before touching issue,2026-03-27T06:45:23Z +issue,178388,[torch.compile] Dynamo fails to trace assignment of custom tensor subclass to module attribute:self.custom_weights = CustomTensor(self.layer1.weight.data)``,https://github.com/pytorch/pytorch/issues/178388,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-25T14:26:34Z +issue,178317,[inductor] Tolerances in test_remove_no_ops have to be lowered with native matmul,https://github.com/pytorch/pytorch/issues/178317,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, module: tests, oncall: pt2",review linked PR state before touching issue,2026-03-24T19:35:22Z +issue,134385,FlopCounterMode doesn't support HOP,https://github.com/pytorch/pytorch/issues/134385,has linked PR,44,"open PR references this issue; component/oncall labels: module: flop counter, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-03-24T18:39:11Z +issue,177968,Add custom precision or tolerance support for out of tree backends,https://github.com/pytorch/pytorch/issues/177968,has linked PR,44,"open PR references this issue; component/oncall labels: module: backend, module: tests",review linked PR state before touching issue,2026-03-24T14:08:26Z +issue,178248,[TorchDynamo] Failed to trace builtin operator 'setattr' when registering forward hook inside forward method,https://github.com/pytorch/pytorch/issues/178248,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-24T07:23:54Z +issue,161863,torch.testing.assert_close doesn't accept tensor-likes,https://github.com/pytorch/pytorch/issues/161863,has linked PR,44,"open PR references this issue; component/oncall labels: module: python frontend, module: testing",review linked PR state before touching issue,2026-03-24T00:31:37Z +issue,177237,torch.conv1d incorrectly accepts invalid weight shape in meta backend while raising correct error in eager execution,https://github.com/pytorch/pytorch/issues/177237,has linked PR,44,"open PR references this issue; component/oncall labels: module: convolution, module: correctness (silent), module: dynamo, module: error checking",review linked PR state before touching issue,2026-03-23T23:16:38Z +issue,128712,Add Swiglu activation function,https://github.com/pytorch/pytorch/issues/128712,has linked PR,44,open PR references this issue; component/oncall labels: module: nn,review linked PR state before touching issue,2026-03-23T22:17:14Z +issue,172324,[OpenReg] Add documentation for standardized testing patterns.,https://github.com/pytorch/pytorch/issues/172324,has linked PR,44,"open PR references this issue; component/oncall labels: module: docs, module: openreg",review linked PR state before touching issue,2026-03-23T07:48:42Z +issue,160094,torch.library.infer_schema doesn't have full support for list types,https://github.com/pytorch/pytorch/issues/160094,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: pt2-dispatcher, module: vllm, oncall: pt2",review linked PR state before touching issue,2026-03-22T23:24:57Z +issue,177750,[functorch] vmap and functionalize under inference_mode,https://github.com/pytorch/pytorch/issues/177750,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: functionalization, module: functorch, module: pt2-dispatcher",review linked PR state before touching issue,2026-03-20T21:37:00Z +issue,160939,Strange failure modes of while_loop under torch.compile -- possibly due to internal array stride,https://github.com/pytorch/pytorch/issues/160939,has linked PR,44,"open PR references this issue; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-03-20T18:39:03Z +issue,91898,Lazily start worker threads in the autograd engine,https://github.com/pytorch/pytorch/issues/91898,has linked PR,44,open PR references this issue; component/oncall labels: module: autograd,review linked PR state before touching issue,2026-03-19T10:40:27Z +issue,91468,torch.compile incorrect when imperative autograd APIs are used,https://github.com/pytorch/pytorch/issues/91468,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: autograd, module: correctness (silent), module: pt2-dispatcher",review linked PR state before touching issue,2026-03-18T21:56:26Z +issue,177712,[dynamo] [torch.compile] sdpa become slower after torch._dynamo.run in dynamo benchmark when attention_mask is None,https://github.com/pytorch/pytorch/issues/177712,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: performance, module: sdpa, oncall: pt2",review linked PR state before touching issue,2026-03-18T02:54:15Z +issue,167070,Inductor doesn't accept Sympy.Eq as input,https://github.com/pytorch/pytorch/issues/167070,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: inductor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-03-17T23:57:06Z +issue,177127,torch.multinomial: add flag to skip input validation (66% of GPU time),https://github.com/pytorch/pytorch/issues/177127,has linked PR,44,open PR references this issue; component/oncall labels: module: performance,review linked PR state before touching issue,2026-03-17T19:21:10Z +issue,116202,"Dynamo'd `test_mixed_device_dtype` needs higher tolerances for Adagrad, RMSProp",https://github.com/pytorch/pytorch/issues/116202,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: optimizer, module: tests, oncall: pt2",review linked PR state before touching issue,2026-03-17T19:17:55Z +issue,167098,request for faster torch.compile kernels for casting to nvfp4,https://github.com/pytorch/pytorch/issues/167098,has linked PR,44,open PR references this issue; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2026-03-17T14:03:40Z +issue,153825,'torch.sparse.to_sparse_semi_structured' significantly worsens performance on H100 GPUs,https://github.com/pytorch/pytorch/issues/153825,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: performance, module: sparse",review linked PR state before touching issue,2026-03-16T20:01:15Z +issue,177265,Compiling CppExtensions in parallel causes race condition,https://github.com/pytorch/pytorch/issues/177265,has linked PR,44,"open PR references this issue; component/oncall labels: module: build, module: cpp-extensions",review linked PR state before touching issue,2026-03-16T17:15:29Z +issue,175858,RFC: `torch._dynamo.mark_dynamic_dict` — skip key-set guards on dict mutations,https://github.com/pytorch/pytorch/issues/175858,has linked PR,44,"open PR references this issue; component/oncall labels: module: compile ux, module: compile-time, module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-15T00:07:25Z +issue,176244,Bug: gaussian_nll_loss — possible issue with broadcastable variance shapes having multiple unit dimensions,https://github.com/pytorch/pytorch/issues/176244,has linked PR,44,"open PR references this issue; component/oncall labels: module: error checking, module: nn, module: python frontend",review linked PR state before touching issue,2026-03-13T16:41:46Z +issue,177327,AArch64 Unit Test Failure - test_int8_woq_mm_*,https://github.com/pytorch/pytorch/issues/177327,has linked PR,44,"open PR references this issue; component/oncall labels: module: arm, module: cpu, module: inductor, module: linear algebra",review linked PR state before touching issue,2026-03-12T22:19:44Z +issue,177250,AArch64 Unit Test Failure - TestNN test_upsampling_bfloat16,https://github.com/pytorch/pytorch/issues/177250,has linked PR,44,"open PR references this issue; component/oncall labels: module: arm, module: autograd, module: nn",review linked PR state before touching issue,2026-03-12T11:37:25Z +issue,127112,[dynamo] Module tracker test errors in dynamo at runtime,https://github.com/pytorch/pytorch/issues/127112,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-11T21:26:22Z +issue,124344,Inconsistent argument parsing behavior for torch.ops bindings between Tensor and FakeTensor,https://github.com/pytorch/pytorch/issues/124344,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2026-03-11T21:19:25Z +issue,177109,[RFC] Make torch.library.custom_op faster in the common case,https://github.com/pytorch/pytorch/issues/177109,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: performance, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-03-11T17:44:41Z +issue,138141,torch.cond with shape condition in predicate doesn't allow for assertions,https://github.com/pytorch/pytorch/issues/138141,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-03-11T17:14:14Z +issue,154329,MPS Memory Leak,https://github.com/pytorch/pytorch/issues/154329,has linked PR,44,"open PR references this issue; component/oncall labels: module: memory usage, module: mps",review linked PR state before touching issue,2026-03-11T05:49:52Z +issue,145773,Dynamo performance test benchmarks reuse model state between eager and compiled,https://github.com/pytorch/pytorch/issues/145773,has linked PR,44,"open PR references this issue; component/oncall labels: module: benchmark, module: cuda graphs, module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-10T22:09:29Z +issue,175001,torch.cond() and cuda graph mega issue,https://github.com/pytorch/pytorch/issues/175001,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda graphs, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-03-10T22:07:04Z +issue,150063,Torch.check does not preserve original error message in the deferred runtime assert,https://github.com/pytorch/pytorch/issues/150063,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-10T16:04:27Z +issue,138795,compile + allgather with group will fail for stack-style allgather,https://github.com/pytorch/pytorch/issues/138795,has linked PR,44,"open PR references this issue; component/oncall labels: module: vllm, oncall: distributed, oncall: pt2",review linked PR state before touching issue,2026-03-10T05:23:43Z +issue,142228,Scaled Dot-Product Attention Invalid Configuration Error on Large batch size,https://github.com/pytorch/pytorch/issues/142228,has linked PR,44,open PR references this issue; component/oncall labels: module: sdpa,review linked PR state before touching issue,2026-03-06T08:29:48Z +issue,152656,"Add explicit error message for def infer_size(a, b): that specificy that non broadcast path was picked due to unbacked existing in both inputs.",https://github.com/pytorch/pytorch/issues/152656,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2026-03-05T06:52:55Z +issue,167729,Dynamo should support torch.autograd.grad,https://github.com/pytorch/pytorch/issues/167729,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-05T01:49:51Z +issue,169538,`torch.compile` fails on `addmm_` with int64 tensors due to fake-tensor meta error (`expand: attempting to expand a dimension of length 2 -> 1`),https://github.com/pytorch/pytorch/issues/169538,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-05T01:16:32Z +issue,117122,Improve behaviour of `torch.linalg.lstsq` on CUDA GPU for rank defficient matrices,https://github.com/pytorch/pytorch/issues/117122,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: linear algebra",review linked PR state before touching issue,2026-03-04T16:20:50Z +issue,176381,Output difference anomaly under equivalent migration in RReLU operator,https://github.com/pytorch/pytorch/issues/176381,has linked PR,44,open PR references this issue; component/oncall labels: module: nn,review linked PR state before touching issue,2026-03-04T05:00:19Z +issue,176296,[MPS] Binary ops on uint16/uint32/uint64 produce garbage values,https://github.com/pytorch/pytorch/issues/176296,has linked PR,44,"open PR references this issue; component/oncall labels: module: correctness (silent), module: mps, module: unsigned int",review linked PR state before touching issue,2026-03-03T23:00:54Z +issue,175477,make_fx symbolic tracing produces wrong 2nd-order gradients through torch.sqrt,https://github.com/pytorch/pytorch/issues/175477,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: decompositions, module: fx, oncall: pt2",review linked PR state before touching issue,2026-03-03T19:41:05Z +issue,175408,custom_op + TensorSubclass + torch.compile interaction : Better error message or doc,https://github.com/pytorch/pytorch/issues/175408,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-03-03T14:05:41Z +issue,175292,[dynamo] Metaclass not supported correctly,https://github.com/pytorch/pytorch/issues/175292,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-02T21:48:45Z +issue,175267,[dynamo] Audit `x is y` code path,https://github.com/pytorch/pytorch/issues/175267,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-02T21:48:21Z +issue,175264,[dynamo] Investigate why we create VariableTrackers for only sourceful objects in UserDefinedClassVariable,https://github.com/pytorch/pytorch/issues/175264,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-03-02T21:48:11Z +issue,174668,torch.histc is not consistent,https://github.com/pytorch/pytorch/issues/174668,has linked PR,44,"open PR references this issue; component/oncall labels: module: edge cases, module: half",review linked PR state before touching issue,2026-03-02T18:23:05Z +issue,175994,CI: Dead CentOS code in docker setup scripts,https://github.com/pytorch/pytorch/issues/175994,has linked PR,44,open PR references this issue; component/oncall labels: module: ci,review linked PR state before touching issue,2026-02-27T21:24:56Z +issue,175968,[opaque object] Running the compiled function with opaque object for the second time fails.,https://github.com/pytorch/pytorch/issues/175968,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: inductor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-27T12:48:45Z +issue,174183,[test] Add error_inputs for nn.Conv2d module,https://github.com/pytorch/pytorch/issues/174183,has linked PR,44,"open PR references this issue; component/oncall labels: module: nn, module: tests",review linked PR state before touching issue,2026-02-26T13:05:43Z +issue,175521,torch.vmap working with non-tensor leaves in output pytrees,https://github.com/pytorch/pytorch/issues/175521,has linked PR,44,"open PR references this issue; component/oncall labels: module: functorch, module: pytree, module: vmap",review linked PR state before touching issue,2026-02-26T09:01:43Z +issue,175368,`F.embedding_bag` segfaults when intermediate offsets exceed indices length,https://github.com/pytorch/pytorch/issues/175368,has linked PR,44,"open PR references this issue; component/oncall labels: module: error checking, module: nn",review linked PR state before touching issue,2026-02-26T06:25:18Z +issue,175489,"CPU `F.interpolate(..., antialias=True)` raises NotImplementedError for bfloat16/float16 (bilinear/bicubic)",https://github.com/pytorch/pytorch/issues/175489,has linked PR,44,"open PR references this issue; component/oncall labels: module: bfloat16, module: cpu, module: half, module: nn",review linked PR state before touching issue,2026-02-23T18:45:02Z +issue,174984,`torch.fft.rfft` errors out with MKL FFT error when given an empty tensor,https://github.com/pytorch/pytorch/issues/174984,has linked PR,44,"open PR references this issue; component/oncall labels: module: edge cases, module: empty tensor, module: error checking, module: fft",review linked PR state before touching issue,2026-02-23T17:49:21Z +issue,175522,while_loop vmap compatibility,https://github.com/pytorch/pytorch/issues/175522,has linked PR,44,"open PR references this issue; component/oncall labels: module: functorch, module: higher order operators, module: pt2-dispatcher, module: vmap",review linked PR state before touching issue,2026-02-23T10:04:27Z +issue,139707,Adding backend-specific autograd kernel for aten builtin operators breaks `torch.compile`,https://github.com/pytorch/pytorch/issues/139707,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-23T04:00:20Z +issue,138264,`_native_batch_norm_legit`: `aot_eager` backend incorrectly raising an error.,https://github.com/pytorch/pytorch/issues/138264,has linked PR,44,"open PR references this issue; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-22T20:30:55Z +issue,138229,[cond] Cond lifts duplicate symint inputs,https://github.com/pytorch/pytorch/issues/138229,has linked PR,44,"open PR references this issue; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-22T20:27:43Z +issue,122386,[histogram] expand: attempting to expand a dimension of length 6!,https://github.com/pytorch/pytorch/issues/122386,has linked PR,44,"open PR references this issue; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-22T20:21:40Z +issue,133512,functionalization doesn't faithfully reproduce strides of intermediate values,https://github.com/pytorch/pytorch/issues/133512,has linked PR,44,"open PR references this issue; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-22T04:30:40Z +issue,132896,AOTAutograd.process_inputs doesn't work on nested subclasses,https://github.com/pytorch/pytorch/issues/132896,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-22T04:00:05Z +issue,122860,Do not use assert for user triggered issues,https://github.com/pytorch/pytorch/issues/122860,has linked PR,44,"open PR references this issue; component/oncall labels: module: functorch, module: pt2-dispatcher, oncall: export, oncall: pt2",review linked PR state before touching issue,2026-02-22T03:48:28Z +issue,117555,Update decompositions registered to `Autograd` dispatch key to also register `CompositeImplicitAutograd`,https://github.com/pytorch/pytorch/issues/117555,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: decompositions, module: pt2-dispatcher, module: python dispatcher",review linked PR state before touching issue,2026-02-22T03:38:48Z +issue,174541,torch.compile: ConstDictVariable should support runtime dict lookup for dynamic keys to avoid recompilation,https://github.com/pytorch/pytorch/issues/174541,has linked PR,44,"open PR references this issue; component/oncall labels: module: compile-time, module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-02-21T16:09:37Z +issue,90923,dynamo.optimizations.training.aot_autograd does not trace correct overload,https://github.com/pytorch/pytorch/issues/90923,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: export, oncall: pt2",review linked PR state before touching issue,2026-02-21T05:00:03Z +issue,124181,AOTAutograd breaks diffusers VAE decoder,https://github.com/pytorch/pytorch/issues/124181,has linked PR,44,"open PR references this issue; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-21T02:49:33Z +issue,122029,AOTAutograd has high fixed overheads,https://github.com/pytorch/pytorch/issues/122029,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-21T02:20:57Z +issue,90507,Functionalization on inplace_views should properly reflect autograd metadata,https://github.com/pytorch/pytorch/issues/90507,has linked PR,44,"open PR references this issue; component/oncall labels: module: bootcamp, module: functionalization, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-20T19:55:28Z +issue,113899,Tensor.requires_grad_() does not trace with make_fx,https://github.com/pytorch/pytorch/issues/113899,has linked PR,44,"open PR references this issue; component/oncall labels: module: proxytensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-20T19:43:56Z +issue,114415,AOTAutograd should detect aliasing of inputs that happens *below* subclasses,https://github.com/pytorch/pytorch/issues/114415,has linked PR,44,"open PR references this issue; component/oncall labels: module: __torch_dispatch__, module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-20T18:36:37Z +issue,140707,mutating a buffer to change its requires_grad-ness can cause RuntimeError,https://github.com/pytorch/pytorch/issues/140707,has linked PR,44,"open PR references this issue; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-20T03:43:28Z +issue,173816,make should_preserve_node_meta and relevant variable TLS,https://github.com/pytorch/pytorch/issues/173816,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, module: fx, oncall: pt2",review linked PR state before touching issue,2026-02-19T10:37:47Z +issue,175324,Should `overrides.is_tensor_like` return a `TypeGuard[Tensor]`?,https://github.com/pytorch/pytorch/issues/175324,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: typing",review linked PR state before touching issue,2026-02-19T10:24:07Z +issue,172538,Support native structured kernel and out_as_primary: true for PrivateUse1,https://github.com/pytorch/pytorch/issues/172538,has linked PR,44,open PR references this issue; component/oncall labels: module: privateuse1,review linked PR state before touching issue,2026-02-18T20:24:08Z +issue,145383,Windows Pytorch compiler crash some version of cl.exe. Fix provided,https://github.com/pytorch/pytorch/issues/145383,has linked PR,44,"open PR references this issue; component/oncall labels: module: windows, oncall: pt2",review linked PR state before touching issue,2026-02-18T01:34:13Z +issue,172026,`torch.compile(torch.func.vjp)` crashes when creating a new `Tensor`,https://github.com/pytorch/pytorch/issues/172026,has linked PR,44,"open PR references this issue; component/oncall labels: module: autograd, module: functorch, oncall: pt2",review linked PR state before touching issue,2026-02-17T18:53:05Z +issue,174682,spin lint doesn't regenerate cuda headers,https://github.com/pytorch/pytorch/issues/174682,has linked PR,44,"open PR references this issue; component/oncall labels: module: build, module: lint, module: spin",review linked PR state before touching issue,2026-02-17T16:52:13Z +issue,174176,Inductor crashes with INTERNAL ASSERT in FunctionalizeFallbackKernel,https://github.com/pytorch/pytorch/issues/174176,has linked PR,44,"open PR references this issue; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-16T19:39:12Z +issue,174371,"[inductor][aot_eager] Incorrect results with torch.compile when using clone, as_strided followed by add_",https://github.com/pytorch/pytorch/issues/174371,has linked PR,44,"open PR references this issue; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-16T09:31:33Z +issue,140000,torch.argmax slow on CPU,https://github.com/pytorch/pytorch/issues/140000,has linked PR,44,"open PR references this issue; component/oncall labels: module: cpu, module: performance",review linked PR state before touching issue,2026-02-14T21:46:49Z +issue,148651,Avoid fork for TORCHINDUCTOR_COMPILE_THREADS > 1,https://github.com/pytorch/pytorch/issues/148651,has linked PR,44,"open PR references this issue; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2026-02-13T22:31:42Z +issue,171355,torch.utils.checkpoint.CheckpointError with qwen3 MoE,https://github.com/pytorch/pytorch/issues/171355,has linked PR,44,open PR references this issue; component/oncall labels: module: activation checkpointing,review linked PR state before touching issue,2026-02-12T13:36:07Z +issue,174625,document `torch.cuda.memory.mem_get_info` in,https://github.com/pytorch/pytorch/issues/174625,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: docs",review linked PR state before touching issue,2026-02-12T00:05:09Z +issue,168972,Aarch64 unit test failures - test/test_cpu_repro.py CPUReproTests.test_float32_to_uint8,https://github.com/pytorch/pytorch/issues/168972,has linked PR,44,"open PR references this issue; component/oncall labels: module: arm, module: inductor, oncall: pt2",review linked PR state before touching issue,2026-02-11T10:37:54Z +issue,169188,"`torch.compile(backend=""tvm"")` produces wrong result on simple `nn.Parameter` addition (eager = 2.0, compiled = 0.0)",https://github.com/pytorch/pytorch/issues/169188,has linked PR,44,"open PR references this issue; component/oncall labels: module: backend, oncall: pt2",review linked PR state before touching issue,2026-02-07T15:44:11Z +issue,169111,"`torch.compile(backend=""tvm"", fullgraph=True, dynamic=True)` fails on PyTorch 2.9.0 with `RuntimeError: Tracer cannot infer type ... Only tensors ... supported, but got SymInt`",https://github.com/pytorch/pytorch/issues/169111,has linked PR,44,"open PR references this issue; component/oncall labels: module: backend, oncall: pt2",review linked PR state before touching issue,2026-02-07T15:06:51Z +issue,174177,[test] Add error_inputs for nn.Linear module,https://github.com/pytorch/pytorch/issues/174177,has linked PR,44,open PR references this issue; component/oncall labels: module: tests,review linked PR state before touching issue,2026-02-04T15:23:46Z +issue,173912,Wrong mode in `torch.distributions.Kumaraswamy`,https://github.com/pytorch/pytorch/issues/173912,has linked PR,44,open PR references this issue; component/oncall labels: module: distributions,review linked PR state before touching issue,2026-02-02T18:29:03Z +issue,173578,FunctionIdSet should store weakrefs instead of object ids.,https://github.com/pytorch/pytorch/issues/173578,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-01-30T12:13:52Z +issue,146249,Replace `*args : Any` with `typing_extensions.TypeVarTuple`,https://github.com/pytorch/pytorch/issues/146249,has linked PR,44,open PR references this issue; component/oncall labels: module: typing,review linked PR state before touching issue,2026-01-30T12:13:17Z +issue,173881,[Build][ppc64le] Fixes required for compiling PyTorch 2.5.1 on PowerPC with GCC 12,https://github.com/pytorch/pytorch/issues/173881,has linked PR,44,"open PR references this issue; component/oncall labels: module: build, module: power, oncall: releng",review linked PR state before touching issue,2026-01-30T04:51:40Z +issue,170869,Feedback about torch.addcdiv,https://github.com/pytorch/pytorch/issues/170869,has linked PR,44,"open PR references this issue; component/oncall labels: module: docs, module: edge cases",review linked PR state before touching issue,2026-01-29T13:17:37Z +issue,173052,[Inconsistency] aten.linalg_pinv returns inconsistent output shape between eager and inductor,https://github.com/pytorch/pytorch/issues/173052,has linked PR,44,"open PR references this issue; component/oncall labels: module: decompositions, module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-01-29T10:50:06Z +issue,173252,[dynamo] LazyConvTranspose1d fails under torch.compile(dynamic=True) with “SymIntArrayRef expected to contain only concrete integers”,https://github.com/pytorch/pytorch/issues/173252,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-01-27T20:21:03Z +issue,173049,OOM error suggests expandable_segments even when enabled,https://github.com/pytorch/pytorch/issues/173049,has linked PR,44,"open PR references this issue; component/oncall labels: module: cuda, module: cudacachingallocator, module: error checking",review linked PR state before touching issue,2026-01-22T22:24:32Z +issue,172956,Mutation during backward on tensor saved for backward can create invalid graph,https://github.com/pytorch/pytorch/issues/172956,has linked PR,44,"open PR references this issue; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-01-21T17:53:59Z +issue,167818,undefined symbol for `at::meta::_index_put_impl_` when running or compiling executable on my own torch-related project.,https://github.com/pytorch/pytorch/issues/167818,has linked PR,44,"open PR references this issue; component/oncall labels: module: binaries, module: privateuse1",review linked PR state before touching issue,2026-01-21T10:14:01Z +issue,172869,[opaque obj] Figure out subclassing OpaqueBase in pybinded classes,https://github.com/pytorch/pytorch/issues/172869,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-01-20T22:00:15Z +issue,172428,`torch.compile` on a `custom_op`'s fake when using `torch.func.vjp` crashes,https://github.com/pytorch/pytorch/issues/172428,has linked PR,44,"open PR references this issue; component/oncall labels: module: custom-operators, module: functorch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-01-18T04:02:12Z +issue,170509,Dynamo fails on unit tests: test/test_torch.py.test_geometric_kstest,https://github.com/pytorch/pytorch/issues/170509,has linked PR,44,"open PR references this issue; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-01-16T00:42:27Z +issue,172321,[OpenReg] Extend test_autograd.py coverage,https://github.com/pytorch/pytorch/issues/172321,has linked PR,44,"open PR references this issue; component/oncall labels: module: ci, module: openreg",review linked PR state before touching issue,2026-01-13T15:08:31Z +issue,119191,AOTAutograd: functionalization should emit foreach_copy_() instead of copy_() when possible,https://github.com/pytorch/pytorch/issues/119191,has linked PR,44,"open PR references this issue; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: functionalization, module: mta, module: pt2-dispatcher",review linked PR state before touching issue,2025-11-19T11:27:31Z +issue,135011,"dynamo (re)compilation issues: shape (1,1), nn.Parameter, mark_dynamic",https://github.com/pytorch/pytorch/issues/135011,has linked PR,44,"open PR references this issue; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-10-10T16:58:07Z +issue,3867,"[Feature Request] Implement ""same"" padding for convolution operations?",https://github.com/pytorch/pytorch/issues/3867,has linked PR,44,"open PR references this issue; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",route to design/research decision before patch planning,2025-09-29T22:55:35Z +issue,91692,[discussion] Analyzing a list of tensors stored as intermediate values / saved_for_backward in autograd graph,https://github.com/pytorch/pytorch/issues/91692,has linked PR,44,open PR references this issue; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: autograd,review linked PR state before touching issue,2025-07-29T10:19:02Z +issue,129020,Expand Tag Set: views & reductions,https://github.com/pytorch/pytorch/issues/129020,has linked PR,44,open PR references this issue; label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: internals,review linked PR state before touching issue,2025-05-11T15:54:54Z +issue,111414,Minifier doesn't transfer execution states like @torch.no_grad to repro,https://github.com/pytorch/pytorch/issues/111414,has linked PR,44,"open PR references this issue; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: minifier, oncall: pt2",route to design/research decision before patch planning,2024-07-11T02:16:10Z +issue,89489,build: cmake: need to uniformize installation of libraries in CMAKE_INSTALL_LIBDIR (not lib),https://github.com/pytorch/pytorch/issues/89489,has linked PR,44,"open PR references this issue; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",route to design/research decision before patch planning,2022-11-22T17:54:43Z +issue,62027,`x.to(memory_format=torch.contiguous_format)` does not always return a contiguous tensor,https://github.com/pytorch/pytorch/issues/62027,has linked PR,44,"open PR references this issue; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: memory format",route to design/research decision before patch planning,2022-10-31T17:22:24Z +pr,189096,[aot_autograd] Make CPU-side stream syncs real cross-stream barriers,https://github.com/pytorch/pytorch/pull/189096,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-09T02:18:21Z +pr,187940,[xpu][inductor] fix the fallback behavior of maxpool backward for performance,https://github.com/pytorch/pytorch/pull/187940,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk, ciflow/xpu",wait for contributor update,2026-07-09T02:08:16Z +pr,189284,Add torch.autograd.graph.node_creation_hook,https://github.com/pytorch/pytorch/pull/189284,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-09T02:04:54Z +pr,187600,[ROCm][Inductor] Enable NHWC convs by default on CDNA with an Inductor layout-opt gate,https://github.com/pytorch/pytorch/pull/187600,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/inductor-perf-test-nightly-rocm-mi300, ciflow/inductor-perf-test-nightly-rocm-mi350",wait for contributor update,2026-07-09T01:54:02Z +pr,189109,[profiler][cupti] Resolve a version-gated CUPTI header and conditionally install libclang for the field-id codegen,https://github.com/pytorch/pytorch/pull/189109,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-09T01:52:57Z +pr,188742,[ROCm][inductor] Enable scaled grouped GEMM Triton lowering,https://github.com/pytorch/pytorch/pull/188742,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm",wait for contributor update,2026-07-09T01:04:51Z +pr,188600,[ROCm][inductor] Support grouped GEMM Triton lowering,https://github.com/pytorch/pytorch/pull/188600,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300",wait for contributor update,2026-07-08T23:51:01Z +pr,180826,"Remove deprecated `set_group_info`, `get_group_info` and `enable_symm_mem_for_group`",https://github.com/pytorch/pytorch/pull/180826,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/b200-symm-mem, ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T22:50:12Z +pr,186965,[RFC][dynamo] Add ContextVar.get() tracing support (Phase 1),https://github.com/pytorch/pytorch/pull/186965,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-08T22:27:06Z +pr,187450,[ROCm] periodic-rocm-mi350: run only the 4-GPU distributed tests,https://github.com/pytorch/pytorch/pull/187450,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/periodic-rocm-mi350, ciflow/trunk",wait for contributor update,2026-07-08T21:52:15Z +pr,185617,[Profiler] Refactor test_trace_validator.py,https://github.com/pytorch/pytorch/pull/185617,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-08T21:49:37Z +pr,188004,rewrite exception handling to be more closer to CPython,https://github.com/pytorch/pytorch/pull/188004,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T21:06:03Z +pr,188954,[MPS] add matrix exp on MPS,https://github.com/pytorch/pytorch/pull/188954,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/mps,wait for contributor update,2026-07-08T20:50:52Z +pr,189207,Scope cuDNN Engine 5 errata to SM 8.6 GPUs only,https://github.com/pytorch/pytorch/pull/189207,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-08T20:50:47Z +pr,183663,[inductor] _make_launchers: skip rebuilding existing launchers,https://github.com/pytorch/pytorch/pull/183663,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T20:49:42Z +pr,179425,[inductor][autotuning] add logging and config for incremental autotuning (#179425),https://github.com/pytorch/pytorch/pull/179425,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-08T20:41:11Z +pr,182278,[CI] Add native build image for linux-riscv64,https://github.com/pytorch/pytorch/pull/182278,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/docker,wait for contributor update,2026-07-08T20:36:32Z +pr,186540,[MPS] Add segment_reduce support with native Metal kernels,https://github.com/pytorch/pytorch/pull/186540,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/mps",wait for contributor update,2026-07-08T19:52:04Z +pr,180247,Migrate build system from setuptools to scikit-build-core,https://github.com/pytorch/pytorch/pull/180247,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/binaries, ciflow/docker, ciflow/dynamo, ciflow/inductor",wait for contributor update,2026-07-08T19:51:02Z +pr,180248,Remove setup.py and setuptools build helpers,https://github.com/pytorch/pytorch/pull/180248,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/binaries,wait for contributor update,2026-07-08T19:12:47Z +pr,188105,Add support for custom attributes on Exceptions,https://github.com/pytorch/pytorch/pull/188105,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T18:44:03Z +pr,178737,"Enable hipsparselt version, initCusparseltBindings and related UTs",https://github.com/pytorch/pytorch/pull/178737,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/rocm-mi300, ciflow/rocm-mi355, ciflow/rocm-nightly, ciflow/trunk",wait for contributor update,2026-07-08T17:58:34Z +pr,185701,[Test] Make test_packed_sequence.py device agnostic for out-of-tree backends,https://github.com/pytorch/pytorch/pull/185701,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-08T15:58:34Z +pr,189183,[inductor][mtia] Fix fp16 reduction accumulator dtype mismatch,https://github.com/pytorch/pytorch/pull/189183,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-08T06:21:15Z +pr,188747,Python c10d Backend trampoline cleanup,https://github.com/pytorch/pytorch/pull/188747,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-08T02:21:56Z +pr,188374,[HOP] Deduplicate lifted args for switch,https://github.com/pytorch/pytorch/pull/188374,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T23:15:33Z +pr,184779,"[MPS] Rewrite _bincount_mps with native Metal atomic-add (~18 s → 2.1 ms, 8,500× speedup)",https://github.com/pytorch/pytorch/pull/184779,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/mps,wait for contributor update,2026-07-07T21:14:06Z +pr,188042,[CI] Split AI advisor garbage verdict into infra_issue + garbage,https://github.com/pytorch/pytorch/pull/188042,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-07T20:50:42Z +pr,176678,[CI][Windows] Add NVIDIA RTX workflow,https://github.com/pytorch/pytorch/pull/176678,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/win-arm64",wait for contributor update,2026-07-07T20:11:55Z +pr,177540,[inductor] Handle Triton cluster_dims metadata from both binary and metadata layouts,https://github.com/pytorch/pytorch/pull/177540,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/trunk",wait for contributor update,2026-07-07T18:13:39Z +pr,188331,[TEST] Refactor test_profiler_tree.py with hw_classification,https://github.com/pytorch/pytorch/pull/188331,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-07T18:05:28Z +pr,187828,Add privateuse1 backend support to has_triton(),https://github.com/pytorch/pytorch/pull/187828,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-07T17:43:48Z +pr,185813,[ROCm][Windows] Audit license-files: explicit enumeration + audit test (#183434),https://github.com/pytorch/pytorch/pull/185813,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/docker,wait for contributor update,2026-07-07T17:15:10Z +pr,187434,Support autograd for Tensor?[] (optional tensor list) outputs,https://github.com/pytorch/pytorch/pull/187434,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-07T14:52:16Z +pr,182725,[c10d] Throttle TCPStore connect-retry warnings,https://github.com/pytorch/pytorch/pull/182725,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-07T06:06:22Z +pr,188095,[ROCm][FlexAttention] Default PRESCALE_QK and auto-detect contiguous blocks,https://github.com/pytorch/pytorch/pull/188095,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-07T03:23:43Z +pr,187755,[testintro] CI: post tests added/removed as a PR comment,https://github.com/pytorch/pytorch/pull/187755,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-06T20:30:01Z +pr,186324,[FakeTensor] Propagate backend metadata through fake tensor conversion,https://github.com/pytorch/pytorch/pull/186324,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-06T20:11:57Z +pr,186047,[Test][PrivateUse1] Enhance `get_all_devices` to include all available devices,https://github.com/pytorch/pytorch/pull/186047,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-06T08:38:02Z +pr,166396,[xpu][test][1/N] Enable tests of test_nn.py on Intel GPU - instantiate TestNN with instantiate_device_type_tests,https://github.com/pytorch/pytorch/pull/166396,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/mps, ciflow/pull, ciflow/xpu",wait for contributor update,2026-07-05T15:52:59Z +pr,176619,Fix: raise RuntimeError when cudagraph partitions are empty,https://github.com/pytorch/pytorch/pull/176619,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-05T05:09:03Z +pr,178849,Enable eager tests for inductor-unittest - - Enabel more tests in test_ops.py on Intel GPU,https://github.com/pytorch/pytorch/pull/178849,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/xpu,wait for contributor update,2026-07-03T21:38:20Z +pr,187654,[dynamo] Remove hardcoded CUDA references in test_logging.py,https://github.com/pytorch/pytorch/pull/187654,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-03T16:14:17Z +pr,176420,[XPU] Ensure dense (canonical) strides for convolution inputs,https://github.com/pytorch/pytorch/pull/176420,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-03T10:11:35Z +pr,169916,[CPU] Support multiple ncores-per-instance settings in xeon/run_cpu.py,https://github.com/pytorch/pytorch/pull/169916,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-03T01:18:14Z +pr,172603,"Rewrite _get_total_norm with _powsum, add skip_root flag",https://github.com/pytorch/pytorch/pull/172603,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/inductor,wait for contributor update,2026-07-02T18:08:04Z +pr,187553,[MPS] Migrate mse_loss to a native fused Metal kernel,https://github.com/pytorch/pytorch/pull/187553,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/mps,wait for contributor update,2026-07-02T16:52:03Z +pr,188395,Refactor Dynamo side effect replay into registry dispatcher,https://github.com/pytorch/pytorch/pull/188395,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-02T16:36:17Z +pr,188608,[ci] Re-enable cuDSS installation in CI Docker images so CUDA CI can actually build and run cuDSS tests,https://github.com/pytorch/pytorch/pull/188608,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/docker,wait for contributor update,2026-07-02T15:06:46Z +pr,179284,Fix AOT autograd leaf backward mutation for auto_functionalized V2 (#179284),https://github.com/pytorch/pytorch/pull/179284,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-02T14:16:47Z +pr,178550,[DTensor] Support sharding propagation with meta device mesh,https://github.com/pytorch/pytorch/pull/178550,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-02T14:16:41Z +pr,176747,Fix root cause of extraneous termination handler error logging (eval) (#176747),https://github.com/pytorch/pytorch/pull/176747,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-07-02T14:16:36Z +pr,170752,Support torch.accelerator.get_device_capability on CUDA,https://github.com/pytorch/pytorch/pull/170752,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/trunk,wait for contributor update,2026-07-01T17:15:49Z +pr,187787,[MPS] Migrate var/std to native Metal kernels (Welford),https://github.com/pytorch/pytorch/pull/187787,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/mps",wait for contributor update,2026-07-01T16:44:40Z +pr,187573,[TMA] Fix AOTI eager scratch buffer missing CACHE_TORCH_DEVICE declaration,https://github.com/pytorch/pytorch/pull/187573,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-07-01T11:59:50Z +pr,186652,"[ROCm] Fix ""Ignoring invalid value for boolean flag AMD_SERIALIZE_KERNEL: 3""",https://github.com/pytorch/pytorch/pull/186652,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/rocm-mi300, ciflow/trunk",wait for contributor update,2026-07-01T03:02:37Z +pr,187976,[MPS] Fix int64 amax/amin returning 0 for partial-simdgroup reductions,https://github.com/pytorch/pytorch/pull/187976,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/mps,wait for contributor update,2026-06-30T23:23:03Z +pr,187772,perf(dynamo): add opt-in guard lookup token plan,https://github.com/pytorch/pytorch/pull/187772,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-29T21:27:08Z +pr,187223,Add targeted SDPA coverage to H100 CI,https://github.com/pytorch/pytorch/pull/187223,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-29T20:14:11Z +pr,188365,[MPS] Remove support for macOS 14,https://github.com/pytorch/pytorch/pull/188365,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/mps",wait for contributor update,2026-06-29T17:18:35Z +pr,171035,Add 2D Right-hand-side support for cuDSS backend `linalg.solve`,https://github.com/pytorch/pytorch/pull/171035,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-29T13:17:25Z +pr,188015,[SDPA] Fix 32-bit dropout RNG offset overflow in mem-efficient attention,https://github.com/pytorch/pytorch/pull/188015,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-28T23:05:33Z +pr,185015,Add SAC-driven fully overlapped activation CPU offload (primarily aimed at GH200/GB200/GB300 superchips),https://github.com/pytorch/pytorch/pull/185015,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-25T21:56:20Z +pr,184739,Refactor test_sdpa for out-of-tree backends,https://github.com/pytorch/pytorch/pull/184739,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/trunk",wait for contributor update,2026-06-25T17:53:48Z +pr,185505,[inductor] Enable ROCm CK GEMM templates under JIT cpp_wrapper,https://github.com/pytorch/pytorch/pull/185505,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/inductor-rocm-mi300, ciflow/torchtitan",wait for contributor update,2026-06-25T13:55:58Z +pr,187747,Replace CUDA APIs with generic torch.accelerator equivalents,https://github.com/pytorch/pytorch/pull/187747,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-25T03:48:22Z +pr,188086,Migrate torch_export_python to ShapesSpec API,https://github.com/pytorch/pytorch/pull/188086,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-24T21:08:38Z +pr,187603,[executorch][dynamo] Import torch._utils before Dynamo variable registration,https://github.com/pytorch/pytorch/pull/187603,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-24T19:15:01Z +pr,182984,[Inductor] Expose config for sm carveout of templates (#182984),https://github.com/pytorch/pytorch/pull/182984,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300",wait for contributor update,2026-06-24T19:04:36Z +pr,187754,[testintro] Add a test-introspection tool,https://github.com/pytorch/pytorch/pull/187754,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/docker,wait for contributor update,2026-06-24T14:17:39Z +pr,185910,[AARCH64] Optimize AArch64 qlinear performance by triggering blocked layout in prepacked weight,https://github.com/pytorch/pytorch/pull/185910,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-24T13:28:28Z +pr,181138,[Inductor][NFC] Refactor fusion legality checks for `UserDefinedTritonKernel`,https://github.com/pytorch/pytorch/pull/181138,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-24T07:21:46Z +pr,182809,[2/N] Replace data_ptr with mutable_data_ptr,https://github.com/pytorch/pytorch/pull/182809,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300",wait for contributor update,2026-06-24T06:07:07Z +pr,187696,Add DataLoader worker signal message affixes,https://github.com/pytorch/pytorch/pull/187696,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-24T02:59:29Z +pr,178463,[Distributed] Make Ckpt Tests Backend Agnostic,https://github.com/pytorch/pytorch/pull/178463,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/torchtitan,wait for contributor update,2026-06-23T19:14:39Z +pr,180080,Bitwise operator tensorcompare.cpp,https://github.com/pytorch/pytorch/pull/180080,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-21T20:56:43Z +pr,173258,[CUDA] Math.cuh add missing return statement,https://github.com/pytorch/pytorch/pull/173258,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/trunk,wait for contributor update,2026-06-20T09:40:27Z +pr,174491,fix: prevent issue with destruction of std::shared_timed_mutex,https://github.com/pytorch/pytorch/pull/174491,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/trunk,wait for contributor update,2026-06-19T13:44:36Z +pr,182025,[MPS] Use MetalPerformancePrimitives for F.linear on macOS 26+,https://github.com/pytorch/pytorch/pull/182025,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/mps,wait for contributor update,2026-06-19T01:01:52Z +pr,187414,[Inductor][cond] Allow output-input aliasing in torch.cond branches (re-land D106106380 + dev-nosan test fix),https://github.com/pytorch/pytorch/pull/187414,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-18T18:34:15Z +pr,179546,Add structured call_hierarchy metadata to FX nodes,https://github.com/pytorch/pytorch/pull/179546,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-17T21:21:31Z +pr,179422,[inductor] Force contiguous inputs for custom ops in FallbackKernel,https://github.com/pytorch/pytorch/pull/179422,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-17T21:19:45Z +pr,186104,[Inductor] Allow output-input aliasing in torch.cond branches (#186104),https://github.com/pytorch/pytorch/pull/186104,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-17T21:07:05Z +pr,186950,Fixed issue with pickling of custom decomp dict,https://github.com/pytorch/pytorch/pull/186950,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-17T19:18:49Z +pr,183673,[ROCm][Windows] torchgen: fix Windows DLL linkage for native and namespaced forward declarations,https://github.com/pytorch/pytorch/pull/183673,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-17T14:34:13Z +pr,187492,Avoid std::views::reverse in CUDACachingAllocator,https://github.com/pytorch/pytorch/pull/187492,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-17T01:16:14Z +pr,173519,Add on-demand ppc64le wheel build support,https://github.com/pytorch/pytorch/pull/173519,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-16T13:06:41Z +pr,185741,[PyTorch][Profiler] Fix use-after-free in global RecordFunction callback teardown (#185741),https://github.com/pytorch/pytorch/pull/185741,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/trunk,wait for contributor update,2026-06-15T23:19:55Z +pr,180129,[scaled_grouped_mm] add support for out argument,https://github.com/pytorch/pytorch/pull/180129,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300",wait for contributor update,2026-06-15T22:07:32Z +pr,184654,Short-circuit pytree.tree_map_only when no leaves match,https://github.com/pytorch/pytorch/pull/184654,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-15T16:34:26Z +pr,186642,[inductor][rocm] make AMD MM matrix_instr_nonkdim configurable,https://github.com/pytorch/pytorch/pull/186642,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-15T13:09:55Z +pr,183669,[inductor] Make experimental benchmarker parameters configurable via env vars,https://github.com/pytorch/pytorch/pull/183669,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-11T17:02:53Z +pr,183556,[inductor] Fix MemoryTracker assertion with shared storages in reordered schedules,https://github.com/pytorch/pytorch/pull/183556,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-11T16:59:49Z +pr,178048,"[inductor] Add custom_pass_context for scoped, composable custom pass registration",https://github.com/pytorch/pytorch/pull/178048,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/inductor,wait for contributor update,2026-06-11T16:26:19Z +pr,182233,Fix torch.combinations dynamic shape support by replacing numel with sym_numel and propagating SymInt,https://github.com/pytorch/pytorch/pull/182233,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/trunk",wait for contributor update,2026-06-11T10:59:22Z +pr,184895,[Inductor] Index expression codegen for user Triton kernel fusion,https://github.com/pytorch/pytorch/pull/184895,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-11T05:29:28Z +pr,186808,[inductor] Re-key range_tree_nodes on IterationRangesEntry rename,https://github.com/pytorch/pytorch/pull/186808,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-10T14:44:27Z +pr,182712,[MPS] Enhance Linear tensor ops to avoid unnecessary .contiguous() calls on weight tensor,https://github.com/pytorch/pytorch/pull/182712,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/mps,wait for contributor update,2026-06-09T23:57:34Z +pr,186138,Support data-dependent 0-dim tensor indexing under tracing,https://github.com/pytorch/pytorch/pull/186138,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-09T23:08:41Z +pr,186251,Wait for Docker images before starting builds,https://github.com/pytorch/pytorch/pull/186251,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/trunk,wait for contributor update,2026-06-09T00:18:59Z +pr,182208,Add missing stacklevel to 57 warnings.warn() calls in torch/,https://github.com/pytorch/pytorch/pull/182208,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-06-03T21:42:10Z +pr,172945,Bf16 optimized heuristics,https://github.com/pytorch/pytorch/pull/172945,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-03T15:44:33Z +pr,185475,[pytorch] Nested custom Python process groups (experimental) (#185475),https://github.com/pytorch/pytorch/pull/185475,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/torchtitan,wait for contributor update,2026-06-03T15:25:40Z +pr,181730,"Prevent DCE eliminate ""triton_kernel_wrapper_mutation""",https://github.com/pytorch/pytorch/pull/181730,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-06-02T22:47:07Z +pr,169694,Increase precision for golden solution in transformer tests.,https://github.com/pytorch/pytorch/pull/169694,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/rocm-mi300, ciflow/rocm-mi355",wait for contributor update,2026-06-02T20:35:18Z +pr,184533,Defer AOTAutograd version-check error for forward-only graph,https://github.com/pytorch/pytorch/pull/184533,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/inductor,wait for contributor update,2026-06-02T00:24:56Z +pr,184335,[spmd_types] pipeline parallel support for spmd_types-annotated tensors,https://github.com/pytorch/pytorch/pull/184335,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-29T18:14:50Z +pr,185407,[RFC][dynamo] Narrow GB6297 to only graph-break on differentiable duplicate tensor inputs,https://github.com/pytorch/pytorch/pull/185407,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-28T15:19:54Z +pr,184141,Remove pointer_to_optional_list,https://github.com/pytorch/pytorch/pull/184141,waiting on contributor,43,review decision requests changes; CI lane labels: ciflow/inductor,wait for contributor update,2026-05-28T03:00:31Z +pr,185418,[coor] Represent get_rank as a SymInt,https://github.com/pytorch/pytorch/pull/185418,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-27T23:42:33Z +pr,178204,[PyTorch][OSS][AOTAutograd] Allow None values in saved tensors for vc_check split (#178204),https://github.com/pytorch/pytorch/pull/178204,waiting on contributor,43,"review decision requests changes; CI lane labels: ciflow/inductor, ciflow/trunk",wait for contributor update,2026-05-26T23:23:31Z +pr,182576,Improved static Type Hints for dataloader,https://github.com/pytorch/pytorch/pull/182576,waiting on contributor,43,review decision requests changes,wait for contributor update,2026-05-25T15:01:16Z +pr,185011,[inductor] Pair maxpool value and offset lowering,https://github.com/pytorch/pytorch/pull/185011,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-24T06:14:38Z +pr,184930,[inductor] Score broadcast reuse for expensive pointwise tiling,https://github.com/pytorch/pytorch/pull/184930,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-24T03:28:29Z +pr,184999,[inductor] Avoid pointwise cat for complex output branches,https://github.com/pytorch/pytorch/pull/184999,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-24T01:26:01Z +pr,185012,[inductor] Add FP64-heavy pointwise autotune candidate,https://github.com/pytorch/pytorch/pull/185012,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-24T01:25:36Z +pr,184949,Remove allow-untyped-defs from pyi files,https://github.com/pytorch/pytorch/pull/184949,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-05-23T17:18:49Z +pr,184933,[inductor] Fold sparse one-hot sum reductions,https://github.com/pytorch/pytorch/pull/184933,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-23T16:46:11Z +pr,184962,[ROCm] Drop version cap on bf16/fp16 mm/bmm accuracy skip,https://github.com/pytorch/pytorch/pull/184962,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/h100, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-23T09:13:26Z +pr,184937,[inductor] Prototype channel-resident reduction suffix codegen,https://github.com/pytorch/pytorch/pull/184937,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-23T02:58:55Z +pr,184873,Scope third-party install rules out of the wheel via CMake components,https://github.com/pytorch/pytorch/pull/184873,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-22T18:52:43Z +pr,180904,Enable MyST Markdown docstrings in Sphinx docs,https://github.com/pytorch/pytorch/pull/180904,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-20T12:46:39Z +pr,183925,Covert notes/cuda from .rst to .md,https://github.com/pytorch/pytorch/pull/183925,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-20T12:46:29Z +pr,178628,[DRAFT]Add TorchInductor documentation to User Guide,https://github.com/pytorch/pytorch/pull/178628,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-20T12:16:49Z +pr,184464,Cache PyCompilerInterface method lookups during graph capture,https://github.com/pytorch/pytorch/pull/184464,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-05-20T11:05:43Z +pr,184440,[dynamic_spec]Add optimizer to ease user's dynamic_spec adoption,https://github.com/pytorch/pytorch/pull/184440,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-20T03:01:16Z +pr,182921,[DO NOT MERGE][TESTING ONLY] Set MI300 ROCm build arch to SPIR-V,https://github.com/pytorch/pytorch/pull/182921,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/rocm-mi300,"identify whether block is CI, merge conflict, or review gate",2026-05-19T20:45:28Z +pr,177605,Default to fused implementation when available,https://github.com/pytorch/pytorch/pull/177605,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-19T15:43:47Z +pr,178412,Move steps to CPU for fused optims,https://github.com/pytorch/pytorch/pull/178412,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/mps,"identify whether block is CI, merge conflict, or review gate",2026-05-19T15:37:43Z +pr,184308,Route privateuse1-overrideable ops through AOTI dispatcher (,https://github.com/pytorch/pytorch/pull/184308,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-19T06:20:42Z +pr,183798,[ROCm] Fix HIP stream pool aliasing when GPU tracing is active,https://github.com/pytorch/pytorch/pull/183798,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-18T19:50:37Z +pr,183479,TMA scatter_add: add multi-row-per-warp path for small rows,https://github.com/pytorch/pytorch/pull/183479,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-16T21:13:13Z +pr,183966,[CUDA] Change argmax to int64_t for better numerics,https://github.com/pytorch/pytorch/pull/183966,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-16T19:37:12Z +pr,177705,[Do Not Land] Lookup Table Recorder,https://github.com/pytorch/pytorch/pull/177705,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:56Z +pr,179011,[Inductor][NVGEMM] Add fast path for GEMM dispatch,https://github.com/pytorch/pytorch/pull/179011,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:55Z +pr,179012,[Inductor][NVGEMM] Add supplement heuristic configs behind flag,https://github.com/pytorch/pytorch/pull/179012,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:54Z +pr,178840,Add parallel subprocess compilation and disk caching for NVGEMM backend,https://github.com/pytorch/pytorch/pull/178840,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:54Z +pr,179167,[Inductor][NVGEMM] Fix FP4 autotuning tensor creation,https://github.com/pytorch/pytorch/pull/179167,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:51Z +pr,179014,[Inductor][NVGEMM] Wrap GroupedGemmArguments in try/except,https://github.com/pytorch/pytorch/pull/179014,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:51Z +pr,179013,[Inductor][NVGEMM] Fix FP4 autotuning tensor creation,https://github.com/pytorch/pytorch/pull/179013,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:51Z +pr,179015,[Inductor][NVGEMM] Add fast path for GEMM dispatch,https://github.com/pytorch/pytorch/pull/179015,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:50Z +pr,179016,[Inductor][NVGEMM] Add supplement heuristic configs behind flag,https://github.com/pytorch/pytorch/pull/179016,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:49Z +pr,179049,[Inductor][NVGEMM] Add supplement heuristic configs behind flag,https://github.com/pytorch/pytorch/pull/179049,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:48Z +pr,179175,[Inductor][NVGEMM] Fix FP4 autotuning tensor creation,https://github.com/pytorch/pytorch/pull/179175,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-16T01:01:46Z +pr,183797,fix internal tests for [BE][Ez]: Make more dataclass slots=True (#183449),https://github.com/pytorch/pytorch/pull/183797,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,"identify whether block is CI, merge conflict, or review gate",2026-05-15T04:28:17Z +pr,166063,Add Gluon based grouped MM kernel for Blackwell into Inductor,https://github.com/pytorch/pytorch/pull/166063,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300","identify whether block is CI, merge conflict, or review gate",2026-05-15T03:00:43Z +pr,180891,[ROCm] Migrate from rocm_smi to amd_smi,https://github.com/pytorch/pytorch/pull/180891,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/rocm-mi300, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-14T16:21:13Z +pr,178354,[XPU] Add XPU dispatch for _scaled_grouped_mm,https://github.com/pytorch/pytorch/pull/178354,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk, ciflow/xpu","identify whether block is CI, merge conflict, or review gate",2026-05-14T12:16:43Z +pr,183612,"Update NVPL versions: blas 0.3.0 → 0.5.0.1, lapack 0.2.3.1 → 0.3.2",https://github.com/pytorch/pytorch/pull/183612,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/binaries,"identify whether block is CI, merge conflict, or review gate",2026-05-14T05:59:26Z +pr,183374,"[ROCm] [ TEST , PLEASE DON""T MERGE ] TheRock 7.13 nightly deb install path",https://github.com/pytorch/pytorch/pull/183374,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/op-benchmark,"identify whether block is CI, merge conflict, or review gate",2026-05-14T05:58:28Z +pr,177690,"DO NOT MERGE Default pre_grad_pass_timing to ""late"" in fbcode",https://github.com/pytorch/pytorch/pull/177690,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-13T22:28:58Z +pr,181306,Integrate profiler documentation into torch.profiler docs,https://github.com/pytorch/pytorch/pull/181306,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-13T21:16:39Z +pr,183456,[ROCm] [opbench] Temporary profiler workaround for rocprofiler crash (#182719),https://github.com/pytorch/pytorch/pull/183456,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/op-benchmark,"identify whether block is CI, merge conflict, or review gate",2026-05-13T11:54:00Z +pr,183361,Prototype lazy module backward hook compile,https://github.com/pytorch/pytorch/pull/183361,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-13T02:15:55Z +pr,183452,[BE]: Update cutlass submodule to 4.5.0rc,https://github.com/pytorch/pytorch/pull/183452,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-12T21:53:53Z +pr,180874,Add CPU offload support for selective activation checkpointing,https://github.com/pytorch/pytorch/pull/180874,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-05-11T20:06:23Z +pr,183264,[WIP][Dynamo] Add mark_batch_invariant API,https://github.com/pytorch/pytorch/pull/183264,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan","identify whether block is CI, merge conflict, or review gate",2026-05-11T14:19:47Z +pr,181089,[Inductor] Disable split-reduction lowering under batch_invariant,https://github.com/pytorch/pytorch/pull/181089,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk","identify whether block is CI, merge conflict, or review gate",2026-05-11T14:19:44Z +pr,182779,[WIP] Test playground,https://github.com/pytorch/pytorch/pull/182779,draft/noise,43,"PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/dtensor, ciflow/h100-symm-mem, ciflow/inductor, ciflow/inductor-pallas","identify whether block is CI, merge conflict, or review gate",2026-05-11T06:05:27Z +pr,183054,coverage-td plan,https://github.com/pytorch/pytorch/pull/183054,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-05-10T14:28:20Z +pr,130856,[pipelining] lazy shape inference for stage,https://github.com/pytorch/pytorch/pull/130856,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,"identify whether block is CI, merge conflict, or review gate",2026-03-21T22:38:50Z +pr,172109,stage executor,https://github.com/pytorch/pytorch/pull/172109,draft/noise,43,PR is draft; failed checks or blocked/unstable merge state; PR has not updated for at least 45 days,"identify whether block is CI, merge conflict, or review gate",2026-03-20T22:32:07Z +pr,174864,[XPU] Fix accuracy issue in addmm for bf16/f16 dtypes,https://github.com/pytorch/pytorch/pull/174864,draft/noise,41,PR is draft; review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/xpu,wait for contributor update,2026-07-03T07:46:44Z +pr,133419,"Keep is_mkldnn, add is_onednn",https://github.com/pytorch/pytorch/pull/133419,draft/noise,41,"PR is draft; review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-18T12:58:26Z +pr,133392,Update DispatchKey and namespace,https://github.com/pytorch/pytorch/pull/133392,draft/noise,41,"PR is draft; review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-18T12:58:24Z +pr,133296,Update file name with onednn,https://github.com/pytorch/pytorch/pull/133296,draft/noise,41,"PR is draft; review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-18T12:58:20Z +pr,133315,Update non user APIs,https://github.com/pytorch/pytorch/pull/133315,draft/noise,41,"PR is draft; review decision requests changes; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-18T12:58:17Z +issue,189014,[RFC]: Generalize memory snapshot APIs under `torch.accelerator.memory`,https://github.com/pytorch/pytorch/issues/189014,needs design decision,38,needs design/research label present; component/oncall labels: module: accelerator,route to design/research decision before patch planning,2026-07-06T07:55:00Z +issue,114403,refactor TracingContext to take a more limited subset of ViewAndMutationMeta,https://github.com/pytorch/pytorch/issues/114403,good volunteer slice,38,"label says issue is suitable for a scoped external contribution; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2026-07-03T09:11:23Z +issue,102207,skipIfTorchInductor Tracking Issue,https://github.com/pytorch/pytorch/issues/102207,good volunteer slice,38,"label says issue is suitable for a scoped external contribution; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2026-07-01T23:50:03Z +issue,99149,"Spectral Normalization can not be applied to Conv{1,2,3}d",https://github.com/pytorch/pytorch/issues/99149,needs design decision,38,"needs design/research label present; component/oncall labels: module: convolution, module: nn, module: norms and normalization",route to design/research decision before patch planning,2026-07-01T12:27:04Z +issue,106571,Enable more flake8-bugbear lints,https://github.com/pytorch/pytorch/issues/106571,good volunteer slice,38,label says issue is suitable for a scoped external contribution; component/oncall labels: module: lint,scan when time allows,2026-07-01T09:53:28Z +issue,9886,[doc] functionalities not documented,https://github.com/pytorch/pytorch/issues/9886,good volunteer slice,38,label says issue is suitable for a scoped external contribution; component/oncall labels: module: docs,scan when time allows,2026-06-30T08:53:16Z +issue,158968,foreach_map enhancements,https://github.com/pytorch/pytorch/issues/158968,good volunteer slice,38,"label says issue is suitable for a scoped external contribution; component/oncall labels: module: dynamo, module: inductor, oncall: pt2",scan when time allows,2026-06-27T21:11:51Z +issue,188216,Feature request: built-in loss extrapolator,https://github.com/pytorch/pytorch/issues/188216,needs design decision,38,needs design/research label present; component/oncall labels: module: loss,route to design/research decision before patch planning,2026-06-26T00:50:41Z +issue,122192,extra checks for reduce-overhead inconsistent when warmup/recording,https://github.com/pytorch/pytorch/issues/122192,needs design decision,38,"needs design/research label present; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2",route to design/research decision before patch planning,2026-06-23T14:21:46Z +issue,170839,Varlen attention API with custom varlen `attn_mask`,https://github.com/pytorch/pytorch/issues/170839,needs design decision,38,"needs design/research label present; component/oncall labels: module: flex attention, module: nn, module: sdpa, oncall: pt2",route to design/research decision before patch planning,2026-06-23T04:55:20Z +issue,176837,FlexAttention: extend AuxRequest to return min_scores and key indices for row-wise max/min,https://github.com/pytorch/pytorch/issues/176837,needs design decision,38,"needs design/research label present; component/oncall labels: module: flex attention, oncall: pt2",route to design/research decision before patch planning,2026-06-23T04:54:39Z +issue,178586,`make_graphed_callables` silently drops in-place mutations on user-provided input tensors,https://github.com/pytorch/pytorch/issues/178586,needs design decision,38,"needs design/research label present; component/oncall labels: module: correctness (silent), module: cuda, module: cuda graphs, oncall: pt2",route to design/research decision before patch planning,2026-06-18T16:13:13Z +issue,160230,Offer official Pytorch Vulkan backend on pytorch.org,https://github.com/pytorch/pytorch/issues/160230,needs design decision,38,needs design/research label present; component/oncall labels: module: vulkan,route to design/research decision before patch planning,2026-06-16T02:56:13Z +issue,50688,[feature request] `torch.scan` (also port `lax.fori_loop` / `lax.while_loop` / `lax.associative_scan` and hopefully parallelized associative scans),https://github.com/pytorch/pytorch/issues/50688,needs design decision,38,"needs design/research label present; component/oncall labels: module: functional ux, module: functorch, module: numpy, module: pt2-dispatcher",route to design/research decision before patch planning,2026-06-09T18:43:18Z +issue,136312,Fault tolerance with k8s,https://github.com/pytorch/pytorch/issues/136312,needs design decision,38,"needs design/research label present; component/oncall labels: oncall: distributed, oncall: distributed infra",route to design/research decision before patch planning,2026-06-05T17:02:45Z +issue,152323,compile generates inefficient code when mutating small slice of a graph input,https://github.com/pytorch/pytorch/issues/152323,needs design decision,38,"needs design/research label present; component/oncall labels: module: functionalization, module: inductor, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-06-04T20:21:56Z +issue,183434,Audit license-files glob in pyproject.toml for over-collection,https://github.com/pytorch/pytorch/issues/183434,needs design decision,38,"needs design/research label present; component/oncall labels: module: binaries, module: third_party",route to design/research decision before patch planning,2026-06-01T13:07:22Z +issue,184517,LRScheduler doesn't use _get_closed_form_lr when initalizing scheduler,https://github.com/pytorch/pytorch/issues/184517,needs design decision,38,needs design/research label present; component/oncall labels: module: lrscheduler,route to design/research decision before patch planning,2026-05-28T23:10:56Z +issue,158657,extend torch.util.checkpoint to support offload to cpu/nvme,https://github.com/pytorch/pytorch/issues/158657,needs design decision,38,"needs design/research label present; component/oncall labels: module: activation checkpointing, module: autograd",route to design/research decision before patch planning,2026-05-24T00:42:42Z +issue,141452,[Dynamo] Keep weak reference of parameters instead of accessing them by attributes,https://github.com/pytorch/pytorch/issues/141452,needs design decision,38,"needs design/research label present; component/oncall labels: module: dynamo, module: vllm, oncall: pt2",route to design/research decision before patch planning,2026-05-21T12:33:44Z +issue,180845,[Dynamo] Unsupported Tensor.item() call in list comprehension when capture_scalar_outputs=False,https://github.com/pytorch/pytorch/issues/180845,needs design decision,38,"needs design/research label present; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",route to design/research decision before patch planning,2026-05-19T15:48:14Z +issue,180107,Allow getting `AtenTensorHandle` from the tensor object via AOTI stable C API,https://github.com/pytorch/pytorch/issues/180107,needs design decision,38,"needs design/research label present; component/oncall labels: module: aotinductor, module: cpp, oncall: pt2",route to design/research decision before patch planning,2026-05-19T00:55:49Z +issue,151142,AOT Dispatcher converts a single `detach` call to multiple `aten.detach`,https://github.com/pytorch/pytorch/issues/151142,needs design decision,38,"needs design/research label present; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-05-15T03:07:02Z +issue,146804,"`torch.library.register_fake` respects only positional order, but not kwargs order",https://github.com/pytorch/pytorch/issues/146804,needs design decision,38,"needs design/research label present; component/oncall labels: module: library, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-05-15T03:06:38Z +issue,149258,Auto-selective activation checkpointing is not optimal for speed (issue with min_cut_rematerialization_partition),https://github.com/pytorch/pytorch/issues/149258,needs design decision,38,"needs design/research label present; component/oncall labels: module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-05-15T02:04:06Z +issue,140835,auto-grad graph replicate split_with_sizes(lengths) X times where X = len(lengths) effecting compile time,https://github.com/pytorch/pytorch/issues/140835,needs design decision,38,"needs design/research label present; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-05-14T22:22:08Z +issue,132651,fix unnecessary re-trace in AOTDispatcher in subclass code path,https://github.com/pytorch/pytorch/issues/132651,needs design decision,38,"needs design/research label present; component/oncall labels: module: __torch_dispatch__, module: aotdispatch, module: compile-time, module: pt2-dispatcher",route to design/research decision before patch planning,2026-05-14T22:21:58Z +issue,113022,Operator with only `Tensor[][]` args unsupported by dispatcher,https://github.com/pytorch/pytorch/issues/113022,needs design decision,38,"needs design/research label present; component/oncall labels: module: custom-operators, module: dispatch, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-05-14T22:20:21Z +issue,149559,Improve handling for custom ops with no returns,https://github.com/pytorch/pytorch/issues/149559,needs design decision,38,"needs design/research label present; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-05-14T22:20:10Z +issue,139444,registering op through function does not work,https://github.com/pytorch/pytorch/issues/139444,needs design decision,38,"needs design/research label present; component/oncall labels: module: custom-operators, module: docs, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-05-14T22:19:57Z +issue,158827,[RFC] Graph generalization,https://github.com/pytorch/pytorch/issues/158827,needs design decision,38,"needs design/research label present; component/oncall labels: module: accelerator, module: privateuse1",route to design/research decision before patch planning,2026-05-12T02:21:44Z +issue,148077,Feature Request: CUDA-Optimized Queue Buffer for PyTorch,https://github.com/pytorch/pytorch/issues/148077,needs design decision,38,needs design/research label present; component/oncall labels: module: cuda,route to design/research decision before patch planning,2026-05-06T20:41:02Z +issue,164469,[RFC] Adopt Spin as developer cli,https://github.com/pytorch/pytorch/issues/164469,needs design decision,38,needs design/research label present; component/oncall labels: module: build,route to design/research decision before patch planning,2026-04-30T09:55:49Z +issue,146414,MX basic dtypes in pytorch/pytorch,https://github.com/pytorch/pytorch/issues/146414,needs design decision,38,needs design/research label present; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2026-04-27T01:44:32Z +issue,180215,[RFC] Best practices for backend installations requiring +cpu / non-PyPI wheels,https://github.com/pytorch/pytorch/issues/180215,needs design decision,38,"needs design/research label present; component/oncall labels: module: binaries, module: privateuse1",route to design/research decision before patch planning,2026-04-13T17:33:41Z +issue,120376,Stochastic rounding in bfloat16,https://github.com/pytorch/pytorch/issues/120376,needs design decision,38,needs design/research label present; component/oncall labels: module: bfloat16,route to design/research decision before patch planning,2026-04-06T06:50:45Z +issue,77171,Allow users to express fused matmul/bias/relu,https://github.com/pytorch/pytorch/issues/77171,needs design decision,38,needs design/research label present; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2026-03-28T19:02:17Z +issue,41625,[feature request] New function `torch.slice(...)` mirroring TorchScript op signature or add step argument to `torch.narrow`,https://github.com/pytorch/pytorch/issues/41625,needs design decision,38,needs design/research label present,route to design/research decision before patch planning,2026-03-28T08:51:48Z +issue,173382,Make torch.cuda.empty_cache() actually empty the GPU cache,https://github.com/pytorch/pytorch/issues/173382,needs design decision,38,"needs design/research label present; component/oncall labels: module: cuda, module: cudacachingallocator",route to design/research decision before patch planning,2026-03-25T16:54:02Z +issue,158346,Compatibility with C++20 `std::format`,https://github.com/pytorch/pytorch/issues/158346,needs design decision,38,needs design/research label present; component/oncall labels: module: build,route to design/research decision before patch planning,2026-03-24T18:05:16Z +issue,177623,add support for Predictive Coding as alternative to Backpropagation,https://github.com/pytorch/pytorch/issues/177623,needs design decision,38,"needs design/research label present; component/oncall labels: module: autograd, module: optimizer",route to design/research decision before patch planning,2026-03-17T07:06:03Z +issue,177537,[RFC] nn.BlockAttentionResidual: depth-wise softmax attention over layer outputs as a drop-in residual replacement,https://github.com/pytorch/pytorch/issues/177537,needs design decision,38,needs design/research label present; component/oncall labels: module: nn,route to design/research decision before patch planning,2026-03-16T19:22:08Z +issue,47117,ctx.save_for_backward doesn't save torch.Tensor subclasses fully,https://github.com/pytorch/pytorch/issues/47117,needs design decision,38,"needs design/research label present; component/oncall labels: module: __torch_function__, module: autograd, module: docs",route to design/research decision before patch planning,2026-03-08T04:23:36Z +issue,175862,Adding Laplacian smoothing to MultiHeadAttention,https://github.com/pytorch/pytorch/issues/175862,needs design decision,38,"needs design/research label present; component/oncall labels: module: nn, module: sdpa",route to design/research decision before patch planning,2026-02-27T20:46:50Z +issue,110080,Devices API,https://github.com/pytorch/pytorch/issues/110080,needs design decision,38,needs design/research label present; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2026-02-25T17:33:38Z +issue,169803,[RFC] Add RGE256ctr: A Fast Counter-Mode ARX RNG for PyTorch GeneratorImpl,https://github.com/pytorch/pytorch/issues/169803,needs design decision,38,needs design/research label present; component/oncall labels: module: random,route to design/research decision before patch planning,2026-02-23T21:29:46Z +issue,62530,[feature request] `torch.switch` (to mirror `lax.switch`) and `torch.cond` - with jax arithmetic coder as a case study,https://github.com/pytorch/pytorch/issues/62530,needs design decision,38,needs design/research label present; component/oncall labels: module: nn,route to design/research decision before patch planning,2026-02-23T20:40:30Z +issue,113737,torch.compile + SAC: mutations in backward are not preserved,https://github.com/pytorch/pytorch/issues/113737,needs design decision,38,"needs design/research label present; component/oncall labels: module: activation checkpointing, module: aotdispatch, module: pt2-dispatcher, oncall: distributed",route to design/research decision before patch planning,2026-02-23T04:21:30Z +issue,115516,[FakeTensor] FakeTensorMode dispatch won't change generator state,https://github.com/pytorch/pytorch/issues/115516,needs design decision,38,"needs design/research label present; component/oncall labels: module: faketensor, module: pt2-dispatcher, module: random, oncall: pt2",route to design/research decision before patch planning,2026-02-22T18:20:05Z +issue,124296,Request for modifying native allocator to use nvidia unified memory `cudaMallocManage`,https://github.com/pytorch/pytorch/issues/124296,needs design decision,38,"needs design/research label present; component/oncall labels: module: cuda, module: cudacachingallocator",route to design/research decision before patch planning,2026-02-22T08:08:13Z +issue,170224,Way to ask Dynamo to hoist some warmup logic to the beginning of the compile region,https://github.com/pytorch/pytorch/issues/170224,needs design decision,38,"needs design/research label present; component/oncall labels: module: dynamo, oncall: pt2",route to design/research decision before patch planning,2026-02-21T01:04:53Z +issue,101082,Multiple Learning Rate Scheduler for Specific Parameters Groups,https://github.com/pytorch/pytorch/issues/101082,needs design decision,38,"needs design/research label present; component/oncall labels: module: lrscheduler, module: optimizer",route to design/research decision before patch planning,2026-02-21T00:30:01Z +issue,110485,[export] `torch.tensor(0)` should not get burned in as a constant,https://github.com/pytorch/pytorch/issues/110485,needs design decision,38,"needs design/research label present; component/oncall labels: module: aotdispatch, module: dynamo, module: pt2-dispatcher, oncall: export",route to design/research decision before patch planning,2026-02-20T16:47:30Z +issue,110175,Module-level bufferization for torch dynamo module spanning multiple `fx.GraphModule`,https://github.com/pytorch/pytorch/issues/110175,needs design decision,38,"needs design/research label present; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2026-02-20T16:40:39Z +issue,105053,Multiple dimensions support for `torch.max`,https://github.com/pytorch/pytorch/issues/105053,needs design decision,38,"needs design/research label present; component/oncall labels: module: numpy, module: python frontend",route to design/research decision before patch planning,2026-01-30T07:32:41Z +issue,173559,Secondary Optimizer for Muon,https://github.com/pytorch/pytorch/issues/173559,needs design decision,38,needs design/research label present; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2026-01-28T17:12:21Z +issue,117844,torch.fft.rfft: Unsupported dtype BFloat16,https://github.com/pytorch/pytorch/issues/117844,needs design decision,38,"needs design/research label present; component/oncall labels: module: bfloat16, module: cuda, module: fft",route to design/research decision before patch planning,2026-01-13T10:43:46Z +issue,171676,"PyTorch allows `tensor.storage().resize_(0)`, leading to segmentation fault on subsequent tensor operations",https://github.com/pytorch/pytorch/issues/171676,needs design decision,38,"needs design/research label present; component/oncall labels: module: edge cases, module: error checking, module: molly-guard",route to design/research decision before patch planning,2026-01-12T16:40:45Z +issue,143412,[Inductor test failure] torch:inductor/test_select_algorithm TestSelectAlgorithm.test_convolution1 with cuda 12.6.3,https://github.com/pytorch/pytorch/issues/143412,has linked PR,38,"open PR references this issue; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-26T08:23:37Z +issue,113800,InternalTorchDynamoError rewrapping loses exception chaining,https://github.com/pytorch/pytorch/issues/113800,has linked PR,38,"open PR references this issue; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, module: logging, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-12T21:32:10Z +issue,115484,torch.compile() breaks when using DeepSpeed ZeRO Level 3 sharding,https://github.com/pytorch/pytorch/issues/115484,has linked PR,38,"open PR references this issue; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-27T00:17:11Z +issue,97078,[compile] TypeError: __init__() missing 1 required positional argument: 'parent_module',https://github.com/pytorch/pytorch/issues/97078,has linked PR,38,open PR references this issue; needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-13T18:41:08Z +pr,189310,Generalize GeneratorState input device type,https://github.com/pytorch/pytorch/pull/189310,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-08T22:01:33Z +pr,189288,[inductor] Fix compute_ancestors KeyError from self-referential mutation copy,https://github.com/pytorch/pytorch/pull/189288,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-08T21:22:14Z +pr,189119,Fix test-jsons artifact upload path,https://github.com/pytorch/pytorch/pull/189119,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T21:13:59Z +pr,189010,[Dynamo] Support relax frontend in tvm backend,https://github.com/pytorch/pytorch/pull/189010,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-08T21:12:14Z +pr,189158,Skip GHA stats artifacts from other workflow attempts,https://github.com/pytorch/pytorch/pull/189158,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T21:09:22Z +pr,189154,Bundle unversioned libnuma for ROCm wheels,https://github.com/pytorch/pytorch/pull/189154,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T21:08:02Z +pr,189125,[Dynamo] Raise clear error for dynamic shapes in tvm relay path,https://github.com/pytorch/pytorch/pull/189125,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-08T20:50:41Z +pr,189127,[inductor] Decompose broadcast-bias baddbmm before pointwise,https://github.com/pytorch/pytorch/pull/189127,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-08T20:38:12Z +pr,189124,[inductor] Cache DSR-expanded single-config reductions,https://github.com/pytorch/pytorch/pull/189124,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-08T20:36:12Z +pr,189285,Fix MIOpen RNN dropout: gradient stall and missing per-step mask variation,https://github.com/pytorch/pytorch/pull/189285,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T20:34:34Z +pr,185794,Fix torch.kron RuntimeError on non-contiguous tensors - noah,https://github.com/pytorch/pytorch/pull/185794,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T20:31:48Z +pr,189145,[Dynamo] Preserve Stream/Event subclass types in graph reconstruction,https://github.com/pytorch/pytorch/pull/189145,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-08T20:27:40Z +pr,189174,Fix dcp privateuse1 stream support,https://github.com/pytorch/pytorch/pull/189174,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan,review when ready,2026-07-08T20:19:18Z +pr,189234,"Fix 1+erf catastrophic cancellation in gelu, Normal.cdf, and ndtr",https://github.com/pytorch/pytorch/pull/189234,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps, ciflow/torchtitan",review when ready,2026-07-08T20:12:59Z +pr,188795,Fix `torch.asarray` to preserve an input's device when a default device is set,https://github.com/pytorch/pytorch/pull/188795,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T19:33:54Z +pr,182945,Add pattern matcher to deduplicate var and std calculation,https://github.com/pytorch/pytorch/pull/182945,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",review when ready,2026-07-08T17:58:53Z +pr,189045,Fix NaN gradients in nn.MultiheadAttention for fully-masked rows (#41…,https://github.com/pytorch/pytorch/pull/189045,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T17:31:38Z +pr,184094,Clone aliased mm/addmm inputs on CPU to avoid in-place gemm,https://github.com/pytorch/pytorch/pull/184094,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T16:56:12Z +pr,181720,[MPS] Make pin_memory return CPU-aliased storage backed by a unified MTLBuffer,https://github.com/pytorch/pytorch/pull/181720,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-08T15:34:47Z +pr,181308,[CPU] Fix std_mean/var_mean all-reduce ~4x slower than std + mean (#122191),https://github.com/pytorch/pytorch/pull/181308,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T15:27:39Z +pr,188053,test: add svdvals NaN propagation test (fixes #187759),https://github.com/pytorch/pytorch/pull/188053,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T11:37:03Z +pr,185751,"[nn] Raise ValueError early for invalid (ndim, pad_size) in non-constant F.pad modes",https://github.com/pytorch/pytorch/pull/185751,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/trunk,review when ready,2026-07-08T05:20:35Z +pr,185756,[clamp] Fix float16 scalar overflow check inconsistency between CPU and GPU,https://github.com/pytorch/pytorch/pull/185756,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/trunk,review when ready,2026-07-08T05:20:15Z +pr,187104,Fix embedding_bag CPU segfault when offsets is empty and indices is not,https://github.com/pytorch/pytorch/pull/187104,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-08T03:34:58Z +pr,184830,[Dynamo] Guard source-backed identity comparisons,https://github.com/pytorch/pytorch/pull/184830,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-07T23:45:32Z +pr,181027,Factor nanmean dtype checks into a helper + add kwarg coverage,https://github.com/pytorch/pytorch/pull/181027,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T22:17:01Z +pr,181062,Validate grad_output shape in multi_margin_loss_backward,https://github.com/pytorch/pytorch/pull/181062,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T22:15:59Z +pr,188039,autograd: fix PyObject* refcount leak in setup_context path,https://github.com/pytorch/pytorch/pull/188039,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T22:06:39Z +pr,148674,Set /NODEFAULTLIB:vcomp for MSVC when linking caffe2::mkl with libiomp5md.lib,https://github.com/pytorch/pytorch/pull/148674,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T21:45:46Z +pr,184780,[MPS] Fast atomic-free path for flat torch.unique (fixes #97310 perf + #111173 correctness),https://github.com/pytorch/pytorch/pull/184780,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-07T21:14:05Z +pr,185054,[autograd] Validate forward-mode jacobian/hessian require vectorize=T…,https://github.com/pytorch/pytorch/pull/185054,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T21:10:35Z +pr,186039,Fix Dynamo verbose guard checks for frame locals,https://github.com/pytorch/pytorch/pull/186039,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-07T13:27:02Z +pr,182273,[c10d] Fix device_id not propagated through ProcessGroupWrapper,https://github.com/pytorch/pytorch/pull/182273,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T09:47:10Z +pr,188906,Add torch.utils.checkpoint.register_determinism_check,https://github.com/pytorch/pytorch/pull/188906,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan,review when ready,2026-07-07T09:13:44Z +pr,184809,Improve custom op Tensor subclass fake error,https://github.com/pytorch/pytorch/pull/184809,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-07T08:21:23Z +pr,178844,[c10d] Support stacked output tensors in Gloo all_gather_into_tensor …,https://github.com/pytorch/pytorch/pull/178844,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T05:58:57Z +pr,185782,fix: [PT2] [Hardening] Track recompiles alongside graph breaks in our actu...,https://github.com/pytorch/pytorch/pull/185782,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-07T05:44:08Z +pr,186375,[CUDA] Change nvmlDeviceGetGpuFabricInfoV assertion to warning,https://github.com/pytorch/pytorch/pull/186375,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-07T04:56:49Z +pr,180408,Fix torch.eye CPU performance cliff at d=182 (#48251),https://github.com/pytorch/pytorch/pull/180408,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T20:31:27Z +pr,188945,Fix assert_close to support comparisons involving Python and NumPy scalars,https://github.com/pytorch/pytorch/pull/188945,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T18:59:36Z +pr,188923,test: add gradient health validation tests (fixes #177116),https://github.com/pytorch/pytorch/pull/188923,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T18:59:05Z +pr,188969,[docs] document __torch_function__ behavior during autograd hooks,https://github.com/pytorch/pytorch/pull/188969,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T18:57:22Z +pr,188883,Handle empty autograd sections in backend YAML,https://github.com/pytorch/pytorch/pull/188883,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T18:51:38Z +pr,188831,Fix module re-export in tensor/experimental API,https://github.com/pytorch/pytorch/pull/188831,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-06T18:50:44Z +pr,188893,[ONNX] Clarify that output_names labels outputs but does not reorder them,https://github.com/pytorch/pytorch/pull/188893,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T18:47:17Z +pr,188857,Fix type confusion in Stream comparison against non-Stream objects,https://github.com/pytorch/pytorch/pull/188857,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T18:46:58Z +pr,188843,Fix scatter/scatter_add/scatter_reduce src gradient when index is sma…,https://github.com/pytorch/pytorch/pull/188843,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T14:33:01Z +pr,188968,Fix record_stack_traces returning empty trace when no forward frame found,https://github.com/pytorch/pytorch/pull/188968,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-06T14:32:32Z +pr,188637,[inductor] Reindex transpose-contiguous clone to enable vertical fusion,https://github.com/pytorch/pytorch/pull/188637,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-06T07:13:02Z +pr,188642,Fix __qualname__ for OpOverloadPacket,https://github.com/pytorch/pytorch/pull/188642,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-05T23:21:14Z +pr,182541,add regression test for weight_norm meta initialization (#96409),https://github.com/pytorch/pytorch/pull/182541,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-05T20:49:12Z +pr,187325,[Metal][memory] add configurable memory budget API for MPS allocator (#105839),https://github.com/pytorch/pytorch/pull/187325,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-05T15:46:24Z +pr,188066,test: add F.normalize zero-input correctness tests (fixes #184575),https://github.com/pytorch/pytorch/pull/188066,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-05T15:20:50Z +pr,187453,[MPS] Add error handling for complex dtypes in max unpooling operations,https://github.com/pytorch/pytorch/pull/187453,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-05T12:44:29Z +pr,182296,fix(nn): add input dim validation to adaptive_avg_pool2d/3d,https://github.com/pytorch/pytorch/pull/182296,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T22:22:31Z +pr,182291,fix(nn): InstanceNorm always raises ValueError on channel mismatch,https://github.com/pytorch/pytorch/pull/182291,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T22:21:46Z +pr,188579,optimize split reduction heuristics on GB200,https://github.com/pytorch/pytorch/pull/188579,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-04T20:26:06Z +pr,188791,Fix MKL VML first-call dispatch race producing silent low-precision results,https://github.com/pytorch/pytorch/pull/188791,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T20:17:38Z +pr,182268,autograd: add `list_saved_tensors` utility to traverse saved tensors in autograd graph,https://github.com/pytorch/pytorch/pull/182268,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T18:50:12Z +pr,182262,Use TypeAliasType for type aliases and @exposed_in for function re-exports,https://github.com/pytorch/pytorch/pull/182262,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T18:50:11Z +pr,182204,torch.cat/concat/concatenate: support dim=None (flatten before concat),https://github.com/pytorch/pytorch/pull/182204,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T18:50:08Z +pr,182133,Allow torch.arange() to return empty tensor when bounds are inconsistent with step sign,https://github.com/pytorch/pytorch/pull/182133,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T18:50:05Z +pr,182076,[Fix] Add overflow and validation checks for pixel_shuffle in decomposition and meta,https://github.com/pytorch/pytorch/pull/182076,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T15:13:40Z +pr,181756,[Docs] Replace outdated flake8 and git-pre-commit references with lintrunner,https://github.com/pytorch/pytorch/pull/181756,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T15:13:39Z +pr,176265,Add backward batching rules for SDPA and fix attn_bias under vmap,https://github.com/pytorch/pytorch/pull/176265,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-04T10:38:32Z +pr,188719,Fix Inductor KeyError on view_as_complex of non-contiguous CUDA tensors,https://github.com/pytorch/pytorch/pull/188719,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-03T07:04:20Z +pr,188633,Draft: Update CUTLASS fence proxy API usage,https://github.com/pytorch/pytorch/pull/188633,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-03T02:40:12Z +pr,188528,cuda: reject negative caching allocator sizes,https://github.com/pytorch/pytorch/pull/188528,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-03T01:38:20Z +pr,166388,Fixing assertion fail by checking if no_sync is true before DDPsink,https://github.com/pytorch/pytorch/pull/166388,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-02T23:22:38Z +pr,188762,Fix multinomial without replacement silently sampling zero-probability categories,https://github.com/pytorch/pytorch/pull/188762,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-02T23:01:26Z +pr,188200,[MPS] Vectorized contiguous cat fast path for any dimension,https://github.com/pytorch/pytorch/pull/188200,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-02T20:29:24Z +pr,186524,Validate conv transpose output padding,https://github.com/pytorch/pytorch/pull/186524,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review when ready,2026-07-02T16:36:56Z +pr,188560,[MPS] Fix constant address space for scalar params in binary_dense_scalar kernels,https://github.com/pytorch/pytorch/pull/188560,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-02T12:56:21Z +pr,187327,[Metal][cache] dual-hash verification for MPSGraphCache and MPSKernelCache (#77176),https://github.com/pytorch/pytorch/pull/187327,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-02T12:49:24Z +pr,187330,[Metal][memory] clear graph/kernel compilation caches in empty_cache() (#154329),https://github.com/pytorch/pytorch/pull/187330,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-02T12:23:41Z +pr,188349,[HOP] Adding length parameter to scan,https://github.com/pytorch/pytorch/pull/188349,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-02T07:59:50Z +pr,172219,[XPU] Adapt test_basic_chrome_trace to XPU,https://github.com/pytorch/pytorch/pull/172219,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-02T05:15:03Z +pr,188647,torch/fx: fix _format_target to handle Python keywords and embedded quotes,https://github.com/pytorch/pytorch/pull/188647,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-02T05:07:39Z +pr,182627,Handle bool inputs in Scalar::log,https://github.com/pytorch/pytorch/pull/182627,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-02T03:43:44Z +pr,188531,pooling: guard max_pool1d integer parameters,https://github.com/pytorch/pytorch/pull/188531,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-02T01:32:30Z +pr,188732,[MPS] Enable ConvTranspose3d for fp16/bf16 via internal fp32 upcast,https://github.com/pytorch/pytorch/pull/188732,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-07-02T00:38:47Z +pr,187544,Fix torch.gcd returning negative results for INT_MIN inputs,https://github.com/pytorch/pytorch/pull/187544,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-01T18:10:34Z +pr,188260,[nn] Fix RReLU noise tensor not being set in eval mode,https://github.com/pytorch/pytorch/pull/188260,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-07-01T13:56:24Z +pr,187257,[DCP] Raise RuntimeError on load dtype mismatch to prevent precision …,https://github.com/pytorch/pytorch/pull/187257,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan,review when ready,2026-07-01T13:30:53Z +pr,179273,[optim] support of dynamic coefficients and normalization in Muon,https://github.com/pytorch/pytorch/pull/179273,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-01T11:36:49Z +pr,188543,Remove deprecated torch.utils.mobile_optimizer from API index,https://github.com/pytorch/pytorch/pull/188543,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-01T03:45:16Z +pr,187309,Add maximize parameter to LBFGS optimizer,https://github.com/pytorch/pytorch/pull/187309,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-07-01T01:54:07Z +pr,188526,error checking: validate _pdist_forward dimensions,https://github.com/pytorch/pytorch/pull/188526,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-30T22:49:04Z +pr,188524,test: centralize optimizer comparison tolerances,https://github.com/pytorch/pytorch/pull/188524,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-30T18:06:14Z +pr,188525,test: add Linear module error inputs,https://github.com/pytorch/pytorch/pull/188525,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-30T18:06:02Z +pr,176434,[export] Move schema deserialization to C++ for faster export.load(),https://github.com/pytorch/pytorch/pull/176434,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review when ready,2026-06-30T16:15:57Z +pr,188369,Add DTensor sharding strategy for FFT c2c,https://github.com/pytorch/pytorch/pull/188369,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",review when ready,2026-06-30T04:26:57Z +pr,188399,Fix #171905: Use typing_extensions.TypeAliasType for better reexport of `__module__,https://github.com/pytorch/pytorch/pull/188399,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-29T21:15:17Z +pr,188373,Enforce 1D tensor validation for batch_norm parameters,https://github.com/pytorch/pytorch/pull/188373,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-29T20:46:06Z +pr,188363,Add kernel_size > 0 validation to FractionalMaxPool2d matching Fracti…,https://github.com/pytorch/pytorch/pull/188363,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-29T16:53:32Z +pr,188381,aten: add output_size kwarg to masked_select for meta/eager shape oracle support,https://github.com/pytorch/pytorch/pull/188381,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-29T16:50:49Z +pr,188347,Preserve tied parameter and buffer aliasing in load_state_dict(assign=True),https://github.com/pytorch/pytorch/pull/188347,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-29T16:28:37Z +pr,188334,Handle negative integer exponents in SymPy utilities,https://github.com/pytorch/pytorch/pull/188334,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-29T16:28:31Z +pr,188271,Allow PrivateUse1 backends to opt out of CEA decomposition,https://github.com/pytorch/pytorch/pull/188271,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-27T22:05:19Z +pr,181681,Fix UBSan null-pointer UB in vec256 partial-load Vec::loadu (#181510),https://github.com/pytorch/pytorch/pull/181681,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-27T19:56:28Z +pr,188183,[nn] Avoid mutating state_dict metadata in load_state_dict(assign=True),https://github.com/pytorch/pytorch/pull/188183,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-27T17:47:49Z +pr,179377,Fix gaussian_nll_loss rejecting valid multi-dim broadcastable var shapes,https://github.com/pytorch/pytorch/pull/179377,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-27T10:17:30Z +pr,185056,Raise error when class attributes shadow registered parameters or modules,https://github.com/pytorch/pytorch/pull/185056,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-27T08:37:47Z +pr,187795,Guard release_available_cached_blocks during CUDA graph capture,https://github.com/pytorch/pytorch/pull/187795,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-27T08:06:18Z +pr,183743,Add drift-based tolerance fallback to op tests,https://github.com/pytorch/pytorch/pull/183743,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-26T21:51:47Z +pr,188099,[JIT] Use native fill/zero ops for outplace tracing of fill_/zero_,https://github.com/pytorch/pytorch/pull/188099,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T20:26:35Z +pr,188258,"Fix _foreach_{add,mul,div}.Tensor autograd",https://github.com/pytorch/pytorch/pull/188258,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T19:29:57Z +pr,188149,Handle mixed Tensor/Scalar min/max bounds in torch.clamp,https://github.com/pytorch/pytorch/pull/188149,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T18:24:30Z +pr,188138,fix: Docs are little bit outdated for torch logs,https://github.com/pytorch/pytorch/pull/188138,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T17:22:45Z +pr,188270,Fix inplace modification error with softmax when combining forward and backward AD,https://github.com/pytorch/pytorch/pull/188270,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:54:53Z +pr,182601,docs: clarify tile dims argument,https://github.com/pytorch/pytorch/pull/182601,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:38:32Z +pr,182602,docs: cross-reference sigmoid APIs,https://github.com/pytorch/pytorch/pull/182602,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:38:24Z +pr,182603,docs: clarify RNN hidden state dimensions,https://github.com/pytorch/pytorch/pull/182603,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:38:15Z +pr,182604,docs: clarify multithreaded autograd guidance,https://github.com/pytorch/pytorch/pull/182604,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:38:03Z +pr,185398,docs: clean up multilabel soft margin loss args,https://github.com/pytorch/pytorch/pull/185398,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:37:51Z +pr,185744,docs: clarify LSTM equation notation,https://github.com/pytorch/pytorch/pull/185744,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:37:39Z +pr,185745,docs: clarify scalar fill values,https://github.com/pytorch/pytorch/pull/185745,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T15:37:31Z +pr,182605,test: strengthen optimizer step hook ordering checks,https://github.com/pytorch/pytorch/pull/182605,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T14:59:02Z +pr,183316,Fix/signbit cuda half nan,https://github.com/pytorch/pytorch/pull/183316,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-26T09:13:26Z +pr,187246,serialization: fix nondeterministic pickle bytes in _legacy_save (gh#39383),https://github.com/pytorch/pytorch/pull/187246,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-25T18:00:04Z +pr,187630,fx/passes: fix non-deterministic ordering in get_source_partitions(),https://github.com/pytorch/pytorch/pull/187630,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-24T19:05:24Z +pr,187803,fix(autograd): sinc_backward NaN for second derivative at x=0,https://github.com/pytorch/pytorch/pull/187803,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-24T18:38:42Z +pr,181051,Fix heap-buffer-overflow in quantized_max_pool2d for 1d args,https://github.com/pytorch/pytorch/pull/181051,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-24T02:31:02Z +pr,175525,vmap compatibility with non-tensor leaves,https://github.com/pytorch/pytorch/pull/175525,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T18:28:04Z +pr,186929,[fx] fix _patch_function using wrong function when forward is wrapped,https://github.com/pytorch/pytorch/pull/186929,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T14:29:39Z +pr,187764,[MPS] Add matrix_exp support,https://github.com/pytorch/pytorch/pull/187764,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-23T14:24:24Z +pr,187814,Reject factor=0 in ConstantLR to avoid later ZeroDivisionError,https://github.com/pytorch/pytorch/pull/187814,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T14:22:29Z +pr,187757,Narrow ModuleDict.__getattr__ return type to Module for type checkers,https://github.com/pytorch/pytorch/pull/187757,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T14:21:26Z +pr,187790,[ao][pruning] Add experimental topology-aware modules,https://github.com/pytorch/pytorch/pull/187790,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T14:00:31Z +pr,187783,fix(nn): raise ValueError instead of ZeroDivisionError for norm_type=0 in lp_pool1d/2d/3d and LPPool modules,https://github.com/pytorch/pytorch/pull/187783,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T13:59:01Z +pr,187748,[MPS] Build only the reduced Q in linalg.qr (fix tall-skinny perf cliff),https://github.com/pytorch/pytorch/pull/187748,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-23T13:58:56Z +pr,187153,jit: add bounds checking for textIndex in SourceRangeDeserializer,https://github.com/pytorch/pytorch/pull/187153,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T13:57:55Z +pr,187570,Fixed heaviside nan propagation issue,https://github.com/pytorch/pytorch/pull/187570,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T13:20:10Z +pr,187597,Add return type annotation to nn.Module.named_modules,https://github.com/pytorch/pytorch/pull/187597,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-23T03:43:47Z +pr,177985,Fix SDPA mem-efficient attention crash when num_heads >= 65536,https://github.com/pytorch/pytorch/pull/177985,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-22T19:18:16Z +pr,187771,Warn when convert_sync_batchnorm drops _BatchNorm subclass behavior,https://github.com/pytorch/pytorch/pull/187771,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-22T13:36:13Z +pr,187198,Add target range validation in binary_cross_entropy_with_logits,https://github.com/pytorch/pytorch/pull/187198,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/dtensor, ciflow/trunk",review when ready,2026-06-22T13:01:20Z +pr,180184,Export auto-generated unstructured inplace meta dispatch symbols,https://github.com/pytorch/pytorch/pull/180184,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-22T11:23:16Z +pr,184641,Fix GroupNorm validation for non-positive num-groups,https://github.com/pytorch/pytorch/pull/184641,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-22T07:34:41Z +pr,178099,Add TypeGuard[Tensor] return type to is_tensor_like,https://github.com/pytorch/pytorch/pull/178099,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-21T23:32:58Z +pr,179827,Replace BF16 matmul OpInfo checks with a feature query,https://github.com/pytorch/pytorch/pull/179827,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-21T21:58:37Z +pr,180781,Fix integer divide-by-zero UB in CUDA binary kernels,https://github.com/pytorch/pytorch/pull/180781,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-21T20:07:02Z +pr,186571,Fix #176298: support unsigned add and sub on CPU,https://github.com/pytorch/pytorch/pull/186571,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-20T22:51:02Z +pr,180601,"Fix pydocstyle errors in _guards.py, _ops.py, _jit_internal.py, functional.py, _tensor_str.py, library.py",https://github.com/pytorch/pytorch/pull/180601,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-20T20:02:57Z +pr,185319,Sort record names before hashing in writeSerializationId,https://github.com/pytorch/pytorch/pull/185319,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-20T09:15:33Z +pr,187595,Fix torch.gcd/lcm integer overflow on INT_MIN inputs,https://github.com/pytorch/pytorch/pull/187595,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-19T23:33:16Z +pr,180188,Validate opt_dtype in nanmean to reject non-floating-point types,https://github.com/pytorch/pytorch/pull/180188,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-19T17:18:04Z +pr,177455,Add nn.SwiGLU activation function,https://github.com/pytorch/pytorch/pull/177455,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-19T16:40:56Z +pr,180764,cpp_extension: add CUDA dlink flags to JIT load APIs,https://github.com/pytorch/pytorch/pull/180764,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-19T13:44:47Z +pr,187656,Reject complex scale/zero_point tensors in quantize_per_tensor with an explicit dtype error instead of value cannot be converted to type double without overflow.,https://github.com/pytorch/pytorch/pull/187656,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-19T06:01:06Z +pr,187662,Reject aliased Q/R out tensors in torch.linalg.qr,https://github.com/pytorch/pytorch/pull/187662,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-18T16:58:06Z +pr,187286,nn: add torch.nn.functional.rotary_embedding and torch.nn.RotaryEmbedding,https://github.com/pytorch/pytorch/pull/187286,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-17T14:54:27Z +pr,172539,[torchgen] Enable Structured Kernels and Out-as-Primary Wrappers for PrivateUse1,https://github.com/pytorch/pytorch/pull/172539,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-17T07:00:03Z +pr,186050,Output Shape Check added to transposed Conv,https://github.com/pytorch/pytorch/pull/186050,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-17T05:12:18Z +pr,185807,Stable sorting for consistent gradient for torch.quantile on tied values,https://github.com/pytorch/pytorch/pull/185807,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-17T05:11:36Z +pr,187253,Align dynamic qparams test helper with C++ overflow check,https://github.com/pytorch/pytorch/pull/187253,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-16T21:10:10Z +pr,187238,Add docs pages for torch.e and torch.pi,https://github.com/pytorch/pytorch/pull/187238,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-16T14:49:12Z +pr,186589,Fix serialization errors in SubTest calls,https://github.com/pytorch/pytorch/pull/186589,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-16T14:25:09Z +pr,187329,Raise ValueError for non-positive kernel_size/dilation/stride in unfold/fold,https://github.com/pytorch/pytorch/pull/187329,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-16T14:24:24Z +pr,187243,Reject p < 1 in triplet_margin_loss to prevent NaN from p-norm overflow,https://github.com/pytorch/pytorch/pull/187243,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-16T14:24:23Z +pr,187242,Validate LocalResponseNorm parameters to prevent silent inf output,https://github.com/pytorch/pytorch/pull/187242,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-16T14:24:23Z +pr,186930,Fix NaN propagation behavior in torch.sign for CPU and CUDA backends,https://github.com/pytorch/pytorch/pull/186930,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-16T10:13:22Z +pr,173894,[nn] Fix softplus numerical instability with large beta values,https://github.com/pytorch/pytorch/pull/173894,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-15T18:54:17Z +pr,179941,[inductor] Add reduction loop peeling for Triton codegen,https://github.com/pytorch/pytorch/pull/179941,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-06-15T16:51:54Z +pr,176746,Added 'gelsd' to CUDA torch.linalg.lstsq driver so that underdetermined least square systems can be solved on CUDA,https://github.com/pytorch/pytorch/pull/176746,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-15T12:42:30Z +pr,177866,Sparse semi structured setup/teardown latency,https://github.com/pytorch/pytorch/pull/177866,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-15T12:41:56Z +pr,178797,Fix meta conv shape checks for invalid groups vs out_channels,https://github.com/pytorch/pytorch/pull/178797,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-15T12:33:52Z +pr,179160,Fix uint32 overflow in CUDA scan _with_indices kernels,https://github.com/pytorch/pytorch/pull/179160,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-15T10:35:26Z +pr,179379,[DTensor] Convert type aliases to TypeAliasType in experimental/,https://github.com/pytorch/pytorch/pull/179379,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-14T16:59:28Z +pr,187101,Add missing dispatch for bf16 in tensor_cpu,https://github.com/pytorch/pytorch/pull/187101,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-14T16:22:04Z +pr,187187,[typing] Replace TypeAlias with TypeAliasType in torch/types.py,https://github.com/pytorch/pytorch/pull/187187,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-13T07:15:22Z +pr,187090,Fix nn.Module.to() failure on subclass parameters held by Python weakrefs,https://github.com/pytorch/pytorch/pull/187090,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-12T17:59:41Z +pr,178371,[Profiler] Fix profile_memory not working with profile_all_threads,https://github.com/pytorch/pytorch/pull/178371,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-12T03:48:16Z +pr,186940,[ShardedTensor] use device-agnostic fallback in tensor_device (fixes #186938),https://github.com/pytorch/pytorch/pull/186940,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/torchtitan,review when ready,2026-06-11T20:01:15Z +pr,175509,Add fast path for min/max dim reductions with size 2 on CPU,https://github.com/pytorch/pytorch/pull/175509,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-11T19:24:58Z +pr,173239,Test openreg inference hooks,https://github.com/pytorch/pytorch/pull/173239,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-11T19:24:28Z +pr,174196,Add validation for lstm.data hidden state shapes to prevent segfaults,https://github.com/pytorch/pytorch/pull/174196,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-10T23:56:04Z +pr,186863,[MPS] Register unsigned binary Metal kernels,https://github.com/pytorch/pytorch/pull/186863,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-10T19:20:14Z +pr,174559,[lint] Enable Ruff PYI056 by replacing __all__.append/extend usage,https://github.com/pytorch/pytorch/pull/174559,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-06-09T15:34:39Z +pr,186177,[autograd] Defer device thread creation when multithreading is disabled,https://github.com/pytorch/pytorch/pull/186177,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-09T15:31:17Z +pr,186607,Raise a clear error for malformed allocator config instead of INTERNAL_ASSERT,https://github.com/pytorch/pytorch/pull/186607,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-09T03:28:06Z +pr,186207,Fix segfault and invalid cast in UntypedStorage,https://github.com/pytorch/pytorch/pull/186207,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-08T14:19:24Z +pr,172829,Fixes registration of custom sharding strategies when using register_sharding with TensorList,https://github.com/pytorch/pytorch/pull/172829,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-06T15:09:53Z +pr,185187,Validate forward-mode autograd.functional vectorize usage,https://github.com/pytorch/pytorch/pull/185187,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-05T21:45:21Z +pr,185864,Fix segfault in native_batch_norm with empty running_mean/running_var,https://github.com/pytorch/pytorch/pull/185864,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-05T18:37:05Z +pr,186073,Support weights_only=True in torch.export.load,https://github.com/pytorch/pytorch/pull/186073,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/inductor,review when ready,2026-06-05T05:57:30Z +pr,186128,Reject dtype conversion of quantized TypedStorage with a clear error,https://github.com/pytorch/pytorch/pull/186128,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-04T22:18:57Z +pr,186159,Restore Any return type for Module.__getattr__ to fix typing false positives,https://github.com/pytorch/pytorch/pull/186159,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-04T13:43:08Z +pr,185959,Fix SIGFPE in torch.lcm with the integer type minimum,https://github.com/pytorch/pytorch/pull/185959,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-03T22:23:51Z +pr,185868,"Update clamp, clamp_min, clamp_max to use the minimum norm subgradient",https://github.com/pytorch/pytorch/pull/185868,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-03T22:01:26Z +pr,184577,nn: validate num_groups > 0 in GroupNorm.__init__,https://github.com/pytorch/pytorch/pull/184577,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/dtensor,review when ready,2026-06-03T10:48:25Z +pr,185117,nn: add missing memory_format overload to Module.to(),https://github.com/pytorch/pytorch/pull/185117,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-03T10:48:12Z +pr,184898,autograd: raise ValueError for forward-mode without vectorize in jacobian/hessian,https://github.com/pytorch/pytorch/pull/184898,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-03T10:48:10Z +pr,185747,Use at::mps::OptionalMPSGuard for MPS dispatch keys in codegen,https://github.com/pytorch/pytorch/pull/185747,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-03T06:03:40Z +pr,175542,[DTensor] Register DTensorSpec as pytree constant at module level,https://github.com/pytorch/pytorch/pull/175542,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-03T03:46:21Z +pr,185792,test/distributed/pipelining: add end-to-end tests with HuggingFace GPT-2 and BERT,https://github.com/pytorch/pytorch/pull/185792,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-01T22:37:41Z +pr,180444,Add `validate` parameter to `torch.multinomial` to skip input validation,https://github.com/pytorch/pytorch/pull/180444,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-06-01T01:55:05Z +pr,180496,feat: Add OpenTelemetry trace export support for torch.profiler,https://github.com/pytorch/pytorch/pull/180496,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-06-01T01:53:04Z +pr,185454,"fix clamp and clamp_{min,max} gradient at scalar bound to be 0",https://github.com/pytorch/pytorch/pull/185454,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-29T15:35:41Z +pr,184698,Materialize lazy modules fully on load_state_dict instead of leaving them in a partially initialized state.,https://github.com/pytorch/pytorch/pull/184698,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-29T09:45:11Z +pr,183269,[DTensor] Switch test_embedding_ops to DTensorOpTestBase (MultiThreadedTestCase),https://github.com/pytorch/pytorch/pull/183269,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/dtensor,review when ready,2026-05-29T05:58:52Z +pr,148023,torch.utils.checkpoint preserves torch function mode stack during recompute,https://github.com/pytorch/pytorch/pull/148023,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-28T12:47:26Z +pr,184064,Fix silent int64 precision loss in F.pad constant mode,https://github.com/pytorch/pytorch/pull/184064,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-27T20:32:14Z +pr,178031,[MPS] Add adaptive_avg_pool3d support for MPS,https://github.com/pytorch/pytorch/pull/178031,has linked issue,35,"PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/mps",review when ready,2026-05-27T05:44:35Z +pr,182384,[autograd] Refresh ctx.needs_input_grad for partial backward,https://github.com/pytorch/pytorch/pull/182384,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-26T19:44:04Z +pr,185026,[typing] Add Iterable base class to ModuleList,https://github.com/pytorch/pytorch/pull/185026,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-26T13:51:35Z +pr,184885,docs: correct torch.addcdiv prose to match implementation order,https://github.com/pytorch/pytorch/pull/184885,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-26T13:44:56Z +pr,184857,"Raise ValueError early when strategy=""forward-mode"" used without vectorize=True",https://github.com/pytorch/pytorch/pull/184857,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-26T13:44:24Z +pr,185193,Validate forward-mode functional autograd arguments,https://github.com/pytorch/pytorch/pull/185193,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-26T13:36:34Z +pr,184840,Add Module.to overload for memory_format,https://github.com/pytorch/pytorch/pull/184840,has linked issue,35,PR closes or references open issue numbers,review when ready,2026-05-26T13:35:02Z +pr,181993,[MPS] fix device binary shaders,https://github.com/pytorch/pytorch/pull/181993,has linked issue,35,PR closes or references open issue numbers; CI lane labels: ciflow/mps,review when ready,2026-05-26T10:06:18Z +pr,177961,[ROCm] Enable native AsyncTP,https://github.com/pytorch/pytorch/pull/177961,waiting on contributor,35,"review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/h100-symm-mem, ciflow/periodic-rocm-mi300, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-05-21T14:38:28Z +pr,181352,Re-enable nightly tests on s390x,https://github.com/pytorch/pytorch/pull/181352,waiting on contributor,35,"review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/s390",wait for contributor update,2026-05-20T06:57:57Z +pr,169107,Add failing test for PolynomialLR total_iters=0 closed-form crash,https://github.com/pytorch/pytorch/pull/169107,waiting on contributor,35,review decision requests changes; PR has not updated for at least 45 days,wait for contributor update,2026-05-19T19:11:12Z +pr,183749,Support auto_functionalized_v2 in can_fuse_into_auto_functionalized,https://github.com/pytorch/pytorch/pull/183749,waiting on contributor,35,"review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-19T14:46:12Z +pr,181520,dataloader: add performance warning for in-memory datasets with num_workers>0,https://github.com/pytorch/pytorch/pull/181520,waiting on contributor,35,review decision requests changes; PR has not updated for at least 45 days,wait for contributor update,2026-05-16T01:07:15Z +pr,181501,docs: add DataLoader performance guidance for in-memory datasets,https://github.com/pytorch/pytorch/pull/181501,waiting on contributor,35,review decision requests changes; PR has not updated for at least 45 days,wait for contributor update,2026-05-16T01:07:10Z +pr,183433,[inductor] Fix backslash escaping order in user-defined Triton kernel…,https://github.com/pytorch/pytorch/pull/183433,waiting on contributor,35,"review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-14T16:04:45Z +pr,181110,Workaround for nvrtcCompileProgram changing locale,https://github.com/pytorch/pytorch/pull/181110,waiting on contributor,35,review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/binaries,wait for contributor update,2026-05-12T15:38:03Z +pr,182612,Ignore decoding errors when reading CMake cache files (errors='ignore'),https://github.com/pytorch/pytorch/pull/182612,waiting on contributor,35,review decision requests changes; PR has not updated for at least 45 days,wait for contributor update,2026-05-12T02:00:14Z +pr,178390,[inductor] fold half addmm+relu into _addmm_activation,https://github.com/pytorch/pytorch/pull/178390,waiting on contributor,35,"review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-10T20:49:42Z +pr,134942,Enabling ATen Distribution kernels for AARCH64 using OpenRNG,https://github.com/pytorch/pytorch/pull/134942,waiting on contributor,35,review decision requests changes; PR has not updated for at least 45 days,wait for contributor update,2026-04-20T18:28:18Z +pr,138388,Extending SVE VEC Backend Support in PyTorch to SVE128 and SVE512.,https://github.com/pytorch/pytorch/pull/138388,waiting on contributor,35,"review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/linux-aarch64",wait for contributor update,2025-08-11T18:47:31Z +issue,177952,Fix onnx tests running on GPU/CUDA,https://github.com/pytorch/pytorch/issues/177952,needs reproduction,32,needs reproduction label present; component/oncall labels: module: onnx,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-08T14:55:43Z +issue,130253,"RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling `cublasGemmEx( handle, opa, opb, m, n, k, &falpha, a, CUDA_R_16F, lda, b, CUDA_R_16F, ldb, &fbeta, c, CUDA_R_16F, ldc, CUDA_R_32F, CUBLAS_GEMM_DEFAULT_TENSOR_OP)`",https://github.com/pytorch/pytorch/issues/130253,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cublas, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-05T07:13:33Z +issue,188411,RTX 5090 cudaErrorLaunchTimeout,https://github.com/pytorch/pytorch/issues/188411,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: cudnn, module: reductions",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-07-02T08:35:28Z +issue,187824,libcudnn_engines_tensor_ir.so.9 not included in the PyTorch based on CUDA 13.0,https://github.com/pytorch/pytorch/issues/187824,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: binaries, module: cudnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-30T18:59:19Z +issue,184030,torch.nn.PoissonNLLLoss returns NaN for negative input when log_input=False,https://github.com/pytorch/pytorch/issues/184030,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: edge cases, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-22T13:04:06Z +issue,170138,Performance regression of depthwise Conv1d with fp16/fp32 AMP from 2.7.1 -> 2.8.0 and the issue remains in 2.9.1,https://github.com/pytorch/pytorch/issues/170138,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: convolution, module: cuda, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-10T08:20:07Z +issue,164342,Official support for sm_120 (RTX 50-series / Blackwell) in stable PyTorch builds,https://github.com/pytorch/pytorch/issues/164342,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: windows",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T18:39:38Z +issue,88251,Weird random SIGTERM occurance,https://github.com/pytorch/pytorch/issues/88251,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: ddp, module: elastic, oncall: distributed, oncall: distributed parallelisms",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:10:51Z +issue,98542,Training runs 50% slower when using 2 GPUs comparing to 1,https://github.com/pytorch/pytorch/issues/98542,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:09:47Z +issue,114861,Torchrun threads with different computing speed.,https://github.com/pytorch/pytorch/issues/114861,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:06:33Z +issue,120383,Numeric stability issue with full parameter fine tuning of Llama 2 7B model with FSDP,https://github.com/pytorch/pytorch/issues/120383,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:05:47Z +issue,121606,"RuntimeError: Rank 2 successfully reached monitoredBarrier, but received errors while waiting for send/recv from rank 0. Please check rank 0 logs for faulty rank.",https://github.com/pytorch/pytorch/issues/121606,needs reproduction,32,"needs reproduction label present; component/oncall labels: oncall: distributed, oncall: distributed infra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:05:06Z +issue,121944,raise RendezvousConnectionError( torch.distributed.elastic.rendezvous.api.RendezvousConnectionError: The connection to the C10d store has failed. See inner exception for details.,https://github.com/pytorch/pytorch/issues/121944,needs reproduction,32,"needs reproduction label present; component/oncall labels: oncall: distributed, oncall: distributed infra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:05:01Z +issue,139579,There is a bug when training SwinIR model in pychram and running it,https://github.com/pytorch/pytorch/issues/139579,needs reproduction,32,"needs reproduction label present; component/oncall labels: oncall: distributed, oncall: distributed infra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T17:02:04Z +issue,158724,Problem with RPC on dual stack networks,https://github.com/pytorch/pytorch/issues/158724,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: rpc, oncall: distributed, oncall: distributed infra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T16:57:10Z +issue,165959,"torch.distributed.DistStoreError: use_libuv was requested but PyTorch was built without libuv support, run with USE_LIBUV=0 to disable it.",https://github.com/pytorch/pytorch/issues/165959,needs reproduction,32,"needs reproduction label present; component/oncall labels: oncall: distributed, oncall: distributed infra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-06-05T16:54:11Z +issue,171499,bug in torch.stack a long 5d video and yields a misordered video output,https://github.com/pytorch/pytorch/issues/171499,needs reproduction,32,needs reproduction label present; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-19T15:37:45Z +issue,138422,vmap with torch.autograd.grad does not work on output of compiled function,https://github.com/pytorch/pytorch/issues/138422,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: functorch, module: pt2-dispatcher, module: vmap, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-15T22:49:16Z +issue,148811,Custom Triton operator not registered when loading AOT-compiled .so in C++ environment,https://github.com/pytorch/pytorch/issues/148811,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: aotdispatch, module: custom-operators, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-14T22:00:31Z +issue,147115,Unable to print in a branch run by torch.cond,https://github.com/pytorch/pytorch/issues/147115,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-14T19:06:16Z +issue,181901,[Inductor][NVGEMM] tile_M<128 EFC kernels broadcast incorrectly,https://github.com/pytorch/pytorch/issues/181901,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: correctness (silent), module: dependency bug, module: inductor, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-14T17:14:36Z +issue,141545,Missing Fake/Meta/Abstract Kernels for Operators,https://github.com/pytorch/pytorch/issues/141545,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-14T15:46:36Z +issue,146994,subclasses + HOPs fail with `Attempting to use FunctionalTensor on its own`,https://github.com/pytorch/pytorch/issues/146994,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-14T03:49:00Z +issue,145451,[Dynamo]while_loop raise an exception,https://github.com/pytorch/pytorch/issues/145451,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-14T03:46:44Z +issue,143579,`assert_size_stride` failing in `_inductor/utils.py` `return model(new_inputs)`,https://github.com/pytorch/pytorch/issues/143579,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-13T23:41:36Z +issue,145782,Behavior Difference with Flex Attention + Sequence Packing,https://github.com/pytorch/pytorch/issues/145782,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-13T23:37:38Z +issue,144803,Something is fishy with discard_graph_changes,https://github.com/pytorch/pytorch/issues/144803,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-13T04:03:54Z +issue,182760,"[Build] UnicodeDecodeError in cmake_utils.py due to non-standard library version suffixes (e.g., OpenSSL 1.1.1wa)",https://github.com/pytorch/pytorch/issues/182760,needs reproduction,32,needs reproduction label present; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-11T20:21:35Z +issue,173389,Inference latency increase on ARM after upgrading from PyTorch 1.13.1 to 2.x,https://github.com/pytorch/pytorch/issues/173389,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: arm, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-05T12:40:22Z +issue,159207,Add official support for CUDA sm_120 (RTX 5090 / Blackwell architecture),https://github.com/pytorch/pytorch/issues/159207,needs reproduction,32,needs reproduction label present; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-05-01T11:40:06Z +issue,136344,Composite Compliance Tensor ut check fails in privateuse1 device,https://github.com/pytorch/pytorch/issues/136344,needs reproduction,32,needs reproduction label present; component/oncall labels: module: privateuse1,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-28T04:15:35Z +issue,177020,tensor cuda memory is not released when using nested_tensor_from_jagged,https://github.com/pytorch/pytorch/issues/177020,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: memory usage, module: python frontend, module: python version",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-22T14:42:29Z +issue,162957,torch.linalg.eigh uses a large amount of memory in pytorch 2.8.0,https://github.com/pytorch/pytorch/issues/162957,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: linear algebra, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-17T07:28:54Z +issue,177819,"MPS unavailable on macOS 26.3.1 arm64, and `device=""mps""` reports incorrect macOS requirement",https://github.com/pytorch/pytorch/issues/177819,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: collect_env.py, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-16T06:26:56Z +issue,175057,torch.compile / Inductor generates invalid C++ (undeclared zuf0/zuf1) during compile (CppCompileError) on RTX 3090,https://github.com/pytorch/pytorch/issues/175057,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-16T05:31:36Z +issue,179904,PyTorch 2.6 weights_only=True breaks loading of existing checkpoints,https://github.com/pytorch/pytorch/issues/179904,needs reproduction,32,needs reproduction label present; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-04-13T16:40:16Z +issue,177028,tensor cuda memory cannot be release by gc when a tensor hold ref of another,https://github.com/pytorch/pytorch/issues/177028,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-31T08:01:55Z +issue,178145,"Large fp32 discrepancy in LeNet5Fashion conv2 output on CUDA eager baseline, with similar error against both CUDA compile and CPU eager",https://github.com/pytorch/pytorch/issues/178145,needs reproduction,32,needs reproduction label present,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-25T03:36:08Z +issue,178251,RuntimeError: Function 'ScaledDotProductEfficientAttentionBackward0' returned nan values in its 0th output when using attn_mask in SDPA backward,https://github.com/pytorch/pytorch/issues/178251,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: autograd, module: cuda, module: sdpa",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-24T19:23:26Z +issue,178049,print is repeated sometimes when the calc is done on gpu,https://github.com/pytorch/pytorch/issues/178049,needs reproduction,32,needs reproduction label present,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-24T19:15:51Z +issue,177738,C++ Can't capture Transformer CUDA Graph,https://github.com/pytorch/pytorch/issues/177738,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cpp, module: cuda graphs, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-24T19:10:21Z +issue,126654,torch.nn.functional.scaled_dot_product_attention returns NaN values after backward pass.,https://github.com/pytorch/pytorch/issues/126654,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: numerical-stability, module: sdpa",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-24T07:12:40Z +issue,133548,Floating point exception on H20,https://github.com/pytorch/pytorch/issues/133548,needs reproduction,32,needs reproduction label present; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-19T08:51:28Z +issue,135196,ModuleNotFoundError: No module named 'torch.fx.experimental.shape_inference',https://github.com/pytorch/pytorch/issues/135196,needs reproduction,32,needs reproduction label present; component/oncall labels: module: fx,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-18T23:59:48Z +issue,97635,JAX + PyTorch produces `OMP: Error #13: Assertion failure at kmp_affinity.cpp(532)`,https://github.com/pytorch/pytorch/issues/97635,needs reproduction,32,needs reproduction label present,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-11T14:54:54Z +issue,171672,torch.compile max-autotune and reduce-overhead much slower than default due to repeated CUDA graph instantiation,https://github.com/pytorch/pytorch/issues/171672,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda graphs, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-10T22:10:29Z +issue,165532,PyTorch CPU / DataLoader: “Too many open files” (`EMFILE`),https://github.com/pytorch/pytorch/issues/165532,needs reproduction,32,needs reproduction label present; component/oncall labels: module: dataloader,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-09T18:36:57Z +issue,166633,"Command '['ninja', '-v']' returned non-zero exit status 255.",https://github.com/pytorch/pytorch/issues/166633,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cpp-extensions, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-07T13:16:23Z +issue,175637,Conv2d output inconsistency across different execution modes (device/dtype/optimization),https://github.com/pytorch/pytorch/issues/175637,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: convolution, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-07T01:41:30Z +issue,118092,torch.load with weights_only=True show error for pickle protocol 4,https://github.com/pytorch/pytorch/issues/118092,needs reproduction,32,needs reproduction label present; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-05T06:49:24Z +issue,173315,[cuda] Numerical Inconsistency: NaN in Eager CUDA vs Valid Output in Compiled CPU for nn.LSTM,https://github.com/pytorch/pytorch/issues/173315,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: edge cases, module: nans and infs, module: rnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-03T16:15:33Z +issue,173476,[Hang] Unrecoverable CUDA deadlock in torch.ops.aten.lstm - process becomes unresponsive,https://github.com/pytorch/pytorch/issues/173476,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: rnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-03T06:51:02Z +issue,125954,Error: command buffer exited with error status.,https://github.com/pytorch/pytorch/issues/125954,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: intel, module: macos, module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-03-02T08:37:19Z +issue,173922,[Numerical Accuracy] 100% relative error in LSTM output between CUDA and CPU backends in eval mode,https://github.com/pytorch/pytorch/issues/173922,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cpu, module: cuda, module: rnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-28T06:40:08Z +issue,174010,Numerical divergence in GPT2 wpe layer: 2.19e-4 relative error between float32 and float16 on CUDA,https://github.com/pytorch/pytorch/issues/174010,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: embedding, module: half, module: numerical-stability",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-28T06:38:47Z +issue,153172,triggered internal assert in matmul,https://github.com/pytorch/pytorch/issues/153172,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: cudacachingallocator",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-23T06:54:04Z +issue,167414,C10_CUDA_KERNEL_LAUNCH_CHECK uses getLastError without clearing the error.,https://github.com/pytorch/pytorch/issues/167414,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: error checking, module: rocm",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-12T17:27:28Z +issue,108041,"Undefined Symobl: pybind11::detail::type_caster::load(pybind11::handle, bool)",https://github.com/pytorch/pytorch/issues/108041,needs reproduction,32,needs reproduction label present; component/oncall labels: module: pybind,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-10T15:20:27Z +issue,170676,"CUBLAS_STATUS_INVALID_VALUE at F.linear((14,4096), (4096,4096)) device='cuda:0', dtype=torch.float16 on 4090 and B300",https://github.com/pytorch/pytorch/issues/170676,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: cuda, module: linear algebra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-05T20:15:09Z +issue,172217,test_numba_integration consistently fails with SIGSEGV on CUDA-enabled environments (Ubuntu 22.04 & RHEL 9),https://github.com/pytorch/pytorch/issues/172217,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: ci, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-02-02T03:15:46Z +issue,137001,BCE loss mps device,https://github.com/pytorch/pytorch/issues/137001,needs reproduction,32,"needs reproduction label present; component/oncall labels: module: correctness (silent), module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-30T23:34:08Z +issue,173066,Training accuracy may suddenly fall to near 0 in the early training stage using pytorch 2.9 and AMP (bfloat16),https://github.com/pytorch/pytorch/issues/173066,needs reproduction,32,needs reproduction label present; component/oncall labels: module: optimizer,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-22T23:56:18Z +issue,172944,Tried to access the schema for which doesn't have a schema registered yet,https://github.com/pytorch/pytorch/issues/172944,needs reproduction,32,needs reproduction label present; component/oncall labels: module: dispatch,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-22T04:02:54Z +issue,172547,Cold start models with compile-cache have significant overhead,https://github.com/pytorch/pytorch/issues/172547,needs reproduction,32,needs reproduction label present; component/oncall labels: oncall: pt2,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-18T09:34:56Z +issue,172628,pip download torch report nvidia_nccl_cu11 checksum mismatch,https://github.com/pytorch/pytorch/issues/172628,needs reproduction,32,needs reproduction label present; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-17T01:39:47Z +issue,168897,Can't install on python 3.13,https://github.com/pytorch/pytorch/issues/168897,needs reproduction,32,needs reproduction label present; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-15T18:19:34Z +issue,111182,Training iresnet with torch.compile is slower than eager mode for torch 2.1.0,https://github.com/pytorch/pytorch/issues/111182,needs reproduction,32,"needs reproduction label present; needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: performance, oncall: pt2",route to design/research decision before patch planning,2024-11-09T02:17:39Z +pr,181926,Enable XPU for sparse_sampled_addmm,https://github.com/pytorch/pytorch/pull/181926,draft/noise,31,PR is draft; review decision requests changes,wait for contributor update,2026-07-08T20:47:00Z +pr,187165,[ROCm] Enable FMA inductor lowering on ROCm,https://github.com/pytorch/pytorch/pull/187165,draft/noise,31,"PR is draft; review decision requests changes; CI lane labels: ciflow/inductor, ciflow/inductor-perf-test-nightly-rocm-mi350, ciflow/rocm-mi200, ciflow/rocm-mi300",wait for contributor update,2026-07-07T17:14:35Z +pr,176726,Fix Muon momentum/update numerics for XPU state_dict parity,https://github.com/pytorch/pytorch/pull/176726,draft/noise,31,PR is draft; review decision requests changes,wait for contributor update,2026-07-06T15:46:35Z +pr,184372,[Draft] Remove ACL,https://github.com/pytorch/pytorch/pull/184372,draft/noise,31,"PR is draft; review decision requests changes; CI lane labels: ciflow/docker, ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-07-03T11:46:42Z +pr,188242,[ROCm] Support USE_ASAN=1 build,https://github.com/pytorch/pytorch/pull/188242,draft/noise,31,PR is draft; review decision requests changes; CI lane labels: ciflow/rocm-mi300,wait for contributor update,2026-07-01T20:52:52Z +pr,182109,[MPS] Replace MPSGraph LayerNorm backward with native Metal kernels,https://github.com/pytorch/pytorch/pull/182109,draft/noise,31,PR is draft; review decision requests changes; CI lane labels: ciflow/mps,wait for contributor update,2026-06-21T22:49:50Z +pr,133753,"Keep mkldnn_rnn_layer, add onednn_rnn_layer",https://github.com/pytorch/pytorch/pull/133753,draft/noise,31,"PR is draft; review decision requests changes; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/torchtitan, ciflow/trunk",wait for contributor update,2026-06-18T15:22:00Z +pr,133667,Update non user APIs - 2,https://github.com/pytorch/pytorch/pull/133667,draft/noise,31,"PR is draft; review decision requests changes; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/mps, ciflow/torchtitan",wait for contributor update,2026-06-18T15:21:56Z +pr,133423,"Keep torch.backends.mkldnn, add torch.backends.onednn",https://github.com/pytorch/pytorch/pull/133423,draft/noise,31,"PR is draft; review decision requests changes; CI lane labels: ciflow/inductor, ciflow/linux-aarch64, ciflow/mps, ciflow/torchtitan",wait for contributor update,2026-06-18T15:21:52Z +issue,180948,DISABLED test_emulate_precision_casts_preserves_explicit_precision_cast_float16 (__main__.CudaReproTests),https://github.com/pytorch/pytorch/issues/180948,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: tests, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-09T01:40:51Z +issue,180194,DISABLED test_vjp_grad_cpu (__main__.TestComposabilityCPU),https://github.com/pytorch/pytorch/issues/180194,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-09T01:16:58Z +issue,174507,PyTorch Stable ABI support,https://github.com/pytorch/pytorch/issues/174507,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: abi,"decide whether this is bug, use case, docs, or support",2026-07-09T00:24:27Z +issue,189164,[CI][B200] B200 Runner Maintenance Notification,https://github.com/pytorch/pytorch/issues/189164,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: infra","decide whether this is bug, use case, docs, or support",2026-07-09T00:23:26Z +issue,188898,DISABLED test_vmap_over_vmap_two_inputs (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/188898,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T19:02:54Z +issue,189279,[CPU] operator_benchmark: eager transcendental unary (cosh/sinh/atan/log2/log10) ~3-5x slower since ~Nov 2025,https://github.com/pytorch/pytorch/issues/189279,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-07-08T17:30:21Z +issue,189262,DISABLED rocm-nightly / linux-noble-rocm-nightly-py3.12-gfx942 / test,https://github.com/pytorch/pytorch/issues/189262,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: rocm","decide whether this is bug, use case, docs, or support",2026-07-08T17:29:46Z +issue,167822,[RFD] Composable Kernel (CK) moving to C++20 — request for feedback/concerns,https://github.com/pytorch/pytorch/issues/167822,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-07-08T17:25:42Z +issue,188756,"[MPS] index_select / F.embedding silently return all-zero rows when the source tensor exceeds 2**31 elements (fixed on main, still broken in release/2.13)",https://github.com/pytorch/pytorch/issues/188756,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: 64-bit, module: correctness (silent), module: embedding, module: mps","decide whether this is bug, use case, docs, or support",2026-07-08T17:18:22Z +issue,189270,[inductor] Emit per-kernel physical op structure (with rematerialization) in provenance,https://github.com/pytorch/pytorch/issues/189270,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: trace, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T15:16:17Z +issue,188496,torch.fft.irfft2 ONNX export produces graph that onnxruntime cannot execute.,https://github.com/pytorch/pytorch/issues/188496,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fft, module: onnx","decide whether this is bug, use case, docs, or support",2026-07-08T14:48:17Z +issue,189261,DISABLED periodic-rocm-mi300 / linux-noble-rocm-py3.12-mi300 / test,https://github.com/pytorch/pytorch/issues/189261,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: rocm","decide whether this is bug, use case, docs, or support",2026-07-08T14:28:01Z +issue,189255,DISABLED test_delayed_optim_step_offload_true_no_shard_cuda (__main__.TestParityWithDDPCUDA),https://github.com/pytorch/pytorch/issues/189255,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-07-08T13:12:54Z +issue,179799,DISABLED test_grad_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/179799,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T13:12:35Z +issue,160740,MPS torch.argmax/torch.argmin fail for non-contiguous input,https://github.com/pytorch/pytorch/issues/160740,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: mps","decide whether this is bug, use case, docs, or support",2026-07-08T11:16:06Z +issue,143954,"No ONNX function found for ",https://github.com/pytorch/pytorch/issues/143954,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: onnx,"decide whether this is bug, use case, docs, or support",2026-07-08T10:40:54Z +issue,187455,[MPS] Migrate remaining MPSGraph-backed ops to native Metal to bound graph-cache memory,https://github.com/pytorch/pytorch/issues/187455,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: memory usage, module: mps, module: performance","decide whether this is bug, use case, docs, or support",2026-07-08T09:10:44Z +issue,168995,MXFP8 + torch.compile error: NotImplementedError: make_reindexer NYI on DtypeView,https://github.com/pytorch/pytorch/issues/168995,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T08:04:22Z +issue,189244,DISABLED test_cat_pointwise (__main__.NumBytesMetricTests),https://github.com/pytorch/pytorch/issues/189244,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T07:30:24Z +issue,189223,DISABLED test_vmap_free_tensor (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/189223,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T07:30:08Z +issue,187289,DISABLED test_col2im_uint32_overflow (__main__.TestMPS),https://github.com/pytorch/pytorch/issues/187289,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: macos, module: mps","decide whether this is bug, use case, docs, or support",2026-07-08T07:30:03Z +issue,152632,DISABLED test_torchvision_models_efficientnet_v2_l (__main__.TestVisionTracing),https://github.com/pytorch/pytorch/issues/152632,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx","decide whether this is bug, use case, docs, or support",2026-07-08T07:30:01Z +issue,179879,DISABLED test_vmap_vmap_cpu (__main__.TestComposabilityCPU),https://github.com/pytorch/pytorch/issues/179879,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T07:29:55Z +issue,189236,[Feature Request] Add 'symmetric' padding_mode (repeats boundary sample),https://github.com/pytorch/pytorch/issues/189236,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: padding","decide whether this is bug, use case, docs, or support",2026-07-08T06:22:05Z +issue,187175,[feature request] Autotune TMA and non-TMA path for pointwise and reduction kernels,https://github.com/pytorch/pytorch/issues/187175,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: inductor, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T05:08:55Z +issue,163650,[DO NOT CLOSE] Autorevert actions shadow mode stream,https://github.com/pytorch/pytorch/issues/163650,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-07-08T01:30:49Z +issue,189221,DISABLED test_nested_dict_partial_load (__main__.TestCheckpointer),https://github.com/pytorch/pytorch/issues/189221,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-07-08T01:09:24Z +issue,179742,DISABLED test_grad_freevar_python_scalar (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/179742,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T01:09:11Z +issue,180213,DISABLED test_vmap_vjp_cpu (__main__.TestComposabilityCPU),https://github.com/pytorch/pytorch/issues/180213,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-08T01:09:09Z +issue,76185,Ensure custom Function are correct in double backward setting,https://github.com/pytorch/pytorch/issues/76185,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: autograd,"decide whether this is bug, use case, docs, or support",2026-07-07T23:37:07Z +issue,187073,torch.compile recompiles when input alternates between AsyncCollectiveTensor and Tensor,https://github.com/pytorch/pytorch/issues/187073,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: fsdp, module: guards, module: performance","decide whether this is bug, use case, docs, or support",2026-07-07T22:55:54Z +issue,189189,Subtraction between two `torch.complex64` tensors fails on modules compiled with `aoti_compile_and_package`,https://github.com/pytorch/pytorch/issues/189189,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, module: complex, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T22:04:29Z +issue,50282,Add MultiScheduler,https://github.com/pytorch/pytorch/issues/50282,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: optimizer,"decide whether this is bug, use case, docs, or support",2026-07-07T19:02:00Z +issue,189161,[Inductor] Remove Redundant NaN Propagation in `max` op in Softmax,https://github.com/pytorch/pytorch/issues/189161,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T15:30:19Z +issue,189151,DISABLED test_pending_fusion_pro_and_epi_use_async_compile_False (__main__.TestPrologueFusion),https://github.com/pytorch/pytorch/issues/189151,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T13:25:32Z +issue,152944,DISABLED test_compiler_collectives_automatic_dynamic_tensor (__main__.TestMultiProc),https://github.com/pytorch/pytorch/issues/152944,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: dynamo, module: flaky-tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T13:25:25Z +issue,186865,DISABLED test_dynamic_shape_backend_eager_device_cpu (__main__.TestPackage),https://github.com/pytorch/pytorch/issues/186865,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: flaky-tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T13:25:18Z +issue,180195,DISABLED test_vjp_vmap_cpu (__main__.TestComposabilityCPU),https://github.com/pytorch/pytorch/issues/180195,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T13:25:13Z +issue,189137,[Sub-RFC] PrivateUse1 Inductor codegen registration interfaces (RFC 3/3),https://github.com/pytorch/pytorch/issues/189137,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, module: backend, module: inductor, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-07-07T12:01:59Z +issue,189136,[Sub-RFC] PrivateUse1 Dynamo tracing via registry splicing (RFC 2/3),https://github.com/pytorch/pytorch/issues/189136,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: privateuse1, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T12:01:30Z +issue,189135,[Sub-RFC] PrivateUse1 device identity/capability in torch.compile (RFC 1/3),https://github.com/pytorch/pytorch/issues/189135,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: backend, module: dynamo, module: inductor, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-07-07T12:00:59Z +issue,189138,[RFC] DeviceInterface: Registry Support for PrivateUse1 Backends in torch.compile,https://github.com/pytorch/pytorch/issues/189138,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: backend, module: dynamo, module: inductor, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-07-07T11:20:54Z +issue,188798,DISABLED test_linearize_jvp_fn (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/188798,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T07:52:39Z +issue,176966,DISABLED test_trace_return_namedtuple (__main__.TestFX),https://github.com/pytorch/pytorch/issues/176966,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T07:52:34Z +issue,124152,DISABLED test_torchvision_models_efficientnet_v2_m (__main__.TestVisionTracing),https://github.com/pytorch/pytorch/issues/124152,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx","decide whether this is bug, use case, docs, or support",2026-07-07T07:52:32Z +issue,188987,DISABLED test_optimizer_non_static_param (__main__.LoggingTests),https://github.com/pytorch/pytorch/issues/188987,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: logging, module: tests, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T06:22:00Z +issue,189046,Treat PrivateUse1 as CUDA-like under env override to reuse CUDA optimization paths in shared `at::native` operators,https://github.com/pytorch/pytorch/issues/189046,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: advanced indexing, module: autograd, module: backend, module: dispatch","decide whether this is bug, use case, docs, or support",2026-07-07T02:58:57Z +issue,188602,[RFC] Optimizing CUDA+AArch64 builds using LLVM-BOLT,https://github.com/pytorch/pytorch/issues/188602,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: build, module: ci, module: cuda","decide whether this is bug, use case, docs, or support",2026-07-07T02:04:59Z +issue,189067,DISABLED test_pyobject_redispatch_requires_keyset (__main__.TestCustomOpWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/189067,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T01:15:59Z +issue,180337,DISABLED test_vmap_grad_cpu (__main__.TestComposabilityCPU),https://github.com/pytorch/pytorch/issues/180337,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T01:15:46Z +issue,188239,DISABLED test_comprehensive_linalg_polar_xpu_float32 (__main__.TestInductorOpInfoXPU),https://github.com/pytorch/pytorch/issues/188239,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: linear algebra, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-07T01:05:29Z +issue,189078,[RFC] Build profiling for PyTorch,https://github.com/pytorch/pytorch/issues/189078,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: performance","decide whether this is bug, use case, docs, or support",2026-07-06T21:57:17Z +issue,189093,Misleading full-backward-hook warning when only `register_full_backward_pre_hook` is used,https://github.com/pytorch/pytorch/issues/189093,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: nn","decide whether this is bug, use case, docs, or support",2026-07-06T21:25:42Z +issue,188669,fx_graph_cse skipping duplicate element-wise CUDA optimization.,https://github.com/pytorch/pytorch/issues/188669,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fx.passes, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T19:44:16Z +issue,189077,operator_microbenchmark + attention_op_microbenchmark: b200 jobs never complete -- run ~8h and are cancelled at the workflow timeout,https://github.com/pytorch/pytorch/issues/189077,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: benchmark, module: ci, module: compile-time, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T19:39:16Z +issue,189075,periodic cuda13.0 slow-gradcheck: default test shards time out (~270 min) before finishing -- most shards killed,https://github.com/pytorch/pytorch/issues/189075,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-07-06T19:27:49Z +issue,189066,DISABLED test_jacrev (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/189066,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T19:15:07Z +issue,140556,DISABLED test_backward_prod_cuda_float32 (__main__.TestNestedTensorOpInfoCUDA),https://github.com/pytorch/pytorch/issues/140556,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: nestedtensor","decide whether this is bug, use case, docs, or support",2026-07-06T19:11:19Z +issue,188564,rocm-navi31 gfx1100 test (default) jobs time out after 270 min on inductor/test_torchinductor (red since 2026-05-20),https://github.com/pytorch/pytorch/issues/188564,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: compile-time, module: convolution, module: inductor","decide whether this is bug, use case, docs, or support",2026-07-06T17:19:11Z +issue,188382,torch.special.logit with eps > 0.5 has tiny vectorized-vs-scalar exact parity mismatch on non-contiguous float32 input,https://github.com/pytorch/pytorch/issues/188382,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: special, module: vectorization, module: windows","decide whether this is bug, use case, docs, or support",2026-07-06T16:59:45Z +issue,189032,nightly / update-metamates-rule failing since ~2026-06-29 — UPDATEBOT_TOKEN can't read the pytorch/metamates team,https://github.com/pytorch/pytorch/issues/189032,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-07-06T15:34:13Z +issue,120308,ROCm: foreach is substantially slower than for loop MI250x,https://github.com/pytorch/pytorch/issues/120308,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, module: rocm","decide whether this is bug, use case, docs, or support",2026-07-06T15:13:23Z +issue,188983,DISABLED test_comprehensive_combinations_cuda_float16 (__main__.TestInductorOpInfoCUDA),https://github.com/pytorch/pytorch/issues/188983,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T13:57:24Z +issue,189015,DISABLED test_jacfwd_two_tensors_argnums (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/189015,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T13:57:21Z +issue,188705,DISABLED test_reduction_not_fused (__main__.TestNVUniversalGemmEpilogueFusion),https://github.com/pytorch/pytorch/issues/188705,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T10:24:27Z +issue,188677,DISABLED test_graph_partition_view_fallback (__main__.CudaGraphTreeTests),https://github.com/pytorch/pytorch/issues/188677,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T10:19:00Z +issue,89431,The problem caused by the parameter dim of torch.norm,https://github.com/pytorch/pytorch/issues/89431,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: error checking","decide whether this is bug, use case, docs, or support",2026-07-06T09:25:16Z +issue,140718,'lobcpg' gives an output on CPU but fails on GPU('mps'),https://github.com/pytorch/pytorch/issues/140718,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: mps","decide whether this is bug, use case, docs, or support",2026-07-06T07:21:35Z +issue,171659,[RFC] RISC-V Architecture Support Roadmap for PyTorch,https://github.com/pytorch/pytorch/issues/171659,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: risc-v","decide whether this is bug, use case, docs, or support",2026-07-06T05:41:43Z +issue,188953,"Intel Arc B580 (Battlemage): ""could not create an engine"" on XPU — torch 2.6.0+xpu",https://github.com/pytorch/pytorch/issues/188953,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dependency bug, module: linear algebra, module: xpu","decide whether this is bug, use case, docs, or support",2026-07-06T05:08:42Z +issue,188985,DISABLED test_output_grad_match_combinations_mps_float16 (__main__.TestConsistencyMPS),https://github.com/pytorch/pytorch/issues/188985,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: macos, module: mps","decide whether this is bug, use case, docs, or support",2026-07-06T01:15:06Z +issue,188041,DISABLED test_jacfwd_randomness (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/188041,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-06T01:14:57Z +issue,175891,[RFC] torch.switch,https://github.com/pytorch/pytorch/issues/175891,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-05T21:09:29Z +issue,188926,[CPU] torch.unravel_index raises ZeroDivisionError when shape has a zero dimension,https://github.com/pytorch/pytorch/issues/188926,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: edge cases, module: empty tensor, module: error checking","decide whether this is bug, use case, docs, or support",2026-07-05T05:26:05Z +issue,188319,[RFC] Publish official ROCm Docker images on pytorch/pytorch,https://github.com/pytorch/pytorch/issues/188319,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: ci, module: docker, module: rocm","decide whether this is bug, use case, docs, or support",2026-07-04T16:30:49Z +issue,68506,cudnn_convolution_relu is 17x slower for 7x7 convolutions and channels last,https://github.com/pytorch/pytorch/issues/68506,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cudnn","decide whether this is bug, use case, docs, or support",2026-07-04T08:23:51Z +issue,188935,find_library(ROCM_ROCTX_LIB roctx64 ...) should be gated behind USE_ROCTRACER,https://github.com/pytorch/pytorch/issues/188935,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: rocm","decide whether this is bug, use case, docs, or support",2026-07-04T00:51:38Z +issue,187738,DISABLED test_grad_two_tensor_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/187738,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-03T20:01:08Z +issue,188899,DISABLED test_cache_hot_load_caching_precompile_dynamic_True_device_cuda_bfloat16 (__main__.TestFxGraphCache),https://github.com/pytorch/pytorch/issues/188899,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-03T13:18:10Z +issue,188897,DISABLED test_addmm_relu_tunableop_cuda_float32 (__main__.TestLinalgCudaOnlyCUDA),https://github.com/pytorch/pytorch/issues/188897,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: linear algebra, module: rocm","decide whether this is bug, use case, docs, or support",2026-07-03T13:18:03Z +issue,188895,DISABLED test_conv2d_backward_parametrized_dilation_2_stride_1_padding_1_kernel_3_channels_groups4_nhwc_True_cuda (__main__.GPUTests),https://github.com/pytorch/pytorch/issues/188895,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-03T13:17:59Z +issue,187718,DISABLED test_grad_two_tensor_all_grad_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/187718,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-03T13:17:52Z +issue,159113,DISABLED test_graph_two_successive (__main__.TestCuda),https://github.com/pytorch/pytorch/issues/159113,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-07-03T13:17:49Z +issue,188670,[RFC] Inductor / AOTI tests: decouple “correct codegen path” from tight eager-fp16 equality,https://github.com/pytorch/pytorch/issues/188670,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, module: inductor, module: tests, oncall: export","decide whether this is bug, use case, docs, or support",2026-07-03T12:31:22Z +issue,187658,DISABLED test_grad_non_tensor_input (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/187658,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-03T07:45:47Z +issue,180414,DISABLED test_torchvision_models_efficientnet_b2 (__main__.TestVisionTracing),https://github.com/pytorch/pytorch/issues/180414,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx","decide whether this is bug, use case, docs, or support",2026-07-03T07:45:40Z +issue,181629,DISABLED test_torchvision_models_efficientnet_b5 (__main__.TestVisionTracing),https://github.com/pytorch/pytorch/issues/181629,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx","decide whether this is bug, use case, docs, or support",2026-07-03T07:45:39Z +issue,188871,[FlexAttention] FLASH backend: cute.where Int64/Int32 dtype mismatch in captured-buffer negative-index wrap under dynamic shapes,https://github.com/pytorch/pytorch/issues/188871,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: flex attention, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-03T03:47:46Z +issue,188851,[CI][ROCm] DebertaV2ForMaskedLM fails to import from transformers on gfx950 (dynamic_inductor_huggingface eager_fail_to_run),https://github.com/pytorch/pytorch/issues/188851,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: benchmark, module: ci, module: dependency bug, module: rocm","decide whether this is bug, use case, docs, or support",2026-07-03T00:20:40Z +issue,188648,"treespec_dumps does not preserve tuple keys in dict/OrderedDict/defaultdict pytrees, so the round-trip fails with `unhashable type: 'list'`",https://github.com/pytorch/pytorch/issues/188648,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pytree, module: serialization","decide whether this is bug, use case, docs, or support",2026-07-02T21:43:19Z +issue,188840,DISABLED test_eager_equivalence_nn_functional_gelu_backend_inductor_default_cuda_float32 (__main__.TestOpInfoPropertiesCUDA),https://github.com/pytorch/pytorch/issues/188840,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: nn, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-02T19:43:18Z +issue,188837,DISABLED test_dropout_parity_cudagraphs_reduce_overhead (__main__.TestDropoutAlignRandomEager),https://github.com/pytorch/pytorch/issues/188837,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-02T19:24:35Z +issue,179979,DISABLED test_sparse_tensor_multiprocessing_context_spawn_cuda (__main__.TestDataLoaderDeviceTypeCUDA),https://github.com/pytorch/pytorch/issues/179979,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: multiprocessing, module: sparse","decide whether this is bug, use case, docs, or support",2026-07-02T19:05:11Z +issue,187425,DISABLED test_grad_closure_scalar (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/187425,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-02T19:02:55Z +issue,179798,DISABLED test_memory_plots_free_segment_stack (__main__.TestCudaAllocator),https://github.com/pytorch/pytorch/issues/179798,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-07-02T19:02:53Z +issue,188676,DISABLED test_float8_e4m3fn_uint8_decode_codegen_cuda (__main__.TestFP8TypesCUDA),https://github.com/pytorch/pytorch/issues/188676,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-02T18:48:43Z +issue,153236,DISABLED test_clip_grad_norm_2d (__main__.TestClipGradNormWorldSize4),https://github.com/pytorch/pytorch/issues/153236,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-07-02T16:41:08Z +issue,168168,[CD][ARM][CUDA] PYTORCH Released Docker Images Should Ideally Support Multi-ARCH But Currently It is only x86_64,https://github.com/pytorch/pytorch/issues/168168,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: cuda","decide whether this is bug, use case, docs, or support",2026-07-02T15:22:10Z +issue,188727,torch.distributed.pipelining _split_block_mask hides batch offset from non-strict tracing,https://github.com/pytorch/pytorch/issues/188727,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-07-02T06:30:59Z +issue,179152,`linalg_cholesky_ex_out` contains some dubious logic,https://github.com/pytorch/pytorch/issues/179152,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-07-02T03:17:36Z +issue,188767,DISABLED test_reconfigure_scale_down_up (__main__.GlooFaultToleranceTest),https://github.com/pytorch/pytorch/issues/188767,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-07-02T01:24:57Z +issue,188707,DISABLED test_main_loop_scaling_shape0_use_fast_accum_False_scaling_block_sizes1_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188707,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-07-02T00:34:16Z +issue,188706,DISABLED test_main_loop_scaling_shape0_use_fast_accum_False_scaling_block_sizes0_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188706,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-07-02T00:34:16Z +issue,188704,DISABLED test_main_loop_scaling_shape2_use_fast_accum_True_scaling_block_sizes1_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188704,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-07-02T00:34:15Z +issue,188740,DISABLED test_linear_block_compile_parity_backward (__main__.TestDropoutAlignRandomEager),https://github.com/pytorch/pytorch/issues/188740,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T22:06:26Z +issue,188721,UNSTABLE B200 Smoke Tests / linux-jammy-cuda13.0-py3.12-gcc11-sm100 / test-osdc (smoke_b200),https://github.com/pytorch/pytorch/issues/188721,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-07-01T19:21:42Z +issue,180542,DISABLED test_vmap_two_inputs_tuple_in_dims (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/180542,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:16:03Z +issue,145344,DISABLED test_cat (__main__.NumBytesMetricTests),https://github.com/pytorch/pytorch/issues/145344,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:16:02Z +issue,188714,DISABLED test_main_loop_scaling_shape1_use_fast_accum_True_scaling_block_sizes1_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188714,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: floatx (formerly float8), module: inductor, module: tests","decide whether this is bug, use case, docs, or support",2026-07-01T19:04:21Z +issue,188715,DISABLED test_main_loop_scaling_shape1_use_fast_accum_True_scaling_block_sizes2_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188715,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:04:19Z +issue,188712,DISABLED test_main_loop_scaling_shape1_use_fast_accum_False_scaling_block_sizes1_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188712,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: floatx (formerly float8), module: inductor, module: tests","decide whether this is bug, use case, docs, or support",2026-07-01T19:04:17Z +issue,188709,DISABLED test_main_loop_scaling_shape0_use_fast_accum_True_scaling_block_sizes0_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188709,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:04:15Z +issue,188713,DISABLED test_main_loop_scaling_shape1_use_fast_accum_False_scaling_block_sizes2_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188713,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: floatx (formerly float8), module: inductor, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-07-01T19:04:13Z +issue,188717,DISABLED test_main_loop_scaling_shape2_use_fast_accum_False_scaling_block_sizes2_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188717,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:04:11Z +issue,188708,DISABLED test_main_loop_scaling_shape0_use_fast_accum_False_scaling_block_sizes2_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188708,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:03:40Z +issue,188716,DISABLED test_main_loop_scaling_shape2_use_fast_accum_False_scaling_block_sizes1_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188716,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:03:27Z +issue,188710,DISABLED test_main_loop_scaling_shape0_use_fast_accum_True_scaling_block_sizes1_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188710,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T19:03:19Z +issue,188693,DISABLED test_vjp_scalar_logging_tensor (__main__.TestAutogradFunctional),https://github.com/pytorch/pytorch/issues/188693,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: tests","decide whether this is bug, use case, docs, or support",2026-07-01T16:19:38Z +issue,188692,DISABLED test_mkldnn_rnn_backward_gru_bias_shape (__main__.TestMetaKernelRegistrations),https://github.com/pytorch/pytorch/issues/188692,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: meta tensors, module: mkldnn, module: rnn, module: tests","decide whether this is bug, use case, docs, or support",2026-07-01T16:16:22Z +issue,188688,DISABLED test_backend_simple (__main__.PrivateUse1BackendTest),https://github.com/pytorch/pytorch/issues/188688,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: privateuse1, module: tests","decide whether this is bug, use case, docs, or support",2026-07-01T16:14:20Z +issue,188687,DISABLED test_linear_block_compile_parity_forward (__main__.TestDropoutAlignRandomEager),https://github.com/pytorch/pytorch/issues/188687,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: random, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T16:05:15Z +issue,188686,DISABLED test_redispatch_nn_functional_conv_transpose3d_cuda_complex32 (__main__.TestTorchFunctionRedispatchOpsCUDA),https://github.com/pytorch/pytorch/issues/188686,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_function__, module: complex, module: convolution, module: cuda","decide whether this is bug, use case, docs, or support",2026-07-01T16:04:51Z +issue,188684,DISABLED test_redispatch_nn_functional_conv_transpose3d_cuda_bfloat16 (__main__.TestTorchFunctionRedispatchOpsCUDA),https://github.com/pytorch/pytorch/issues/188684,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_function__, module: bfloat16, module: convolution, module: cuda","decide whether this is bug, use case, docs, or support",2026-07-01T16:04:00Z +issue,188675,DISABLED test_main_loop_scaling_shape2_use_fast_accum_True_scaling_block_sizes2_cuda (__main__.TestFP8LoweringCUDA),https://github.com/pytorch/pytorch/issues/188675,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T15:21:50Z +issue,188672,[Inductor] Prefer CUDA Toolkit libdevice for Triton-backed Inductor kernels,https://github.com/pytorch/pytorch/issues/188672,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T15:04:50Z +issue,188652,Pytorch can't be built in an air-gapped environment,https://github.com/pytorch/pytorch/issues/188652,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: nnpack","decide whether this is bug, use case, docs, or support",2026-07-01T14:12:10Z +issue,185534,Sub-optimal heuristics in MixOrderReduction prevent fusion,https://github.com/pytorch/pytorch/issues/185534,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: performance, module: reductions, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T14:11:37Z +issue,188668,DISABLED test_extern_kernel_benchmark_valid_timing (__main__.TestExternKernelCaller),https://github.com/pytorch/pytorch/issues/188668,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T13:32:58Z +issue,156818,Tighter invariants for the inductor graph,https://github.com/pytorch/pytorch/issues/156818,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T08:06:35Z +issue,187378,DISABLED test_grad_capture_tensor (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/187378,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T08:03:48Z +issue,180490,DISABLED test_vmap_two_inputs (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/180490,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T08:03:41Z +issue,182868,[RFC] [Inductor] L2 cache aware welford/two-pass selction,https://github.com/pytorch/pytorch/issues/182868,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T03:27:25Z +issue,188420,torch.compile produces inconsistent outputs for MultiHeadAttentionModel with scaled_dot_product_attention,https://github.com/pytorch/pytorch/issues/188420,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: determinism, module: random, module: sdpa, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T03:02:28Z +issue,117974,RuntimeError: derivative for aten::_scaled_dot_product_efficient_attention_backward is not implemented,https://github.com/pytorch/pytorch/issues/117974,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sdpa,"decide whether this is bug, use case, docs, or support",2026-07-01T01:35:30Z +issue,188628,DISABLED test_rank0_broadcast_meta_device_init (__main__.TestFullyShardMetaDeviceInit),https://github.com/pytorch/pytorch/issues/188628,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-07-01T01:33:04Z +issue,188627,DISABLED test_control_deps_passthrough_creates_ordering_barrier (__main__.TestControlDeps),https://github.com/pytorch/pytorch/issues/188627,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T01:33:03Z +issue,187000,DISABLED test_upper_bound_i64_cuda (__main__.AOTInductorTestDualWrapper),https://github.com/pytorch/pytorch/issues/187000,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T01:32:55Z +issue,187176,DISABLED test_functional_call (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/187176,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T01:32:49Z +issue,149080,DISABLED test_aoti_debug_printer_codegen_cuda (__main__.AOTInductorTestABICompatibleGpu),https://github.com/pytorch/pytorch/issues/149080,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T01:32:46Z +issue,182260,DISABLED test_assert_size_stride_guarded_by_env_check_cpu_with_stack_allocation (__main__.AOTInductorTestABICompatibleCpuWithStackAllocation),https://github.com/pytorch/pytorch/issues/182260,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-07-01T01:32:43Z +issue,187046,DISABLED test_tensor_with_unbacked_shape_closure (__main__.HigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/187046,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-30T19:16:09Z +issue,180439,DISABLED test_vmap_over_vmap_captured (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/180439,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-30T19:16:06Z +issue,48628,Add docs on PyTorch - NumPy interaction,https://github.com/pytorch/pytorch/issues/48628,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: numpy","decide whether this is bug, use case, docs, or support",2026-06-30T12:15:56Z +issue,188542,PyProcessGroup is missing several `ProcessGroup` collective overrides,https://github.com/pytorch/pytorch/issues/188542,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-30T10:53:52Z +issue,186872,DISABLED test_call_count_tunableop_cuda_float32 (__main__.TestLinalgCudaOnlyCUDA),https://github.com/pytorch/pytorch/issues/186872,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-06-30T07:59:52Z +issue,186909,DISABLED test_trace_only_export_matches_default_use_cuda_True (__main__.TestProfiler),https://github.com/pytorch/pytorch/issues/186909,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: profiler","decide whether this is bug, use case, docs, or support",2026-06-30T07:59:46Z +issue,180366,DISABLED test_vmap_free_const (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/180366,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-30T07:59:45Z +issue,187996,TestDTensorStateDictStager.test_async_save_no_memory_leak fails on GB200,https://github.com/pytorch/pytorch/issues/187996,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-30T07:51:56Z +issue,106665,Inconsistency between CPU and GPU for `Linear()` layer with input size 0,https://github.com/pytorch/pytorch/issues/106665,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: nn","decide whether this is bug, use case, docs, or support",2026-06-30T07:37:01Z +issue,188360,Inconsistent and unclear error messages for `offsets` validation when `input.dim() == 1` in `torch/nn/functional.py`,https://github.com/pytorch/pytorch/issues/188360,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: error checking, module: nn","decide whether this is bug, use case, docs, or support",2026-06-30T06:29:08Z +issue,187290,DISABLED test_Conv2d_depthwise_kernel_flag_cuda_float16 (__main__.TestConvolutionNNCUDACUDA),https://github.com/pytorch/pytorch/issues/187290,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: nn","decide whether this is bug, use case, docs, or support",2026-06-30T01:25:58Z +issue,180324,DISABLED test_jvp_two_tensors_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/180324,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T19:17:43Z +issue,187463,PyTorch has Inconsistent Negative Indexing Semantics,https://github.com/pytorch/pytorch/issues/187463,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: advanced indexing, module: python frontend, module: scatter & gather ops","decide whether this is bug, use case, docs, or support",2026-06-29T18:07:23Z +issue,188446,Add issue de-duplication system,https://github.com/pytorch/pytorch/issues/188446,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: devx,"decide whether this is bug, use case, docs, or support",2026-06-29T17:50:26Z +issue,132845,[PTD BE DAY]Burn Down Distributed Disabled Tests!!,https://github.com/pytorch/pytorch/issues/132845,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: distributed,"decide whether this is bug, use case, docs, or support",2026-06-29T17:39:14Z +issue,187473,[torch 2.13] umbrella issue - vLLM CI failures,https://github.com/pytorch/pytorch/issues/187473,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: vllm, oncall: pt2, oncall: releng","decide whether this is bug, use case, docs, or support",2026-06-29T17:21:49Z +issue,188430,"torch.compile produces large absolute errors for model with LSTM, MultiheadAttention, and torch.reciprocal",https://github.com/pytorch/pytorch/issues/188430,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T17:18:16Z +issue,188432,torch.compile produces inconsistent outputs for Multi-Query Attention model with manually implemented attention,https://github.com/pytorch/pytorch/issues/188432,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T17:17:55Z +issue,188375,"CPU LayerNorm normalized_shape=(1,) gives nonzero weight.grad although output is independent of weight",https://github.com/pytorch/pytorch/issues/188375,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: bfloat16, module: correctness (silent), module: cpu","decide whether this is bug, use case, docs, or support",2026-06-29T17:15:40Z +issue,187974,reserving issue for myself,https://github.com/pytorch/pytorch/issues/187974,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-06-29T17:08:44Z +issue,188074,torch.compile produces inconsistent outputs for stateful SNN model with custom LIF neurons,https://github.com/pytorch/pytorch/issues/188074,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: determinism, module: dynamo, module: nn","decide whether this is bug, use case, docs, or support",2026-06-29T17:08:27Z +issue,176662,C++20 Rollout Tracking,https://github.com/pytorch/pytorch/issues/176662,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cpp,"decide whether this is bug, use case, docs, or support",2026-06-29T17:05:49Z +issue,154449,16KB pagination support for PyTorch Torch Vision and PyTorch Lite,https://github.com/pytorch/pytorch/issues/154449,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: android,"decide whether this is bug, use case, docs, or support",2026-06-29T16:05:30Z +issue,167684,[RFC] Access to ncclDevComm in kernel,https://github.com/pytorch/pytorch/issues/167684,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-29T12:32:44Z +issue,188222,[aot_autograd] `RuntimeError: view size is not compatible ... use .reshape()` under `torch.compile`,https://github.com/pytorch/pytorch/issues/188222,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: functionalization, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T10:36:07Z +issue,188005,[RFC] Protect SymmetricMemory signal pads against cross-stream races,https://github.com/pytorch/pytorch/issues/188005,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-29T09:38:16Z +issue,186030,[DTensor][local_map]Support `None` placements in `local_map` placement specs,https://github.com/pytorch/pytorch/issues/186030,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-29T08:30:15Z +issue,180204,DISABLED test_jacrev_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/180204,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T08:25:25Z +issue,188393,DISABLED test_comprehensive_linalg_matrix_sqrth_xpu_float32 (__main__.TestInductorOpInfoXPU),https://github.com/pytorch/pytorch/issues/188393,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: linear algebra, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T06:56:35Z +issue,188392,DISABLED test_comprehensive_linalg_matrix_sqrth_xpu_float64 (__main__.TestInductorOpInfoXPU),https://github.com/pytorch/pytorch/issues/188392,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: linear algebra, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T06:53:33Z +issue,188391,DISABLED test_compare_cpu_linalg_matrix_sqrth_xpu_float32 (__main__.TestCommonXPU),https://github.com/pytorch/pytorch/issues/188391,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: tests, module: xpu","decide whether this is bug, use case, docs, or support",2026-06-29T06:51:21Z +issue,188238,DISABLED test_comprehensive_linalg_polar_xpu_float64 (__main__.TestInductorOpInfoXPU),https://github.com/pytorch/pytorch/issues/188238,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: linear algebra, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-29T06:50:23Z +issue,188389,[Docs] torch.nn.Module.parameters() does not specify whether the ordering is deterministic,https://github.com/pytorch/pytorch/issues/188389,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-06-29T05:58:42Z +issue,147971,Fp8 scaled-mm row-wise is substantially slower than tensor-wise,https://github.com/pytorch/pytorch/issues/147971,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: floatx (formerly float8), module: performance","decide whether this is bug, use case, docs, or support",2026-06-28T06:22:30Z +issue,175229,Idiomatic way of matching against flex attention regions in SAC policy with wrap_inductor_compiled_regions,https://github.com/pytorch/pytorch/issues/175229,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: flex attention, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-28T03:27:23Z +issue,77764,General MPS op coverage tracking issue,https://github.com/pytorch/pytorch/issues/77764,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: mps,"decide whether this is bug, use case, docs, or support",2026-06-27T11:10:58Z +issue,139792,Speed up `DataLoader` worker start time for large datasets,https://github.com/pytorch/pytorch/issues/139792,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dataloader,"decide whether this is bug, use case, docs, or support",2026-06-27T02:12:17Z +issue,188317,[RFC] Add process group revoke in torch.distributed,https://github.com/pytorch/pytorch/issues/188317,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-27T00:27:05Z +issue,166166,MultiheadAttention fast_path is substantially less efficient than the alternative,https://github.com/pytorch/pytorch/issues/166166,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-06-26T23:55:53Z +issue,179897,DISABLED test_jacfwd (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/179897,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T19:15:43Z +issue,179853,DISABLED test_hessian_argnums (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/179853,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T19:15:36Z +issue,188077,torch._dynamo.disable seems expensive,https://github.com/pytorch/pytorch/issues/188077,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T14:23:21Z +issue,179811,DISABLED test_hessian (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/179811,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T13:21:40Z +issue,188205,Warn if torch.Tensor.backward() or torch.autograd.grad() is called underneath torch.cuda.use_mem_pool() with torch.autograd.grad_mode.set_multithreading_enabled(True),https://github.com/pytorch/pytorch/issues/188205,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: cuda","decide whether this is bug, use case, docs, or support",2026-06-26T12:15:36Z +issue,179990,DISABLED test_jacfwd_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),https://github.com/pytorch/pytorch/issues/179990,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T07:59:07Z +issue,187019,DISABLED test_combo_kernel_no_bench_persistent_reduction (__main__.ComboKernelTests),https://github.com/pytorch/pytorch/issues/187019,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: tests, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T01:52:28Z +issue,188234,DISABLED test_del_attr_multi_stream_sync_dealloc (__main__.TestStreams),https://github.com/pytorch/pytorch/issues/188234,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: flaky-tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T01:30:04Z +issue,188232,[aot_autograd] `linalg.eigh` fails to converge (ill-conditioned) under `torch.compile` but not eager,https://github.com/pytorch/pytorch/issues/188232,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: edge cases, module: linear algebra, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-06-26T01:30:01Z +issue,181322,DISABLED test_trace_return_dataclass_nested (__main__.TestFX),https://github.com/pytorch/pytorch/issues/181322,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T01:29:52Z +issue,179874,DISABLED test_lazy_init_with_view (__main__.LazyFuncionalizationTest),https://github.com/pytorch/pytorch/issues/179874,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, oncall: jit, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T01:29:48Z +issue,188212,[RFC] Add Gluon Backend To Inductor Backend,https://github.com/pytorch/pytorch/issues/188212,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flex attention, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-26T00:13:14Z +issue,162588,scaled_dot_product_attention produces incorrect results with attn_mask on large sequence lengths,https://github.com/pytorch/pytorch/issues/162588,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sdpa,"decide whether this is bug, use case, docs, or support",2026-06-25T22:34:26Z +issue,136275,Please offer `+cpu`-versioned packages for MacOS and linux ARM CPU-only builds,https://github.com/pytorch/pytorch/issues/136275,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: releng,"decide whether this is bug, use case, docs, or support",2026-06-25T22:10:13Z +issue,178368,On the future of triton-windows,https://github.com/pytorch/pytorch/issues/178368,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: ci, module: windows, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-25T21:47:02Z +issue,187862,`OpOverload.__call__` is roughly 100ns overhead,https://github.com/pytorch/pytorch/issues/187862,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: performance, module: pt2-dispatcher, module: python frontend","decide whether this is bug, use case, docs, or support",2026-06-25T19:29:14Z +issue,188151,torch.cond fails in torch.compile with simpliest symbolic shapes (torch 2.12+),https://github.com/pytorch/pytorch/issues/188151,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: dynamo, module: higher order operators, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-06-25T17:40:51Z +issue,158032,extension-cpp claims ROCM support with no additional changes; we should test this.,https://github.com/pytorch/pytorch/issues/158032,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpp-extensions, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-25T16:41:24Z +issue,181499,RocM: Assumed realtime clock rate,https://github.com/pytorch/pytorch/issues/181499,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-25T16:34:28Z +issue,181939,HIP related errors compiling against PyTorch,https://github.com/pytorch/pytorch/issues/181939,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cpp, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-25T16:21:47Z +issue,186350,[Feature Request] Distribute triton with device-local version identifiers instead of separate package names,https://github.com/pytorch/pytorch/issues/186350,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: rocm, module: xpu","decide whether this is bug, use case, docs, or support",2026-06-25T15:52:40Z +issue,187715,`torch.compile(mode='reduce-overhead')` silently degrades training on ROCm — `cudagraph_trees` has no HIP guard,https://github.com/pytorch/pytorch/issues/187715,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda graphs, module: inductor, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-25T15:44:55Z +issue,187907,AOTInductor: reverse-mode autograd.grad through sqrt/reciprocal bakes a symbolic-shaped constant under a dynamic batch (strict export),https://github.com/pytorch/pytorch/issues/187907,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, module: dynamic shapes, module: inductor, oncall: export","decide whether this is bug, use case, docs, or support",2026-06-25T15:30:04Z +issue,188153,"torch.export.export Fails with RuntimeError and ""Please file an issue on github""",https://github.com/pytorch/pytorch/issues/188153,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: pt2,"decide whether this is bug, use case, docs, or support",2026-06-25T14:23:09Z +issue,130330,why not enable expandable_segments by default？,https://github.com/pytorch/pytorch/issues/130330,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: performance","decide whether this is bug, use case, docs, or support",2026-06-25T12:57:50Z +issue,188144,DISABLED test_custom_op_dce_dynamic_shapes_cuda (__main__.DynamicShapesCodegenGPUTests),https://github.com/pytorch/pytorch/issues/188144,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-25T07:55:18Z +issue,188123,DISABLED test_index_cuda_with_cpu_int64_propagate_real_tensors (__main__.PropagateRealTensorsFakeTensorTest),https://github.com/pytorch/pytorch/issues/188123,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: meta tensors, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-25T07:55:02Z +issue,153460,DISABLED test_mempool_ctx_multithread (__main__.TestMemPool),https://github.com/pytorch/pytorch/issues/153460,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-06-25T07:55:00Z +issue,188135,DISABLED test_compare_cpu_linalg_polar_xpu_float32 (__main__.TestCommonXPU),https://github.com/pytorch/pytorch/issues/188135,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: tests, module: xpu","decide whether this is bug, use case, docs, or support",2026-06-25T06:03:25Z +issue,186871,DISABLED test_addmm_relu_tunableop_rocm_cuda_float32 (__main__.TestLinalgCudaOnlyCUDA),https://github.com/pytorch/pytorch/issues/186871,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: linear algebra, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-25T01:27:39Z +issue,180322,DISABLED test_vjp_vjp_cpu (__main__.TestComposabilityCPU),https://github.com/pytorch/pytorch/issues/180322,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-25T01:27:23Z +issue,187615,[FSDP2] Fractional CPU offload policy (PartialOffloadPolicy),https://github.com/pytorch/pytorch/issues/187615,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-25T00:23:11Z +issue,188001,[RFC] Unify rendezvous caching across SymmetricMemory backends,https://github.com/pytorch/pytorch/issues/188001,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-24T18:56:32Z +issue,188087,"torch.nanmedian(..., dim=None) raises a named-dimension RuntimeError while omitting dim works",https://github.com/pytorch/pytorch/issues/188087,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: error checking, module: reductions","decide whether this is bug, use case, docs, or support",2026-06-24T18:13:41Z +issue,187963,Pallas CPU max_unpool2d does not preserve channels-last output layout,https://github.com/pytorch/pytorch/issues/187963,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: memory format, module: pallas, module: pooling, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-24T17:31:33Z +issue,9983,[feature request] Add matrix functions,https://github.com/pytorch/pytorch/issues/9983,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: numpy","decide whether this is bug, use case, docs, or support",2026-06-24T11:33:27Z +issue,187649,[Inductor] Support fx wrapper with non-triton fused kernel.,https://github.com/pytorch/pytorch/issues/187649,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fx, module: higher order operators, module: inductor, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-06-24T08:58:29Z +issue,186825,Bernoulli(probs=0).log_prob(1) returns dtype-dependent finite values instead of -inf (Geometric special-cases this and tests it),https://github.com/pytorch/pytorch/issues/186825,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributions, module: edge cases","decide whether this is bug, use case, docs, or support",2026-06-24T03:49:18Z +issue,188020,Allow platform context in DataLoader worker fatal signal messages,https://github.com/pytorch/pytorch/issues/188020,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dataloader,"decide whether this is bug, use case, docs, or support",2026-06-24T03:00:14Z +issue,117662,dynamo <> autograd.Function assumes that grad_outputs have the same subclass-ness as forward outputs,https://github.com/pytorch/pytorch/issues/117662,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: dynamo, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-23T20:27:09Z +issue,185587,[inductor] Numerical inconsistency: polygamma + cumulative_trapezoid under torch.compile(dynamic=True) produces wrong results after Conv2d + GroupNorm,https://github.com/pytorch/pytorch/issues/185587,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: dynamic shapes, module: inductor, module: special","decide whether this is bug, use case, docs, or support",2026-06-23T17:30:32Z +issue,185295,[Inductor][feature] Block FP8 quantization: fuse scale layout transformation with quant math,https://github.com/pytorch/pytorch/issues/185295,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-06-23T17:27:07Z +issue,186381,[export] No-op contiguous() elided during tracing bakes trace-time layout into the graph; downstream view becomes layout-specific,https://github.com/pytorch/pytorch/issues/186381,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-23T16:30:11Z +issue,186594,torch.ops.higher_order.associative_scan cannot be torch.export'ed,https://github.com/pytorch/pytorch/issues/186594,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-23T16:29:54Z +issue,186999,DISABLED test_bmm_large_batch_dynamic_cuda (__main__.AOTInductorTestDualWrapper),https://github.com/pytorch/pytorch/issues/186999,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-23T16:29:45Z +issue,148819,addition of muon optimizer to torch.optim,https://github.com/pytorch/pytorch/issues/148819,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: optimizer,"decide whether this is bug, use case, docs, or support",2026-06-23T16:25:57Z +issue,187183,Update slow tests job removes tests that may still be slow,https://github.com/pytorch/pytorch/issues/187183,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: tests","decide whether this is bug, use case, docs, or support",2026-06-23T14:31:00Z +issue,187807,[PyTorch 2.13][API-Unstable] Intel XPU Enhancements,https://github.com/pytorch/pytorch/issues/187807,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: windows, module: xpu","decide whether this is bug, use case, docs, or support",2026-06-23T07:36:47Z +issue,180974,[RFC][inductor] Support pre/post kernel execution hooks in compiled graphs,https://github.com/pytorch/pytorch/issues/180974,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-23T06:40:17Z +issue,93531,TorchInductor CPU Performance Dashboard,https://github.com/pytorch/pytorch/issues/93531,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: cpu inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-23T06:13:49Z +issue,173071,Allow privateuse1 devices in FlexAttention device validation,https://github.com/pytorch/pytorch/issues/173071,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flex attention, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-23T04:49:48Z +issue,159547,grouped_mm for float16,https://github.com/pytorch/pytorch/issues/159547,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: half, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-06-23T03:39:52Z +issue,186260,[Accelerator] Maybe host memory should be included in torch.accelerator.mem_stats,https://github.com/pytorch/pytorch/issues/186260,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: accelerator,"decide whether this is bug, use case, docs, or support",2026-06-22T22:54:41Z +issue,187770,torch.nn.attention.varlen: make backend selection configurable,https://github.com/pytorch/pytorch/issues/187770,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sdpa,"decide whether this is bug, use case, docs, or support",2026-06-22T21:14:09Z +issue,178196,libvnrtc error in torch nightly on cuda 13.0,https://github.com/pytorch/pytorch/issues/178196,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: cuda","decide whether this is bug, use case, docs, or support",2026-06-22T16:20:54Z +issue,187721,[Inductor] Silent correctness : F.kl_div reduction over -0.0 yields different signbits in Eager vs Compiled,https://github.com/pytorch/pytorch/issues/187721,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: half, module: inductor, module: loss","decide whether this is bug, use case, docs, or support",2026-06-22T15:23:49Z +issue,187834,reminder to check cu129 binaries aren't released with pytorch 2.14,https://github.com/pytorch/pytorch/issues/187834,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: ci","decide whether this is bug, use case, docs, or support",2026-06-22T14:11:00Z +issue,186939,pitch: fast development mode where we use pre-build nightly .so files,https://github.com/pytorch/pytorch/issues/186939,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: devx,"decide whether this is bug, use case, docs, or support",2026-06-22T13:43:56Z +issue,187827,[pr-review skill] Add MPS/Metal kernel review checks (tracking),https://github.com/pytorch/pytorch/issues/187827,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: mps,"decide whether this is bug, use case, docs, or support",2026-06-22T13:41:29Z +issue,105632,torch.nn.TransformerDecoderLayer lacks parameter validation check,https://github.com/pytorch/pytorch/issues/105632,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-06-22T12:49:53Z +issue,174131,turn on autograd_cache_normalize_inputs in fbcode,https://github.com/pytorch/pytorch/issues/174131,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pt2-dispatcher, module: vllm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-22T11:05:17Z +issue,186465,[RFC] Fix Control flow operators consistency,https://github.com/pytorch/pytorch/issues/186465,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: higher order operators, module: inductor, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-06-22T07:05:47Z +issue,145679,Gamma cdf function,https://github.com/pytorch/pytorch/issues/145679,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: distributions,"decide whether this is bug, use case, docs, or support",2026-06-21T21:14:19Z +issue,121135,[torch.compile] `randperm_index_add_pattern` doesn't check the shape mismatch between self and source tensor,https://github.com/pytorch/pytorch/issues/121135,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-20T13:54:37Z +issue,153809,Optimize rowwise scaled mm CUTLASS kernel for large values of K,https://github.com/pytorch/pytorch/issues/153809,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8),"decide whether this is bug, use case, docs, or support",2026-06-20T08:26:31Z +issue,143281,torch.linalg.qr is significantly slower on GPU compared to CPU and SVD for batched small matrices,https://github.com/pytorch/pytorch/issues/143281,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-06-20T05:18:44Z +issue,181861,nn.Sequential type hints too general for nested lists,https://github.com/pytorch/pytorch/issues/181861,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: typing,"decide whether this is bug, use case, docs, or support",2026-06-20T02:12:37Z +issue,187180,torch.nn.functional.softplus returns +Inf for finite float16 input when beta is very small,https://github.com/pytorch/pytorch/issues/187180,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: edge cases, module: half, module: nn","decide whether this is bug, use case, docs, or support",2026-06-19T18:52:25Z +issue,187470,MemberType LLM readability,https://github.com/pytorch/pytorch/issues/187470,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-19T18:31:39Z +issue,187561,DISABLED test_view_ops (__main__.TestViewOpsWithLocalTensor),https://github.com/pytorch/pytorch/issues/187561,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: flaky-tests, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-19T08:40:27Z +issue,155512,Unable to find caffe2 when building libtorch from the source and then trying to use it in a cmake project,https://github.com/pytorch/pytorch/issues/155512,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cpp","decide whether this is bug, use case, docs, or support",2026-06-19T07:59:54Z +issue,186826,"Distribution.icdf performs no argument validation even with validate_args=True, unlike cdf/log_prob",https://github.com/pytorch/pytorch/issues/186826,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributions, module: error checking","decide whether this is bug, use case, docs, or support",2026-06-19T06:35:43Z +issue,187698,"Vulkan nn.Linear unavailable with requires_grad=True on Windows (inference works, training path blocked)",https://github.com/pytorch/pytorch/issues/187698,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: nn, module: vulkan, module: windows","decide whether this is bug, use case, docs, or support",2026-06-18T21:26:43Z +issue,187241,"[inductor] TypeError: Too few arguments for CSE in cutedsl_kernel.py (torch 2.10.0, Python 3.11)",https://github.com/pytorch/pytorch/issues/187241,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: python version, module: typing, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-18T20:45:58Z +issue,181428,torch.compile + torch.func.grad: superlinear compile time with MaxPool2d → log/clamp → pow → AdaptiveAvgPool2d,https://github.com/pytorch/pytorch/issues/181428,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile-time, module: dynamic shapes, module: functorch, module: inductor","decide whether this is bug, use case, docs, or support",2026-06-18T19:27:18Z +issue,110700,More informative variable names in AOTAutograd,https://github.com/pytorch/pytorch/issues/110700,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: fx, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-18T18:16:25Z +issue,164666,sparse.mm(CSR x dense) backward fails on CUDA for bfloat16 but CSR @ dense backward works,https://github.com/pytorch/pytorch/issues/164666,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: bfloat16, module: sparse","decide whether this is bug, use case, docs, or support",2026-06-18T17:38:04Z +issue,187239,[NCCL][FSDP2] Support parameter-contiguous all-gather output to skip copy-out,https://github.com/pytorch/pytorch/issues/187239,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: fsdp, module: performance, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-18T16:39:08Z +issue,187645,DISABLED test_async_future_results (__main__.TestAsyncCheckpointerSpecific),https://github.com/pytorch/pytorch/issues/187645,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-18T08:23:18Z +issue,179745,DISABLED test_memory_snapshot (__main__.TestCudaAllocator),https://github.com/pytorch/pytorch/issues/179745,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-06-18T08:23:08Z +issue,185452,DISABLED test_torchvision_models_convnext_large (__main__.TestVisionTracing),https://github.com/pytorch/pytorch/issues/185452,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx","decide whether this is bug, use case, docs, or support",2026-06-18T08:22:58Z +issue,177010,[RFC] Introduce the unified RNG-related API to torch.accelerator,https://github.com/pytorch/pytorch/issues/177010,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: accelerator, module: random","decide whether this is bug, use case, docs, or support",2026-06-18T07:55:22Z +issue,167365,PyTorch is no longer importable from `build` folder,https://github.com/pytorch/pytorch/issues/167365,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-06-18T05:41:19Z +issue,184466,Implementing Self-Regulated Kinetic Learning (SRKL) Optimizer using Second-Order Curvature,https://github.com/pytorch/pytorch/issues/184466,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: optimizer","decide whether this is bug, use case, docs, or support",2026-06-18T03:32:00Z +issue,187179,"torch.cdist returns +Inf for small p (p-norm overflow, docstring admits p in [0, inf])",https://github.com/pytorch/pytorch/issues/187179,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distance functions, module: edge cases, module: numerical-stability","decide whether this is bug, use case, docs, or support",2026-06-17T20:51:55Z +issue,187591,[Dynamo] Symbolic Strings,https://github.com/pytorch/pytorch/issues/187591,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: guards, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-17T18:47:23Z +issue,187590,[Dynamo] LazyConstants -- Avoid over-guarding on constant values,https://github.com/pytorch/pytorch/issues/187590,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: guards, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-17T18:35:04Z +issue,118107,[Tracker] Nested tensor op coverage requests,https://github.com/pytorch/pytorch/issues/118107,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nestedtensor,"decide whether this is bug, use case, docs, or support",2026-06-17T18:01:16Z +issue,187582,torch.nested NJTs with identical layout fail on binary operation,https://github.com/pytorch/pytorch/issues/187582,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nestedtensor,"decide whether this is bug, use case, docs, or support",2026-06-17T17:48:25Z +issue,104053,Tracking issue for optimizer graph not being an inference graph,https://github.com/pytorch/pytorch/issues/104053,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-17T15:17:58Z +issue,186601,[RFC]SDMA AllGather and Zero-Copy Output for PyTorch FSDP2,https://github.com/pytorch/pytorch/issues/186601,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-17T14:00:54Z +issue,187148,[Proposal] Fix cpu_fallback returning tensors on wrong device for reverse binary ops,https://github.com/pytorch/pytorch/issues/187148,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: backend, module: dispatch, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-06-17T07:59:05Z +issue,185639,"torch.func.jacfwd gives incorrect Jacobian for torch.pow(torch.abs(complex), complex exponent)",https://github.com/pytorch/pytorch/issues/185639,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: complex, module: functorch","decide whether this is bug, use case, docs, or support",2026-06-17T02:04:28Z +issue,187505,OpaqueObject Fuzzing,https://github.com/pytorch/pytorch/issues/187505,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: faketensor, module: fx, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-06-16T21:53:49Z +issue,163429,[CI][CUDA][B200] test_max_autotune.py failures on B200 runner,https://github.com/pytorch/pytorch/issues/163429,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: cuda, module: tests","decide whether this is bug, use case, docs, or support",2026-06-16T18:56:31Z +issue,187293,"torch.nn.functional.gelu(approximate=""none"") returns NaN for +Inf input",https://github.com/pytorch/pytorch/issues/187293,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: nans and infs, module: nn","decide whether this is bug, use case, docs, or support",2026-06-16T14:08:06Z +issue,187276,Two-node DDP ResNet-50/Adam run ends with matching rank-1 loss but gradient and parameter-fingerprint divergence across CPU/Gloo and CUDA/NCCL,https://github.com/pytorch/pytorch/issues/187276,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-16T10:49:58Z +issue,187279,Two-node DDP shows reproducible FP32 loss divergence between CPU/Gloo and CUDA/NCCL with large lr and resnet18,https://github.com/pytorch/pytorch/issues/187279,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-16T09:22:50Z +issue,187424,DISABLED test_range_based_static_shape_no_cond_dispatch (__main__.TestCustomOpAutoTune),https://github.com/pytorch/pytorch/issues/187424,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-16T08:36:49Z +issue,104701,System memory leak when using different input size of torch.nn.Conv3d,https://github.com/pytorch/pytorch/issues/104701,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cudnn, module: memory usage, module: nn","decide whether this is bug, use case, docs, or support",2026-06-16T06:36:35Z +issue,162859,[RFC] support symmetric memory in torch.compile,https://github.com/pytorch/pytorch/issues/162859,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: symm_mem, module: vllm, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-15T23:55:21Z +issue,187158,[Inductor] AssertionError: Unregistered range symbol 'yindex' in TMA template epilogue fusion,https://github.com/pytorch/pytorch/issues/187158,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-15T23:03:03Z +issue,187227,[SDPA] Fix 32-bit offset overflow in mem-efficient attention forward with attn_bias,https://github.com/pytorch/pytorch/issues/187227,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda, module: int overflow, module: sdpa","decide whether this is bug, use case, docs, or support",2026-06-15T22:17:26Z +issue,156771,Support fp16 and fp32 outputs for rowwise scaled_mm,https://github.com/pytorch/pytorch/issues/156771,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8),"decide whether this is bug, use case, docs, or support",2026-06-15T13:32:37Z +issue,180168,[RFC] XPUGraph Trees,https://github.com/pytorch/pytorch/issues/180168,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: xpu, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-15T03:49:07Z +issue,186449,[RFC]: Contextvars Support in TorchDynamo,https://github.com/pytorch/pytorch/issues/186449,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: dynamo, module: graph breaks, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-14T17:41:28Z +issue,182001,CUDA linalg: use auxiliary streams for independent looped cuSOLVER batch calls,https://github.com/pytorch/pytorch/issues/182001,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: linear algebra, module: performance","decide whether this is bug, use case, docs, or support",2026-06-14T09:58:26Z +issue,187266,scaled_dot_product_attention falls back to math for unbatched (3D) inputs under vmap,https://github.com/pytorch/pytorch/issues/187266,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functorch, module: performance, module: sdpa, module: vmap","decide whether this is bug, use case, docs, or support",2026-06-14T06:07:48Z +issue,187020,DISABLED test_redistribute_cost_latency (__main__.TestCostModel),https://github.com/pytorch/pytorch/issues/187020,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: flaky-tests, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-13T16:39:20Z +issue,173123,"[DTensor] Gradient of unused output becomes plain torch.Tensor in backward, causing type mismatch with torch.compile",https://github.com/pytorch/pytorch/issues/173123,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-13T16:38:34Z +issue,172491,Direct access to DTensor._local_tensor breaks _dynamo_graph_capture_for_export,https://github.com/pytorch/pytorch/issues/172491,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-13T16:38:02Z +issue,172485,[DTensor] Summary of issues related to `Partial` placement,https://github.com/pytorch/pytorch/issues/172485,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-13T16:37:52Z +issue,179374,"[Infra] Missing PEP 700 upload-time metadata breaks supply-chain security tools (uv, pip)",https://github.com/pytorch/pytorch/issues/179374,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: infra,"decide whether this is bug, use case, docs, or support",2026-06-13T03:02:17Z +issue,186143,AOTInductor Lowering Backward Compatibility,https://github.com/pytorch/pytorch/issues/186143,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, module: docs, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-12T21:25:07Z +issue,187203,DISABLED test_disable_external_correlation (__main__.TestProfilerCUDA),https://github.com/pytorch/pytorch/issues/187203,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: profiler","decide whether this is bug, use case, docs, or support",2026-06-12T19:22:08Z +issue,187202,DISABLED test_combo_kernel_fusion_within_same_stream (__main__.TestUserStreamCompile),https://github.com/pytorch/pytorch/issues/187202,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-12T19:22:06Z +issue,170479,nvshmem [GPU 0] Peer GPU 8 is not accessible,https://github.com/pytorch/pytorch/issues/170479,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-12T17:10:47Z +issue,177245,AArch64 Unit Test Failures - Multiple failures in test/nn/test_convolution.py TestConvolutionNNDeviceTypeCPU for oneDNN,https://github.com/pytorch/pytorch/issues/177245,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: convolution, module: cpu, module: mkldnn","decide whether this is bug, use case, docs, or support",2026-06-12T17:10:32Z +issue,169814,DTensor.detach() retains stale DeviceMesh after destroying and reinitializing process groups,https://github.com/pytorch/pytorch/issues/169814,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-12T17:09:18Z +issue,162257,Optimize FSDP1 sharded_state_dict saving performance,https://github.com/pytorch/pytorch/issues/162257,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-12T17:09:12Z +issue,169320,[DTensor] A New Placement -- RaggedShard,https://github.com/pytorch/pytorch/issues/169320,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-12T17:08:21Z +issue,169035,"[Question] Why torch.ops.symm_mem.multimem_all_reduce_() don't support e4m3, e5m2, fp16?",https://github.com/pytorch/pytorch/issues/169035,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-12T17:08:20Z +issue,167917,Operator aten.detach_.default does not have a sharding strategy registered,https://github.com/pytorch/pytorch/issues/167917,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-12T17:07:32Z +issue,167897,Hooks as extension points for torch.pipelining,https://github.com/pytorch/pytorch/issues/167897,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-12T17:07:31Z +issue,167657,make_fx trick for tracing DTensor compute doesn't work with redistribute,https://github.com/pytorch/pytorch/issues/167657,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-12T17:06:57Z +issue,186618,Inconsistent rounding between CUDA and CPU for division operator,https://github.com/pytorch/pytorch/issues/186618,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: cuda, module: edge cases","decide whether this is bug, use case, docs, or support",2026-06-12T08:20:55Z +issue,187156,DISABLED test_load_with_map_location (__main__.TestCheckpointer),https://github.com/pytorch/pytorch/issues/187156,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-12T08:10:37Z +issue,186837,[RFC]: Auto-select and run Operator Performance Benchmarks from PR Diffs,https://github.com/pytorch/pytorch/issues/186837,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: performance","decide whether this is bug, use case, docs, or support",2026-06-12T04:23:15Z +issue,186905,download.pytorch.org packages doesn't have release timestamp,https://github.com/pytorch/pytorch/issues/186905,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: binaries,"decide whether this is bug, use case, docs, or support",2026-06-11T20:08:59Z +issue,187045,DISABLED test_out_of_memory_retry (__main__.TestCuda),https://github.com/pytorch/pytorch/issues/187045,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-06-11T19:31:29Z +issue,167074,DTensor with `torch.compile` fails with `aten.as_strided.default` error,https://github.com/pytorch/pytorch/issues/167074,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:25:46Z +issue,167219,Are there limitations to dtensor's registration strategy?,https://github.com/pytorch/pytorch/issues/167219,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:25:14Z +issue,167090,DTensor `F.conv2d` fails with `bias=None` - AssertionError in sharding propagation,https://github.com/pytorch/pytorch/issues/167090,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:25:12Z +issue,163692,[feature][dtensor] support nn.Module inputs containing DTensor parameters in local_map,https://github.com/pytorch/pytorch/issues/163692,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:24:00Z +issue,166638,[DTensor] calling item on dtensor should perform full tensor first,https://github.com/pytorch/pytorch/issues/166638,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:23:56Z +issue,163688,[torch.distributed.pipelining] Gradients are None in first training step with ScheduleGPipe,https://github.com/pytorch/pytorch/issues/163688,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: amp (automated mixed precision), module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:23:15Z +issue,161974,DTensor communications cost model is way off for H100,https://github.com/pytorch/pytorch/issues/161974,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:23:10Z +issue,162215,consider allowing more dtype in the dtensor opinfo test,https://github.com/pytorch/pytorch/issues/162215,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:22:33Z +issue,161847,Feedback about FSDP Notes,https://github.com/pytorch/pytorch/issues/161847,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:22:06Z +issue,160968,[DTensor] broadcast_to fails with KeyError,https://github.com/pytorch/pytorch/issues/160968,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:21:46Z +issue,161218,[DTensor] Losing TupleStrategy in OpSchema.args_strategy,https://github.com/pytorch/pytorch/issues/161218,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:21:37Z +issue,155657,"[DTensor] DTensor is not well supported on older versions of GPUs, such as A10",https://github.com/pytorch/pytorch/issues/155657,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-11T17:21:09Z +issue,185142,RFC Test class classification,https://github.com/pytorch/pytorch/issues/185142,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: tests","decide whether this is bug, use case, docs, or support",2026-06-11T16:38:14Z +issue,175869,[dynamo] Tracking issue for itertools polyfills,https://github.com/pytorch/pytorch/issues/175869,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-11T13:43:34Z +issue,34130,ExceptionWrapper cannot handle exceptions with more than one argument,https://github.com/pytorch/pytorch/issues/34130,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: error checking","decide whether this is bug, use case, docs, or support",2026-06-11T09:25:14Z +issue,158911,[RFC] Refactor test_c10d_nccl.py to Support Backend-Agnostic Testing for PyTorch,https://github.com/pytorch/pytorch/issues/158911,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: privateuse1, module: testing, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-11T07:02:48Z +issue,180345,Allow registering multiple out-of-tree backends and setting current_accelerator,https://github.com/pytorch/pytorch/issues/180345,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: accelerator, module: backend","decide whether this is bug, use case, docs, or support",2026-06-11T04:44:45Z +issue,186911,DISABLED test_launch_elastic_multiple_agents (__main__.ElasticLaunchTest),https://github.com/pytorch/pytorch/issues/186911,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, oncall: r2p","decide whether this is bug, use case, docs, or support",2026-06-10T22:26:30Z +issue,184352,Python 3.15 support for PyTorch,https://github.com/pytorch/pytorch/issues/184352,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: python frontend, module: python version","decide whether this is bug, use case, docs, or support",2026-06-10T21:31:03Z +issue,160340,Meta device initialization on HF models + FSDP leads to different numerical behaviour,https://github.com/pytorch/pytorch/issues/160340,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:16:37Z +issue,160911,Severe differences between outputs of Linear with DTensor and Tensor in half precision,https://github.com/pytorch/pytorch/issues/160911,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:16:24Z +issue,160169,"When using DP + TP, DP only parameters diverge across TP ranks if using operations with non-deterministic implementations",https://github.com/pytorch/pytorch/issues/160169,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:16:15Z +issue,160068,FSDP2 not compatible with transformers >= 4.54.0 GenericForTokenClassification,https://github.com/pytorch/pytorch/issues/160068,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:15:47Z +issue,160272,Add Green Context Support for fsdp2,https://github.com/pytorch/pytorch/issues/160272,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:15:39Z +issue,160648,How to Use Pipeline Parallelism in Multi-input Models,https://github.com/pytorch/pytorch/issues/160648,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:15:38Z +issue,159836,Missing documentation for device mesh on DDP,https://github.com/pytorch/pytorch/issues/159836,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: docs, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:15:04Z +issue,159959,TensorMetadata not retained during DTensor convolution double backpropagation,https://github.com/pytorch/pytorch/issues/159959,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:15:03Z +issue,156217,[Tracker] AutoParallel's feature request to DTensor,https://github.com/pytorch/pytorch/issues/156217,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:14:25Z +issue,154657,[FSDP2] offer public API to share communication context aross fsdp roots,https://github.com/pytorch/pytorch/issues/154657,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:14:09Z +issue,145218,getting different results when adding `torch.Tensor` or python number to a DTensor - Is that expected?,https://github.com/pytorch/pytorch/issues/145218,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:13:32Z +issue,146884,[DTensor]Operator aten.mT.default/aten.mv.default/aten.dot.default does not have a sharding strategy registered,https://github.com/pytorch/pytorch/issues/146884,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:13:27Z +issue,143748,[DTensor]`Linear` fails on 3D DTensor with `batch size > 1` and `Replicate` input redistributed from `Shard`,https://github.com/pytorch/pytorch/issues/143748,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:13:20Z +issue,143823,Possible race condition found in TailLogTest.test_tail,https://github.com/pytorch/pytorch/issues/143823,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: elastic, module: tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-10T17:13:07Z +issue,143985,[DTensor] Allow multiple dimensions to be sharded together (as if flattened),https://github.com/pytorch/pytorch/issues/143985,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:12:51Z +issue,143647,A very weird bug involving ddp,https://github.com/pytorch/pytorch/issues/143647,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-10T17:12:16Z +issue,185770,"torch.nn.functional.gelu returns NaN for +inf input with approximate=""none""",https://github.com/pytorch/pytorch/issues/185770,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: nans and infs, module: nn","decide whether this is bug, use case, docs, or support",2026-06-10T16:18:18Z +issue,186912,DISABLED test_checkpoint_write_future_state_dict (__main__.TestCheckpointProcess),https://github.com/pytorch/pytorch/issues/186912,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: rocm, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-10T08:02:47Z +issue,186910,DISABLED test_Embedding_discontiguous_cuda (__main__.TestNN),https://github.com/pytorch/pytorch/issues/186910,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: nn, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-10T08:02:41Z +issue,186908,DISABLED test_nested_tensor_multiprocessing_context_forkserver_cuda (__main__.TestDataLoaderDeviceTypeCUDA),https://github.com/pytorch/pytorch/issues/186908,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: flaky-tests, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-10T08:02:35Z +issue,186873,DISABLED test_mm_submatrix_offline_tunableop_cuda_float32 (__main__.TestLinalgCudaOnlyCUDA),https://github.com/pytorch/pytorch/issues/186873,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: linear algebra, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-10T08:02:10Z +issue,86735,DISABLED test_reentrant_parent_error_on_cpu_cuda (__main__.TestAutogradDeviceTypeCUDA),https://github.com/pytorch/pytorch/issues/86735,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-06-10T08:02:00Z +issue,186868,DISABLED test_LSTM_dropout_per_call_randomness_dropout_p_0_5_training_True_cuda (__main__.TestNNDeviceTypeCUDA),https://github.com/pytorch/pytorch/issues/186868,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: nn, module: rocm","decide whether this is bug, use case, docs, or support",2026-06-10T08:01:58Z +issue,104695,DISABLED test_cuda_memory_leak_detection (__main__.TestCudaMultiGPU),https://github.com/pytorch/pytorch/issues/104695,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-06-10T08:01:58Z +issue,181318,DISABLED test_codegen_structure_three_stream_pipeline (__main__.TestUserStreamCompile),https://github.com/pytorch/pytorch/issues/181318,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-10T08:01:55Z +issue,181224,DISABLED test_cross_stream_stride_copy (__main__.TestUserStreamCompile),https://github.com/pytorch/pytorch/issues/181224,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-10T08:01:54Z +issue,186864,DISABLED test_triggers (__main__.CallbackTests),https://github.com/pytorch/pytorch/issues/186864,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: flaky-tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-10T08:01:40Z +issue,183603,torch.compile fails on second-order gradient after autograd.grad(create_graph=True),https://github.com/pytorch/pytorch/issues/183603,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: double backwards, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-10T07:27:02Z +issue,106692,torch.polygamma inconsistent with scipy.special.polygamma for n >= 1,https://github.com/pytorch/pytorch/issues/106692,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: numpy, module: special","decide whether this is bug, use case, docs, or support",2026-06-10T02:55:54Z +issue,182174,DISABLED test_torchvision_models_efficientnet_b0 (__main__.TestVisionTracing),https://github.com/pytorch/pytorch/issues/182174,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx","decide whether this is bug, use case, docs, or support",2026-06-10T01:29:39Z +issue,100074,DISABLED test_checkpointing_resets_persistent_refs (__main__.CudaGraphTreeTests),https://github.com/pytorch/pytorch/issues/100074,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor","decide whether this is bug, use case, docs, or support",2026-06-10T01:29:33Z +issue,138464,`torch.special.zeta` ignores `nan` input when `other=-inf`.,https://github.com/pytorch/pytorch/issues/138464,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: nans and infs, module: python frontend, module: special","decide whether this is bug, use case, docs, or support",2026-06-09T18:51:05Z +issue,164563,[Blackwell][Inductor] Numerical mismatches in test_max_autotune.py::TestMaxAutotune::test_non_contiguous_input_mm_plus_mm,https://github.com/pytorch/pytorch/issues/164563,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-09T17:29:44Z +issue,156523,[Dtensor] handle dtensor ops that only need to operate on certain shard without all_gather first.,https://github.com/pytorch/pytorch/issues/156523,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-09T17:05:00Z +issue,156533,`<<` and `>>` operators seem silently broken for DTensor operand 1 and scalar operand 2,https://github.com/pytorch/pytorch/issues/156533,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-09T17:04:57Z +issue,156784,[FSDP2] allow different dtypes for the model params with gradients,https://github.com/pytorch/pytorch/issues/156784,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-09T17:04:44Z +issue,155589,"pipeline() fails when a sub-module uses ""no_grad()""; impacts RoPE implementation on HF models",https://github.com/pytorch/pytorch/issues/155589,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-09T17:03:54Z +issue,153484,Pytorch PP requires all parameters to have grad in backward,https://github.com/pytorch/pytorch/issues/153484,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-09T17:02:46Z +issue,143721,"add ""enabled=True"" to DistributedDataParallel.no_sync()",https://github.com/pytorch/pytorch/issues/143721,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-09T17:00:45Z +issue,144608,Build breaks on FreeBSD on arm platforms: Unrecognized CMAKE_SYSTEM_NAME = FreeBSD,https://github.com/pytorch/pytorch/issues/144608,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-06-09T10:15:41Z +issue,125604,Compiling GPT2 with tensorrt backend and dynamic shapes results in guard failures.,https://github.com/pytorch/pytorch/issues/125604,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-09T02:05:42Z +issue,182259,DISABLED test_torchvision_models_convnext_base (__main__.TestVisionTracing),https://github.com/pytorch/pytorch/issues/182259,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: fx","decide whether this is bug, use case, docs, or support",2026-06-09T01:24:27Z +issue,178153,[RFC] Refactor Triton Heuristics via Unified Device-Scoped Class Hierarchy,https://github.com/pytorch/pytorch/issues/178153,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-08T23:43:29Z +issue,167885,Ideas for Simple CUDA Dynamic Analysis Tools for Common Bugs,https://github.com/pytorch/pytorch/issues/167885,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: performance","decide whether this is bug, use case, docs, or support",2026-06-08T17:49:37Z +issue,142356,AllReduce hangs due to no device_id into init_process_group,https://github.com/pytorch/pytorch/issues/142356,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:16:53Z +issue,142095,torch.distributed.new_group got stuck when use backend=mpi,https://github.com/pytorch/pytorch/issues/142095,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mpi, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:16:30Z +issue,140069,DTensor support for fused qkv matmul,https://github.com/pytorch/pytorch/issues/140069,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-08T17:16:28Z +issue,139129,[pgnccl] unstable restarting of PGs,https://github.com/pytorch/pytorch/issues/139129,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:16:24Z +issue,137014,[RFC] Allow user to abort group (followed by custom recovery),https://github.com/pytorch/pytorch/issues/137014,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:15:24Z +issue,137010,[RFC] Investigate multi-thread support of c10d,https://github.com/pytorch/pytorch/issues/137010,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:15:18Z +issue,136950,Incomplete Parameter Gathering on Rank 0 with FSDP Model Saving,https://github.com/pytorch/pytorch/issues/136950,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-08T17:15:08Z +issue,135358,"UserWarning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. grad.sizes() = [1, 64, 64, 64], strides() = [1, 4096, 64, 1] bucket_view.sizes() = [1, 64, 64, 64], strides() = [262144, 4096, 64, 1]",https://github.com/pytorch/pytorch/issues/135358,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-08T17:14:44Z +issue,136390,[c10d][nccl][cuda] Regression (unspecific cuda launch error) with test_c10d_nncl,https://github.com/pytorch/pytorch/issues/136390,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:14:20Z +issue,129862,The call to ncclCommSplit does not have the correct config parameters set,https://github.com/pytorch/pytorch/issues/129862,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nccl, module: performance, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:14:00Z +issue,124058,[RFC] Support Torchrun for Out-of-the-tree Devices,https://github.com/pytorch/pytorch/issues/124058,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T17:13:08Z +issue,186004,git fetch error: not our ref a56dcd79c699366e7ac6466792c3025883ff7704,https://github.com/pytorch/pytorch/issues/186004,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: third_party","decide whether this is bug, use case, docs, or support",2026-06-08T14:27:58Z +issue,186521,Feature Request: Add CPU-only PyTorch wheels for Termux (Android ARM64) + Python 3.13 (cp313),https://github.com/pytorch/pytorch/issues/186521,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: binaries","decide whether this is bug, use case, docs, or support",2026-06-08T05:00:29Z +issue,186548,"`ProcessGroupXCCL` never overrides `Backend::supportsSplitting()`, breaking `DeviceMesh._unflatten` (and any `dist.split_group` caller) on XPU",https://github.com/pytorch/pytorch/issues/186548,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: xpu, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-08T03:52:30Z +issue,168044,RFC: Internal Unification of LRScheduler Semantics for Correct Resuming and Composition,https://github.com/pytorch/pytorch/issues/168044,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: lrscheduler, module: optimizer","decide whether this is bug, use case, docs, or support",2026-06-07T20:34:06Z +issue,120158,"The forward accuracy cannot be aligned when using DTensor and ColwiseParallel, RowwiseParallel for TP training on the Llama2",https://github.com/pytorch/pytorch/issues/120158,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-07T16:39:34Z +issue,116683,DTensor tensor subwrapper not be handle with apex,https://github.com/pytorch/pytorch/issues/116683,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-07T16:39:11Z +issue,121594,[DDP] Gradient Synchronization Failure Induced by model.gradient_checkpointing_enable() in multi-task setting,https://github.com/pytorch/pytorch/issues/121594,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: activation checkpointing, module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-07T16:38:58Z +issue,120945,DISABLED distributed/_composable/fsdp/test_fully_shard_training.py::TestFullyShard1DTrainingCompose::test_train_parity_with_activation_checkpointing,https://github.com/pytorch/pytorch/issues/120945,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: activation checkpointing, module: fsdp, module: rocm, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-07T16:38:42Z +issue,120741,[C10D] coalesced op hang does not respect timeout,https://github.com/pytorch/pytorch/issues/120741,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-07T16:38:32Z +issue,114884,Undefined reference to ncclCommSplit,https://github.com/pytorch/pytorch/issues/114884,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-07T16:38:28Z +issue,115861,gather_object does not work with ThreadPoolExecutor,https://github.com/pytorch/pytorch/issues/115861,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-07T16:37:49Z +issue,113281,NCCL p2p ops hung after communicator aborts,https://github.com/pytorch/pytorch/issues/113281,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-07T16:37:47Z +issue,109477,[DTensor] optimizer step performance is still too bad,https://github.com/pytorch/pytorch/issues/109477,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-07T16:37:14Z +issue,106851,RPC all_gather doesn't work with dynamic world size (world_size=None),https://github.com/pytorch/pytorch/issues/106851,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rpc, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-07T16:37:14Z +issue,114437,[Help Needed] [DataParallel/DataLoader/AMP] Severe slowdown on large dataset with fp16 precision and DataParallel wrapper,https://github.com/pytorch/pytorch/issues/114437,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data parallel, module: dataloader, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-07T16:37:08Z +issue,90052,DDP overlapped optimizer: set grads to None enhancements,https://github.com/pytorch/pytorch/issues/90052,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data parallel, module: ddp, module: performance, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-07T16:36:06Z +issue,171938,Device id in distributed process group with multiple backends,https://github.com/pytorch/pytorch/issues/171938,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: distributed,"decide whether this is bug, use case, docs, or support",2026-06-07T15:17:56Z +issue,186496,"Typo in PyTorch 2.12 quantization docs: ""cetralizing"" should be ""centralizing""",https://github.com/pytorch/pytorch/issues/186496,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-06-07T15:05:24Z +issue,56340,log_softmax(x) != x - logsumexp(x),https://github.com/pytorch/pytorch/issues/56340,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: numerical-stability,"decide whether this is bug, use case, docs, or support",2026-06-07T03:04:19Z +issue,128396,torch.prod cannot be used with cudagraphs,https://github.com/pytorch/pytorch/issues/128396,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs,"decide whether this is bug, use case, docs, or support",2026-06-06T20:12:21Z +issue,95895,[PTD] dist.barrier() unreliable when using collectives from multiple threads.,https://github.com/pytorch/pytorch/issues/95895,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:35:30Z +issue,91629,vmap + nn.SyncBatchNorm.convert_sync_batchnorm,https://github.com/pytorch/pytorch/issues/91629,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data parallel, module: ddp, module: functorch, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-06T16:35:22Z +issue,90848,[Distributed] Destruction order fiasco in ProcessGroupNCCL workCleanupLoop(),https://github.com/pytorch/pytorch/issues/90848,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:35:21Z +issue,88838,[RFC] PyTorch DistributedTensor,https://github.com/pytorch/pytorch/issues/88838,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:34:10Z +issue,72025,[c10d] destruction of Store objects,https://github.com/pytorch/pytorch/issues/72025,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:30:28Z +issue,72034,Bug about distributed launch,https://github.com/pytorch/pytorch/issues/72034,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:30:24Z +issue,72099,halt or exit function implementation,https://github.com/pytorch/pytorch/issues/72099,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:30:20Z +issue,72107,DistributedDataParallel creates too many threads,https://github.com/pytorch/pytorch/issues/72107,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:30:16Z +issue,72346,torch.distributed.new_group() should consolidate the type of rank list,https://github.com/pytorch/pytorch/issues/72346,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:30:12Z +issue,72366,Custom ProcessGroup Destructor Not Correctly Called in PT 1.10,https://github.com/pytorch/pytorch/issues/72366,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:30:07Z +issue,72435,ProcessGroupWrapper: Additional Improvements,https://github.com/pytorch/pytorch/issues/72435,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: ddp, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:30:03Z +issue,72835,torch.distributed hangs at barrier(),https://github.com/pytorch/pytorch/issues/72835,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:58Z +issue,73081,Additional out of the box DDP comm hooks,https://github.com/pytorch/pytorch/issues/73081,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:53Z +issue,73332,BatchNorm with LSTM in DistributedDataParallel throws one of the variables needed for gradient computation has been modified by an inplace operation,https://github.com/pytorch/pytorch/issues/73332,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:45Z +issue,73960,processes hang when executing cross-machine asynchronous P2P communication on NCCL backend,https://github.com/pytorch/pytorch/issues/73960,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:36Z +issue,74162,consume_prefix_in_state_dict_if_present can not remove prefix of _metadata.,https://github.com/pytorch/pytorch/issues/74162,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:31Z +issue,74323,[c10d] `gather`/`scatter` inconsistency behavior on non-dst/src rank,https://github.com/pytorch/pytorch/issues/74323,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:27Z +issue,74528,Supports for dist.send/dist.recv sending and recving torch.shorttensor,https://github.com/pytorch/pytorch/issues/74528,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:23Z +issue,74602,My_rank in the implementation of torch.distributed.scatter_object_list should be global,https://github.com/pytorch/pytorch/issues/74602,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:19Z +issue,74666,Re-initializing torch.distributed process_group hangs with destroy_process_group,https://github.com/pytorch/pytorch/issues/74666,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:15Z +issue,74734,RuntimeError: Unconvertible NCCL type Short when sending torch.cuda.ShortTensor.,https://github.com/pytorch/pytorch/issues/74734,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:10Z +issue,74824,C10d Elastic Training master_addr ERROR,https://github.com/pytorch/pytorch/issues/74824,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra, oncall: r2p","decide whether this is bug, use case, docs, or support",2026-06-06T16:29:02Z +issue,75031,comm hook error in BWD pass,https://github.com/pytorch/pytorch/issues/75031,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:58Z +issue,75087,extend torch.distributed's `--tee` to log the nodename,https://github.com/pytorch/pytorch/issues/75087,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra, oncall: r2p","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:53Z +issue,75097,destroy_process_group() has a certain probability of hangs,https://github.com/pytorch/pytorch/issues/75097,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:49Z +issue,75334,ProcessGroupNCCL is relying on UB to support bool data type,https://github.com/pytorch/pytorch/issues/75334,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:45Z +issue,75657,1.11.0 distribution train different with 1.8.1,https://github.com/pytorch/pytorch/issues/75657,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:40Z +issue,75936,AllGather with backward support async_op=True,https://github.com/pytorch/pytorch/issues/75936,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:32Z +issue,76103,Some test failed when running in parallel.,https://github.com/pytorch/pytorch/issues/76103,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:28Z +issue,76287,torch.elastic fails to shutdown despite crashed processes,https://github.com/pytorch/pytorch/issues/76287,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:19Z +issue,76387,[checkpoint] Make prepare_sharded_tensor_read and prepare_sharded_tensor_write public,https://github.com/pytorch/pytorch/issues/76387,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:15Z +issue,76389,Error in DistributedDataParallel with 'CPU' device,https://github.com/pytorch/pytorch/issues/76389,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:11Z +issue,76395,[checkpoint] SPMD distributed checkpoint coordination,https://github.com/pytorch/pytorch/issues/76395,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-06T16:28:07Z +issue,76527,Non target rank receives result of 'reduce' op when backend is 'gloo',https://github.com/pytorch/pytorch/issues/76527,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-06T16:27:56Z +issue,148420,Float8_e4m3fn,https://github.com/pytorch/pytorch/issues/148420,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: mps","decide whether this is bug, use case, docs, or support",2026-06-06T07:18:25Z +issue,125557,result of 2 ** s differs between eager mode and inductor + triton + cuda when in float32 denormal range,https://github.com/pytorch/pytorch/issues/125557,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: pt2,"decide whether this is bug, use case, docs, or support",2026-06-06T06:32:16Z +issue,39242,Add numerically stable log1mexp = log(1 - exp(-|x|)) function,https://github.com/pytorch/pytorch/issues/39242,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: numerical-stability,"decide whether this is bug, use case, docs, or support",2026-06-06T04:26:47Z +issue,186405,arch_id.h: error: template constraint not satisfied,https://github.com/pytorch/pytorch/issues/186405,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cuda, module: dependency bug","decide whether this is bug, use case, docs, or support",2026-06-06T03:21:32Z +issue,113708,`log_softmax` could be `2**124` to `2**1021` times more accurate on small outputs,https://github.com/pytorch/pytorch/issues/113708,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-06-06T00:07:37Z +issue,186031,[torchcomms] FSDP+TP+PP+compile torchtitan flavor fails: functional collective cannot resolve torchcomms-managed TP process group under torch.compile,https://github.com/pytorch/pytorch/issues/186031,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: torchcomms, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T22:37:10Z +issue,76760,hanging process with init_process_group(backend='mpi') cannot be killed,https://github.com/pytorch/pytorch/issues/76760,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:13:03Z +issue,76913,Pytorch return TCPStore( RuntimeError: Connection reset by peer),https://github.com/pytorch/pytorch/issues/76913,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:54Z +issue,76927,Can't pickle model torch._C._distributed_c10d.ProcessGroupNCCL' object,https://github.com/pytorch/pytorch/issues/76927,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: ddp, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:49Z +issue,77154,Distributed Weighted Sampler.,https://github.com/pytorch/pytorch/issues/77154,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributed_tool, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:44Z +issue,77523,DDP window TCP bug [socket.cpp:558] [c10d] The client socket has failed to,https://github.com/pytorch/pytorch/issues/77523,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:39Z +issue,78638,"torch.distributed.init_process_group(backend=""nccl"") NCCL version error",https://github.com/pytorch/pytorch/issues/78638,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:34Z +issue,78842,Add TORCH_SHOW_CPP_STACKTRACES when TORCH_DISTRIBUTED_DEBUG = detail,https://github.com/pytorch/pytorch/issues/78842,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: bootcamp, module: c10d, module: distributed_tool, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:29Z +issue,78885,[distributed_test.py] Improve `test_barrier`,https://github.com/pytorch/pytorch/issues/78885,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:24Z +issue,79388,Init connect timeout when use torch.distributed.run,https://github.com/pytorch/pytorch/issues/79388,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra, oncall: r2p","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:19Z +issue,80496,DDP find_unused_parameters=True does not work for Sparse gradients,https://github.com/pytorch/pytorch/issues/80496,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:14Z +issue,80595,Overlapping Optimizer.step() with DDP backward,https://github.com/pytorch/pytorch/issues/80595,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: optimizer, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:09Z +issue,80774,RPC: Make RRefProxy callable,https://github.com/pytorch/pytorch/issues/80774,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rpc, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:12:04Z +issue,80824,"When running GPT trainning with megatron, the program quit due to torch.distributed.elastic.agent.server.api:Received 1 death signal, shutting down workers",https://github.com/pytorch/pytorch/issues/80824,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:11:59Z +issue,80832,[DDP] doesn't support multiple backwards when static_graph=True,https://github.com/pytorch/pytorch/issues/80832,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:11:54Z +issue,81383,Investigate ncclRedOpCreatePreMulSum operator for gradient reduction,https://github.com/pytorch/pytorch/issues/81383,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: fsdp, module: performance, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-05T17:11:49Z +issue,81622,[Distributed] test_dynamic_rpc_existing_rank_can_communicate_with_new_rank_cuda fails in caching allocator,https://github.com/pytorch/pytorch/issues/81622,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rpc, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:11:43Z +issue,82153,DistributedDataParallel hangs when not using GPU 0,https://github.com/pytorch/pytorch/issues/82153,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:11:32Z +issue,85949,"[Distributed] Loading distributed checkpoint with FSDP fails with varying key errors (pos.embedding, shared.weight)",https://github.com/pytorch/pytorch/issues/85949,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T17:11:12Z +issue,86456,`SyncBatchNorm` doesn't work with subclass of `torch.Tensor`,https://github.com/pytorch/pytorch/issues/86456,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:11:07Z +issue,88006,C++ Extensions can't import c10d/reducer.hpp,https://github.com/pytorch/pytorch/issues/88006,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:57Z +issue,88626,Cloud-based rendezvous backend / distributed store?,https://github.com/pytorch/pytorch/issues/88626,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:46Z +issue,89197,Collective operations do not work with `torch.BoolTensor`s on `gloo` and raise `Invalid scalar type`,https://github.com/pytorch/pytorch/issues/89197,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:41Z +issue,89546,Unable to launch CUDA Graph with DDP model,https://github.com/pytorch/pytorch/issues/89546,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:37Z +issue,89716,DDP hangs on forward pass of transformer,https://github.com/pytorch/pytorch/issues/89716,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:32Z +issue,90171,Option to let DistributedDataParallel know in advance unused parameters at each forward pass,https://github.com/pytorch/pytorch/issues/90171,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:27Z +issue,90245,[Distributed] `Invalid scalar type` when `dist.scatter()` boolean tensor,https://github.com/pytorch/pytorch/issues/90245,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:22Z +issue,90509,Can torchrun have a shell completion?,https://github.com/pytorch/pytorch/issues/90509,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:17Z +issue,90688,In distributed get SIGTERM and run crash,https://github.com/pytorch/pytorch/issues/90688,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:13Z +issue,91408,`NotImplementedError` when using `torch.distributed.launch`,https://github.com/pytorch/pytorch/issues/91408,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: functorch, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:10:08Z +issue,92694,API to check for errors in c10d.ProcessGroupNCCL,https://github.com/pytorch/pytorch/issues/92694,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:57Z +issue,95724,"distributed training: lots of ""Exception ignored"" at the end of each epoch",https://github.com/pytorch/pytorch/issues/95724,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:52Z +issue,99614,Make broadcast_coalesced to a op for processgroup,https://github.com/pytorch/pytorch/issues/99614,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:42Z +issue,99652,DistributedDataParallel doesn't work with complex buffers,https://github.com/pytorch/pytorch/issues/99652,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: complex, module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:37Z +issue,99908,Deprecate torch.distributed.algorithms._optimizer_overlap,https://github.com/pytorch/pytorch/issues/99908,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:32Z +issue,99978,torch.jit.trace can not trace buffer by Module.register_buffer() when use DDP Module.,https://github.com/pytorch/pytorch/issues/99978,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:27Z +issue,100582,"Synchronization issue when combining DPP and RPC - ""Parameter marked twice""",https://github.com/pytorch/pytorch/issues/100582,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: rpc, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:22Z +issue,101402,DataParallel for nested modules,https://github.com/pytorch/pytorch/issues/101402,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data parallel, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:17Z +issue,102197,torch.func.jvp fails when acting on a DistributedDataParallel model,https://github.com/pytorch/pytorch/issues/102197,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: forward ad, module: functorch, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:06Z +issue,102402,[Composable] Unified way to check if modules are managed by composable API,https://github.com/pytorch/pytorch/issues/102402,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:09:01Z +issue,102438,DDP multi node multi gpu inconsistent params,https://github.com/pytorch/pytorch/issues/102438,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:56Z +issue,103622,ARM based GPU support for Distributed Data Parallelism Module,https://github.com/pytorch/pytorch/issues/103622,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:51Z +issue,103625,DISABLED test_backward_ddp_outside (__main__.TensorPipeDdpUnderDistAutogradTest),https://github.com/pytorch/pytorch/issues/103625,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:46Z +issue,103626,2D model checkpointing hangs on a ViT model,https://github.com/pytorch/pytorch/issues/103626,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:41Z +issue,104011,DISABLED test_backward_ddp_outside_uneven_inputs (__main__.TensorPipeDdpUnderDistAutogradTest),https://github.com/pytorch/pytorch/issues/104011,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:36Z +issue,104164,DDP enhancement,https://github.com/pytorch/pytorch/issues/104164,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:31Z +issue,105723,Parameter ... has been marked as ready twice,https://github.com/pytorch/pytorch/issues/105723,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:21Z +issue,107174,Surface NCCL and CUDA version incompatibility,https://github.com/pytorch/pytorch/issues/107174,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:11Z +issue,107177,Timeout during NCCL initialization due to store,https://github.com/pytorch/pytorch/issues/107177,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:07Z +issue,108212,Is there a standard procedure to check the consistency of environment across all nodes in PyTorch DDP training?,https://github.com/pytorch/pytorch/issues/108212,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributed_tool, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:08:02Z +issue,108381,"FSDP always puts parameters to fp32 when loading state_dict, even if state_dict has bf16 params",https://github.com/pytorch/pytorch/issues/108381,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:57Z +issue,109103,"DDP - ""No backend type associated with device type cpu"" with new Model Phi 1.5 despite everything loaded on GPUs",https://github.com/pytorch/pytorch/issues/109103,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:52Z +issue,110281,feat(Pipeline Parallelism): use mincut optimization for local communication optimization,https://github.com/pytorch/pytorch/issues/110281,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:47Z +issue,110775,torch.distributed.pipeline skip module throws assert error that portal.grad is not None,https://github.com/pytorch/pytorch/issues/110775,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:42Z +issue,111131,"""Invalid Scalar type"" when using bf16 allreduce with Gloo backend",https://github.com/pytorch/pytorch/issues/111131,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:37Z +issue,111646,torchrun: elastic training not restarted on missing keep-alive heartbeat/scale-down event,https://github.com/pytorch/pytorch/issues/111646,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:27Z +issue,111834,"Numerical inaccuracies in ""ddp_apply_optim_in_backward"" unit tests for gloo backend",https://github.com/pytorch/pytorch/issues/111834,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:22Z +issue,111943,Implement scatter and broadcast in functional collectives,https://github.com/pytorch/pytorch/issues/111943,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:17Z +issue,112029,DDPWrapper,https://github.com/pytorch/pytorch/issues/112029,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:12Z +issue,112164,[Tracking] torch.compile + torch.distributed + set_grad_enabled/autocast,https://github.com/pytorch/pytorch/issues/112164,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:07:02Z +issue,112277,test_DistributedDataParallel fails with static_graph=True,https://github.com/pytorch/pytorch/issues/112277,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:57Z +issue,113114,Allow to construct distributed Work object from Future,https://github.com/pytorch/pytorch/issues/113114,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:53Z +issue,113376,Add validation for send/recv sizes,https://github.com/pytorch/pytorch/issues/113376,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:48Z +issue,114765,DistributedDataParallel _verify_param_shape_across_processes causes reserved memory on GPU,https://github.com/pytorch/pytorch/issues/114765,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:43Z +issue,114857,recursive search for DDP ignored parameters,https://github.com/pytorch/pytorch/issues/114857,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:38Z +issue,116183,Pipeline parallel doesn't seem to work with mixed precision,https://github.com/pytorch/pytorch/issues/116183,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:28Z +issue,117159,"If error is raised in distributed setting, top level exception formatting should include rank",https://github.com/pytorch/pytorch/issues/117159,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:23Z +issue,117755,[c10d] Enqueue work pointers instead of work copies,https://github.com/pytorch/pytorch/issues/117755,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:17Z +issue,118094,DTensorConverter::successful returns False on no Tensor args causing test failures,https://github.com/pytorch/pytorch/issues/118094,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:12Z +issue,119758,Flight Recorder dump_entries() segfaults when used with coalesced operations,https://github.com/pytorch/pytorch/issues/119758,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:06:06Z +issue,119856,Timeout for process termination upon receiving sigterm should be a user configurable parameter,https://github.com/pytorch/pytorch/issues/119856,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:56Z +issue,119890,dynamo remap collective improvement,https://github.com/pytorch/pytorch/issues/119890,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:51Z +issue,120516,[C10D] coalesced collectives are not well tested,https://github.com/pytorch/pytorch/issues/120516,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:41Z +issue,120787,Tensor Parallel Inference Performance Issue with Llama-7b-chat-hf Model caused by straggler problem,https://github.com/pytorch/pytorch/issues/120787,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:36Z +issue,121181,Profiling tool for DDP and FSDP,https://github.com/pytorch/pytorch/issues/121181,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: distributed_tool, module: fsdp, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:31Z +issue,121486,Kubernetes API server as Rendezvous backend,https://github.com/pytorch/pytorch/issues/121486,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:26Z +issue,121543,[DCP] test_state_dict.test_fsdp is failing on unrelated tests in CI.,https://github.com/pytorch/pytorch/issues/121543,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:20Z +issue,121587,Add clearer docs for `torch.distributed.` vs. `torch.distributed.nn.,https://github.com/pytorch/pytorch/issues/121587,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:16Z +issue,121605,Make the data type of local_used_map_ in the Reducer class configurable,https://github.com/pytorch/pytorch/issues/121605,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:05:11Z +issue,123815,my_role_info = role_infos group_rank index error,https://github.com/pytorch/pytorch/issues/123815,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:56Z +issue,124019,[FSDP+TP] RuntimeError: 'weight' must be 2-D,https://github.com/pytorch/pytorch/issues/124019,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:40Z +issue,124101,[C10D] Standardize torch distributed launcher interface,https://github.com/pytorch/pytorch/issues/124101,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:35Z +issue,124204,RPC + DDP VRAM overflow to rank 0 GPU.,https://github.com/pytorch/pytorch/issues/124204,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:31Z +issue,124260,[FSDP] Unreleased reserved memory after FSDP wrap,https://github.com/pytorch/pytorch/issues/124260,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:26Z +issue,124317,Many tests under test/distributed/elastic not running in OSS CI,https://github.com/pytorch/pytorch/issues/124317,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra, oncall: r2p","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:20Z +issue,124372,Pipeline Parallelism support for ZeroBubble,https://github.com/pytorch/pytorch/issues/124372,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:15Z +issue,124380,Error when running DDP examples to train between nodes,https://github.com/pytorch/pytorch/issues/124380,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: elastic, module: nccl, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:11Z +issue,124468,Race in ProcessGroupNCCL shutdown,https://github.com/pytorch/pytorch/issues/124468,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: nccl, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:04:05Z +issue,124680,Inconsistent results when training a model containing SyncBatchNorm with multiple GPUs,https://github.com/pytorch/pytorch/issues/124680,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:55Z +issue,126302,Make it easier to run distributed tests without run_test.py,https://github.com/pytorch/pytorch/issues/126302,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributed_tool, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:50Z +issue,131737,Intra-Node one-shot/two-shot low latency kernels for fp32,https://github.com/pytorch/pytorch/issues/131737,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:30Z +issue,131739,One Shot Low-Latency kernel is 2x higher than expected,https://github.com/pytorch/pytorch/issues/131739,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:25Z +issue,131754,Coredump for pytorch ddp when job finished,https://github.com/pytorch/pytorch/issues/131754,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:20Z +issue,135178,[DTensor] loss_parallel not worked with shifted logits and labels,https://github.com/pytorch/pytorch/issues/135178,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:14Z +issue,135511,RuntimeError: Modules with uninitialized parameters can't be used with `DistributedDataParallel` when using subclass of tensor.,https://github.com/pytorch/pytorch/issues/135511,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:10Z +issue,136122,[Distributed] Decouple ProcessGroupNCCL's regular exit from abort,https://github.com/pytorch/pytorch/issues/136122,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:05Z +issue,136187,Unable to Specify CUDA Stream for Collective Operations Using with torch.cuda.stream() context,https://github.com/pytorch/pytorch/issues/136187,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:03:00Z +issue,136226,[pipelining] try not to dry run module when creating PipelineStage?,https://github.com/pytorch/pytorch/issues/136226,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:55Z +issue,136248,Async NCCL communication blocks CUDA kernel in the first run,https://github.com/pytorch/pytorch/issues/136248,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:50Z +issue,136363,[pipelining] output shape validation / lazy inference,https://github.com/pytorch/pytorch/issues/136363,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:41Z +issue,136380,[pipelining] Improve recv buffer management,https://github.com/pytorch/pytorch/issues/136380,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:35Z +issue,136408,[RFC] Composable parallel documentation,https://github.com/pytorch/pytorch/issues/136408,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:29Z +issue,136425,Any updates on AsyncCollectiveTensor support for all-gather along non-zero dims?,https://github.com/pytorch/pytorch/issues/136425,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:25Z +issue,136567,[RFC] Offload collectives to NVSwitch when possible,https://github.com/pytorch/pytorch/issues/136567,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:20Z +issue,136866,fused_scaled_matmul_reduce_scatter report error with channel-wise scaling,https://github.com/pytorch/pytorch/issues/136866,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:15Z +issue,139280,Possible bug of tools::flight_recorder,https://github.com/pytorch/pytorch/issues/139280,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:02:09Z +issue,139581,Suggestion : enable one to adjust the interval between calls to the pytorch NCCL watchdog thread via env variable,https://github.com/pytorch/pytorch/issues/139581,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nccl, module: performance, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:59Z +issue,139681,Unexpected type cast from DTensor to Tensor when applying SequenceParallel on RMSNorm.,https://github.com/pytorch/pytorch/issues/139681,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:54Z +issue,139739,[Distributed][collectiveCoalesced] Why function(check_gpu_tensors_same_device) only works in allreduce_coalesced?,https://github.com/pytorch/pytorch/issues/139739,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:50Z +issue,140107,Support Joinable loss functions with DDP,https://github.com/pytorch/pytorch/issues/140107,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:45Z +issue,140227,"`Torch.distributed.tensor.parallel.Style.Rowwiseparallel` when send tensor, the use of the underlying operator is `aten.to.dtype_layout`, lack of distributed strategy. But after adding a policy, `AsyncCollectiveTensor` no placements error",https://github.com/pytorch/pytorch/issues/140227,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:40Z +issue,141237,test_fsdp_tp_integration fails for non-power-of 2 GPUs,https://github.com/pytorch/pytorch/issues/141237,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:30Z +issue,144239,Integrate ULFM and FTI for OpenMPI (torch.distributed),https://github.com/pytorch/pytorch/issues/144239,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:20Z +issue,144779,Adding Infiniband to RDZV Backend for optimal torch run training,https://github.com/pytorch/pytorch/issues/144779,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:11Z +issue,144851,torch.distributed hangs between Linux (X86) and Mac (M2 Pro),https://github.com/pytorch/pytorch/issues/144851,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:07Z +issue,144856,Connection Limitation in PyTorch Distributed (Vanilla) with c10d Rendezvous Backend,https://github.com/pytorch/pytorch/issues/144856,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:01:02Z +issue,145170,CUDA initialization error with vLLM 0.5.4 and PyTorch 2.4.0+cu121,https://github.com/pytorch/pytorch/issues/145170,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:57Z +issue,145376,distributed.new_group with backend GLOO hangs when distributed.split_group was called before,https://github.com/pytorch/pytorch/issues/145376,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:53Z +issue,146094,[CI][Distributed] @skip_if_lt_x_gpu(2) seems to be broken,https://github.com/pytorch/pytorch/issues/146094,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:48Z +issue,146326,Tracking issue for Improvements for DeepSeek,https://github.com/pytorch/pytorch/issues/146326,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: sdpa, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:43Z +issue,146329,DeepSeek: a2a communication with metadata on the GPU,https://github.com/pytorch/pytorch/issues/146329,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:38Z +issue,146332,DeepSeek: fine-grained overlap,https://github.com/pytorch/pytorch/issues/146332,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:28Z +issue,146565,use register_full_backward_hook with c10d in torch.compile and raise error,https://github.com/pytorch/pytorch/issues/146565,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:24Z +issue,146601,Enabling efficient model-level redistribution between FSPD-TP,https://github.com/pytorch/pytorch/issues/146601,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:19Z +issue,146813,Oneshot AllReduce not being triggered when there's nested intra- and inter-node process groups,https://github.com/pytorch/pytorch/issues/146813,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:14Z +issue,152042,distrubuted: false positive Grad strides vs Bucket strides warning,https://github.com/pytorch/pytorch/issues/152042,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T17:00:03Z +issue,152427,Add switch to disable truncation to long list print,https://github.com/pytorch/pytorch/issues/152427,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:58Z +issue,152822,TorchRun: Option to specify which GPUs to run on,https://github.com/pytorch/pytorch/issues/152822,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:53Z +issue,153016,inconsistent grads between two types of `allgather`s,https://github.com/pytorch/pytorch/issues/153016,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:49Z +issue,153122,[CUDA] test_c10d_nccl test_extra_cuda_context failure due to _helper_test_extra_cuda_context_by_memory,https://github.com/pytorch/pytorch/issues/153122,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:44Z +issue,153350,`distributed.checkpoint.state_dict.set_model_state_dict` returning incorrect `incompatible_keys` when `full_state_dict=True`,https://github.com/pytorch/pytorch/issues/153350,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:40Z +issue,153369,Rank {rank} has different values for {name}: {scalar_tensor_value}.,https://github.com/pytorch/pytorch/issues/153369,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:35Z +issue,153615,[DTensor] Backward Pass Failure with 2D Sharded DTensor After Mean Reduction,https://github.com/pytorch/pytorch/issues/153615,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:31Z +issue,153822,Passing string to devices in register_backend causes incorrect behavior,https://github.com/pytorch/pytorch/issues/153822,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:27Z +issue,154102,dist.init_process_group not work,https://github.com/pytorch/pytorch/issues/154102,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:22Z +issue,154116,Timeout in distributed training,https://github.com/pytorch/pytorch/issues/154116,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:18Z +issue,154297,Hangs and timeouts on dist.reduce_scatter on B200 GPU,https://github.com/pytorch/pytorch/issues/154297,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:14Z +issue,154327,`torch.distributed.checkpoint.state_dict.get_model_state_dict` does not update the state_dict._metadata key,https://github.com/pytorch/pytorch/issues/154327,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:09Z +issue,154345,[RFC] : Remove Explicit Backend References from `torch.distributed` (`c10d`),https://github.com/pytorch/pytorch/issues/154345,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:04Z +issue,154370,[C10D] Blackwell B200: NotImplementedError: Could not run '_c10d_functional_autograd::all_to_all_single' with arguments from the 'CUDA' backend.,https://github.com/pytorch/pytorch/issues/154370,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:59:00Z +issue,154563,Distributed Breakpoint doesn't exit safely,https://github.com/pytorch/pytorch/issues/154563,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:55Z +issue,154621,functional all_gather does not work with scalar tensor and fail with torch.compile,https://github.com/pytorch/pytorch/issues/154621,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:50Z +issue,154787,Redistribute DTensor across different DeviceMeshes,https://github.com/pytorch/pytorch/issues/154787,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: devicemesh, module: dtensor, oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:46Z +issue,155467,torch.distributed TCP init method binds socket to all interfaces in all cases,https://github.com/pytorch/pytorch/issues/155467,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:42Z +issue,155547,"[Misc] The keys of the sac_milp return dict were modified, but the test case was not updated",https://github.com/pytorch/pytorch/issues/155547,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: activation checkpointing, module: distributed_tool, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:37Z +issue,155664,[Misc] skip_if decorators aborts the test process in distributed/test_composability.py,https://github.com/pytorch/pytorch/issues/155664,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:33Z +issue,155892,[RFC] `DeviceGroup`: a mixin of `ProcessGroup` and `Backend`,https://github.com/pytorch/pytorch/issues/155892,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:29Z +issue,155916,[`torch.distributed`] Watchdog monitor error handling should match watchdog error handling,https://github.com/pytorch/pytorch/issues/155916,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:24Z +issue,155993,Keep gettting AssertionError: found no DeviceMesh from dtensor args for c10d.broadcast_.default!,https://github.com/pytorch/pytorch/issues/155993,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:20Z +issue,156002,DCP save only saves one shard of tensor parallel model when using DP + TP,https://github.com/pytorch/pytorch/issues/156002,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:16Z +issue,156130,[DDP][FSDP2] add unit test to showcase DDP mixed precision with FSDP2 mixed precision,https://github.com/pytorch/pytorch/issues/156130,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:12Z +issue,156392,DataParallel gather NaN across multiple gpus,https://github.com/pytorch/pytorch/issues/156392,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data parallel, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:08Z +issue,156627,`distributed.all_gather()` support for uneven tensors does not apply to all backends,https://github.com/pytorch/pytorch/issues/156627,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:58:03Z +issue,156842,FSDP2 + TP does not work,https://github.com/pytorch/pytorch/issues/156842,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:59Z +issue,156942,Can't use pinned_memory to do async save ckpt when using torch.distributed.checkpoint.async_save,https://github.com/pytorch/pytorch/issues/156942,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:55Z +issue,156966,Perf drop when running with FSDP and torch.compile,https://github.com/pytorch/pytorch/issues/156966,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:50Z +issue,157027,[RFC] Remove the FSDP data copy from compute stream critical path,https://github.com/pytorch/pytorch/issues/157027,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:46Z +issue,157318,Torch Elastic Wait timeout increase,https://github.com/pytorch/pytorch/issues/157318,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:41Z +issue,157395,[FSDP2] figure out the contract for mp_policy and tensor subclass extention,https://github.com/pytorch/pytorch/issues/157395,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:37Z +issue,157781,torch.distributed.checkpoint.state_dict.set_state_dict stucks with StateDictOptions(full_state_dict=True),https://github.com/pytorch/pytorch/issues/157781,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:33Z +issue,157862,"Spurious ""Grad strides do not match bucket view strides"" warning for 1x1 convolution",https://github.com/pytorch/pytorch/issues/157862,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:28Z +issue,158016,"dist.all_to_all_single, when input tensors of different shapes result in undefined output behavior",https://github.com/pytorch/pytorch/issues/158016,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:24Z +issue,158443,DISABLED test_gradients_synchronizations (__main__.TensorPipeTensorPipeCudaDistAutogradTest),https://github.com/pytorch/pytorch/issues/158443,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:19Z +issue,158698,[fsdp2] Possible to fuse reduce-scatter (in fp32) + div + dtype conversion (to bf16) in post-backward?,https://github.com/pytorch/pytorch/issues/158698,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:15Z +issue,158775,Add docs for torchrun numa binding on website,https://github.com/pytorch/pytorch/issues/158775,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:57:06Z +issue,159563,About init_process_group with different backends,https://github.com/pytorch/pytorch/issues/159563,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:57Z +issue,159686,"c10d broadcast copies at::Tensor, creates footgun for downstream implementations",https://github.com/pytorch/pytorch/issues/159686,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:52Z +issue,159711,Support dict as input/output for pipeline parallelism,https://github.com/pytorch/pytorch/issues/159711,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:48Z +issue,160006,Support NUMA Binding for Callable entrypoints to `elastic_launch`,https://github.com/pytorch/pytorch/issues/160006,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:44Z +issue,160130,Torchrun breaks with virtual environments,https://github.com/pytorch/pytorch/issues/160130,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: elastic, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:39Z +issue,160731,Cannot initialise process group on CPU device with GLOO backend,https://github.com/pytorch/pytorch/issues/160731,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:35Z +issue,160732,GLOO backend should have MPS capability,https://github.com/pytorch/pytorch/issues/160732,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:32Z +issue,160950,fx.node._sort_key dones't work right with torch.distributed._functional_collectives,https://github.com/pytorch/pytorch/issues/160950,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:26Z +issue,161091,DTensor view op results in RuntimeError: shape '[38597376]' is invalid for input of size 38598144,https://github.com/pytorch/pytorch/issues/161091,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:22Z +issue,161153,torch.compile fails in FSDP due to .data assignment,https://github.com/pytorch/pytorch/issues/161153,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:18Z +issue,161368,Incorrect offsets for 2D uneven sharding in DTensor in `compute_local_shape_and_global_offsetproduces`,https://github.com/pytorch/pytorch/issues/161368,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:14Z +issue,161776,FSDP&device_mesh Mesh_dim_names in init_device_mesh not work,https://github.com/pytorch/pytorch/issues/161776,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: devicemesh, module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:09Z +issue,162032,torch thread yield after launch nccl kernel,https://github.com/pytorch/pytorch/issues/162032,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:05Z +issue,162160,[BUG] Experts weight not release after backward when FSDP2 + moe expert parallel,https://github.com/pytorch/pytorch/issues/162160,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:56:00Z +issue,162179,test_fully_shard_logging.py is not being run,https://github.com/pytorch/pytorch/issues/162179,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, module: tests, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:56Z +issue,162237,Does it support new CPU group after passing `device_id` to `init_process_group`,https://github.com/pytorch/pytorch/issues/162237,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:51Z +issue,162290,local_output_for_attn: bool = False ( for Col/Row-wiseParallel ),https://github.com/pytorch/pytorch/issues/162290,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:47Z +issue,162638,Gradient Clipping in Pipeline Parallelism Schedules,https://github.com/pytorch/pytorch/issues/162638,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:42Z +issue,162668,ReduceOp has incorrect py::class_ binding,https://github.com/pytorch/pytorch/issues/162668,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:37Z +issue,162741,`torch.distributed.pipelining` Llama 2 and 3 Results Torch Dynamo Compiled Error,https://github.com/pytorch/pytorch/issues/162741,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:32Z +issue,162745,[CI][CUDA][Distributed] python test/distributed/test_data_parallel.py TestDataParallel.test_data_parallel_rnn p1.allclose(p2) False,https://github.com/pytorch/pytorch/issues/162745,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data parallel, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:28Z +issue,162748,[CI][CUDA][Distributed] test_3_level_hierarchical_model_averager Works on 4-GPU runner but requires explicit WORLD_SIZE=4 on 8-GPU runner,https://github.com/pytorch/pytorch/issues/162748,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:23Z +issue,162755,"Ideally, we should not use `require_exact_world_size(4)` if we only have 8-GPU runners",https://github.com/pytorch/pytorch/issues/162755,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributed_tool, module: tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:19Z +issue,162871,[CI][CUDA][Distributed] TestDistBackendWithSpawn.test_gather_object Requires Exact World Size of 4,https://github.com/pytorch/pytorch/issues/162871,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:14Z +issue,162897,[CI][CUDA][Distributed][NVSHMEM][Triton] NVSHMEMTritonTest.test_triton_fence KeyError,https://github.com/pytorch/pytorch/issues/162897,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:10Z +issue,162917,[CI][CUDA][Distributed][Symmetric Memory] AsyncTPTest.test_fused_all_gather_matmul_native_symm_mem_input_False_is_b_row_major_False Failure,https://github.com/pytorch/pytorch/issues/162917,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: symm_mem, module: tests, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:05Z +issue,162940,[CI][CUDA][Distributed] test_fused_scaled_matmul_reduce_scatter_scatter_dim_0_rowwise_False fails due to accuracy mismatch,https://github.com/pytorch/pytorch/issues/162940,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:55:00Z +issue,163043,`parallelize_module` causing scatter,https://github.com/pytorch/pytorch/issues/163043,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:55Z +issue,163546,Better nccl timeout handling,https://github.com/pytorch/pytorch/issues/163546,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:47Z +issue,163678,RuntimeError: NCCL error: unhandled cuda error,https://github.com/pytorch/pytorch/issues/163678,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:42Z +issue,163722,AdaptiveLogSoftmaxWithLoss with weight tying,https://github.com/pytorch/pytorch/issues/163722,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:38Z +issue,163913,test_triton_wait_until hangs,https://github.com/pytorch/pytorch/issues/163913,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:34Z +issue,164015,[CI][CUDA] test_multimem_one_shot_all_reduce_bfloat16_align_bytes_1 failures on B200 runner,https://github.com/pytorch/pytorch/issues/164015,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: symm_mem, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:29Z +issue,165653,[distributed] test_symmetric_memory fails on odd-numbered GPU configurations (3 GPUs),https://github.com/pytorch/pytorch/issues/165653,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:25Z +issue,165933,[Distributed] fully_shard: support no_shard (ddp) strategy?,https://github.com/pytorch/pytorch/issues/165933,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:16Z +issue,166523,DDP + MPI-based distributed backend: missing work/future completion assertion for Broadcast operations,https://github.com/pytorch/pytorch/issues/166523,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ddp, module: mpi, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:54:02Z +issue,166930,[DTensor] DTensor randn inconsistent with single device behavior,https://github.com/pytorch/pytorch/issues/166930,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:58Z +issue,167036,FSDP2 moves my inputs to device when I don't want it to,https://github.com/pytorch/pytorch/issues/167036,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:53Z +issue,167106,DTensor operations incur unnecessary gathering and memory consumption resulting therefrom,https://github.com/pytorch/pytorch/issues/167106,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: performance, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:49Z +issue,167288,"when use torchrun and set restarts for elastic run, cause NCCL INFO socketPollConnect: connect returned Connection refused",https://github.com/pytorch/pytorch/issues/167288,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:44Z +issue,167305,Pytorch distributed plugin backend _set_default_backend error when use multi-plugin backends,https://github.com/pytorch/pytorch/issues/167305,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:39Z +issue,167353,`use_mem_pool` is not active during backward,https://github.com/pytorch/pytorch/issues/167353,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:35Z +issue,167425,confusion on pipeline parallelism implementation,https://github.com/pytorch/pytorch/issues/167425,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:30Z +issue,167775,Graceful Ctrl+C handling from torchrun,https://github.com/pytorch/pytorch/issues/167775,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: devx, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:26Z +issue,167832,PG splitting does not consider new groups with a different backend,https://github.com/pytorch/pytorch/issues/167832,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:21Z +issue,167994,CI Not Detecting Failing Tests in test/distributed/elastic/*,https://github.com/pytorch/pytorch/issues/167994,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:16Z +issue,168143,FSDP2 does not work with custom Tensor wrapper objects,https://github.com/pytorch/pytorch/issues/168143,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:11Z +issue,169275,GLOO distribute train can not find each ranks,https://github.com/pytorch/pytorch/issues/169275,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:07Z +issue,169752,"if __name__ == ""__main__"": run_tests() cannot be found in the test files under distributed/elastic/rendezvous",https://github.com/pytorch/pytorch/issues/169752,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: tests, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:53:02Z +issue,169841,ProcessGroup hash doesn't incorporate pg_options,https://github.com/pytorch/pytorch/issues/169841,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:58Z +issue,170290,"A lot of ""Failed to check the ""should dump"" flag on TCPStore"" output in ProcessGroupNCCL when rank 0 exit first",https://github.com/pytorch/pytorch/issues/170290,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:49Z +issue,171412,"[Feature Request][DCP] Add ""Broadcast Load"" arg to dcp.load for HSDP storage optimization",https://github.com/pytorch/pytorch/issues/171412,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:44Z +issue,171495,[internode] Symmetric mem rendezvous erro,https://github.com/pytorch/pytorch/issues/171495,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: symm_mem, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:39Z +issue,171604,NotImplementedError when using AsyncCollectiveTensor with torch.func.jvp,https://github.com/pytorch/pytorch/issues/171604,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functorch, oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:34Z +issue,171656,torch.distributed.pipelining fails on models having DynamicCache (esp. Llama),https://github.com/pytorch/pytorch/issues/171656,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:30Z +issue,173164,allgather throw exception with discontigous input tensor,https://github.com/pytorch/pytorch/issues/173164,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:25Z +issue,173362,"`dist.reduce_scatter` throws `ValueError: Tensors must be contiguous` when the output tensor is non-contiguous (e.g., from `permute()` or `transpose()`)",https://github.com/pytorch/pytorch/issues/173362,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:20Z +issue,173608,The program is stuck at function torch.distributed.new_group when parameter use_local_synchronization is True and len(_world.pg_names) is different between NPUs,https://github.com/pytorch/pytorch/issues/173608,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:16Z +issue,173789,[Devicemesh] Bug: `child_to_root_mapping` overwritten due to hash collision in DeviceMesh submeshes,https://github.com/pytorch/pytorch/issues/173789,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: distributed infra","decide whether this is bug, use case, docs, or support",2026-06-05T16:52:11Z +issue,184394,[inductor] Merge SymPy printer path into typed ops/CSE path for Triton codegen,https://github.com/pytorch/pytorch/issues/184394,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-05T16:09:18Z +issue,24422,Label tracking meta-issue (edit me to get automatically CC'ed on issues! cc bot),https://github.com/pytorch/pytorch/issues/24422,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-06-05T09:49:01Z +issue,170562,PadMMTest.test_original_aten_preserved_pad_mm fails,https://github.com/pytorch/pytorch/issues/170562,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-04T20:08:49Z +issue,186259,DISABLED test_redispatch_nn_functional_linear_cross_entropy_chunked_cuda_float16 (__main__.TestTorchFunctionRedispatchOpsCUDA),https://github.com/pytorch/pytorch/issues/186259,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_function__, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-06-04T19:28:03Z +issue,145189,DISABLED test_slice_scatter_reinplace_cuda (__main__.GPUTests),https://github.com/pytorch/pytorch/issues/145189,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-04T13:42:52Z +issue,137560,vmap specializing on dynamic shapes with index_put_,https://github.com/pytorch/pytorch/issues/137560,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-04T09:45:26Z +issue,186213,CUDA expandable IPC receiver reserves ~9/8 of total GPU memory VA per imported handle and can exhaust device virtual address space,https://github.com/pytorch/pytorch/issues/186213,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cudacachingallocator, module: multiprocessing","decide whether this is bug, use case, docs, or support",2026-06-04T09:19:34Z +issue,186203,DISABLED test_redispatch_nn_functional_linear_cross_entropy_cuda_float16 (__main__.TestTorchFunctionRedispatchOpsCUDA),https://github.com/pytorch/pytorch/issues/186203,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_function__, module: flaky-tests","decide whether this is bug, use case, docs, or support",2026-06-04T08:08:13Z +issue,166628,[WinError 1114] A dynamic link library (DLL) initialization routine failed,https://github.com/pytorch/pytorch/issues/166628,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: windows","decide whether this is bug, use case, docs, or support",2026-06-04T07:28:44Z +issue,181851,register_fake to throw error if signature doesn't match op's schema,https://github.com/pytorch/pytorch/issues/181851,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-03T17:47:41Z +issue,186099,Triton linux nightly wheels fail to install on Python 3.15 (requires-python: <3.15),https://github.com/pytorch/pytorch/issues/186099,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: ci, module: python version","decide whether this is bug, use case, docs, or support",2026-06-03T16:56:20Z +issue,174373,Increase NUM_THREADS that embedded ARM OpenBLAS is compiled with,https://github.com/pytorch/pytorch/issues/174373,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: binaries, module: build, module: openblas","decide whether this is bug, use case, docs, or support",2026-06-03T13:38:07Z +issue,181663,torch.softmax is not a true alias for torch.nn.functional.softmax,https://github.com/pytorch/pytorch/issues/181663,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: ux","decide whether this is bug, use case, docs, or support",2026-06-03T09:00:52Z +issue,168652,Test: TestSelectAlgorithm.test_mm_dropout,https://github.com/pytorch/pytorch/issues/168652,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-02T23:20:08Z +issue,177259,Migrate device-agnostic skips from `op_db` to `@skipOps`,https://github.com/pytorch/pytorch/issues/177259,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: tests,"decide whether this is bug, use case, docs, or support",2026-06-02T16:51:31Z +issue,185746,CUDNN version issues in 2.13.0.dev20260530+cu132 nightly wheel,https://github.com/pytorch/pytorch/issues/185746,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: cudnn, module: windows","decide whether this is bug, use case, docs, or support",2026-06-02T15:29:29Z +issue,150121,torch.compile on MPS progress tracker,https://github.com/pytorch/pytorch/issues/150121,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: mps,"decide whether this is bug, use case, docs, or support",2026-06-02T11:05:43Z +issue,185198,[xpu] Intel GPU telemetry management APIs may not work on Arc A-Series and data center GPUs,https://github.com/pytorch/pytorch/issues/185198,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xpu,"decide whether this is bug, use case, docs, or support",2026-06-02T10:13:37Z +issue,160472,PT2 Cache <> CustomOp issue,https://github.com/pytorch/pytorch/issues/160472,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-01T19:27:24Z +issue,185684,🐛 [Bug] Fake Tensor Updater Doesn't Exist on Windows Platform,https://github.com/pytorch/pytorch/issues/185684,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: inductor, module: pt2-dispatcher, module: windows","decide whether this is bug, use case, docs, or support",2026-06-01T16:59:36Z +issue,175252,Slow reading tensor meta-data,https://github.com/pytorch/pytorch/issues/175252,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, module: serialization","decide whether this is bug, use case, docs, or support",2026-06-01T14:54:10Z +issue,170487,Support `.requires_grad_(True)` in `torch.compile(fullgraph = True)`,https://github.com/pytorch/pytorch/issues/170487,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-01T04:01:19Z +issue,181006,[inductor] fails at caching stage with UnboundLocalError (bufXX referenced before assignment) when compiling torch.func.grad with MultiheadAttention + LayerNorm,https://github.com/pytorch/pytorch/issues/181006,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-06-01T03:59:09Z +issue,181653,torch.compile(torch.func.grad(fn)) hangs indefinitely on models with pooling ops — sympy.simplify exponential recursion in stride_at,https://github.com/pytorch/pytorch/issues/181653,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: functorch, module: inductor, module: pooling","decide whether this is bug, use case, docs, or support",2026-06-01T03:58:28Z +issue,175636,`[feature request][discussion][linalg]` Polar decomposition,https://github.com/pytorch/pytorch/issues/175636,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-06-01T01:46:50Z +issue,51002,Consider adding context manager support for torch.set_printoptions,https://github.com/pytorch/pytorch/issues/51002,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: printing,"decide whether this is bug, use case, docs, or support",2026-05-31T21:30:34Z +issue,185698,CUDA extension build failure on glibc 2.38+ and GCC 14+ due to math header noexcept mismatch - proposed clean fix,https://github.com/pytorch/pytorch/issues/185698,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cpp-extensions, module: cuda, module: dependency bug","decide whether this is bug, use case, docs, or support",2026-05-30T04:58:00Z +issue,152028,[CUDA] MultiheadAttention with masks and dropout produces NaNs,https://github.com/pytorch/pytorch/issues/152028,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda, module: sdpa","decide whether this is bug, use case, docs, or support",2026-05-30T03:02:26Z +issue,159151,DISABLED test_fake_crossref_backward_no_amp_nn_functional_bilinear_cuda_float32 (__main__.TestFakeTensorCUDA),https://github.com/pytorch/pytorch/issues/159151,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: unknown, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-29T13:43:53Z +issue,159150,DISABLED test_fake_crossref_backward_amp_nn_functional_bilinear_cuda_float32 (__main__.TestFakeTensorCUDA),https://github.com/pytorch/pytorch/issues/159150,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: flaky-tests, module: pt2-dispatcher, module: unknown","decide whether this is bug, use case, docs, or support",2026-05-29T07:59:21Z +issue,185590,[Test] PyTorch Test Case Refactoring and Track,https://github.com/pytorch/pytorch/issues/185590,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: tests,"decide whether this is bug, use case, docs, or support",2026-05-29T03:39:57Z +issue,74415,"Warning: ""Specified kernel cache directory could not be created""",https://github.com/pytorch/pytorch/issues/74415,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: jiterator,"decide whether this is bug, use case, docs, or support",2026-05-28T22:49:23Z +issue,184361,[pipelining] PP fails when trying to pipeline any HF transformers model that uses RoPE,https://github.com/pytorch/pytorch/issues/184361,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pipelining, oncall: distributed, oncall: distributed parallelisms","decide whether this is bug, use case, docs, or support",2026-05-28T15:51:07Z +issue,155240,Dynamo Fails on torch_scatter.scatter_max with Fake Tensor Allocation Error During Graph Tracing,https://github.com/pytorch/pytorch/issues/155240,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-28T07:28:14Z +issue,179951,DISABLED test_scaled_mm (__main__.DistMatrixOpsTestWithLocalTensor),https://github.com/pytorch/pytorch/issues/179951,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rocm, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-05-28T06:30:50Z +issue,158186,Flash Attention 2 + Dynamo + FSDP accelerate plugin + torch.compile error,https://github.com/pytorch/pytorch/issues/158186,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, module: sdpa, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-27T19:29:07Z +issue,184903,[Inductor] Codegen Gluon,https://github.com/pytorch/pytorch/issues/184903,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-27T17:48:43Z +issue,185215,PyTorch with XPU support fails to build on Windows in Debug/RelWithDebInfo configurations,https://github.com/pytorch/pytorch/issues/185215,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: windows, module: xpu","decide whether this is bug, use case, docs, or support",2026-05-27T06:22:37Z +issue,131009,[cudagraph] an issue related to dynamic shapes for cudagraph backend with sd2.1,https://github.com/pytorch/pytorch/issues/131009,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-27T02:49:26Z +issue,185135,[inductor] Scheduler phases scale poorly on lowered aten.lstm/aten.gru graphs,https://github.com/pytorch/pytorch/issues/185135,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile-time, module: inductor, module: performance, module: rnn","decide whether this is bug, use case, docs, or support",2026-05-26T17:50:58Z +issue,181093,[RFC] Add Minimal torch.compile Backend to Validate PrivateUse1 Integration,https://github.com/pytorch/pytorch/issues/181093,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: backend, module: dynamo, module: openreg, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-05-26T14:43:04Z +issue,173369,Exporting ONNX model with captum heatmaps generation,https://github.com/pytorch/pytorch/issues/173369,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: onnx, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-26T14:13:37Z +issue,183758,[Testing] Expand Testing for Non-Contiguous Tensor Ops Across All Platforms,https://github.com/pytorch/pytorch/issues/183758,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: memory format, module: testing","decide whether this is bug, use case, docs, or support",2026-05-26T13:25:38Z +issue,174469,[RFC] Overview of Test Case Refactoring in PyTorch,https://github.com/pytorch/pytorch/issues/174469,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: backend, module: privateuse1, module: tests","decide whether this is bug, use case, docs, or support",2026-05-26T07:54:29Z +issue,183060,PyTorch pytree equality comparison is too strict.,https://github.com/pytorch/pytorch/issues/183060,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: pytree,"decide whether this is bug, use case, docs, or support",2026-05-26T07:26:01Z +issue,162728,Significant Numerical Inconsistencies After torch.compile() for Local Attention Model,https://github.com/pytorch/pytorch/issues/162728,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: pt2,"decide whether this is bug, use case, docs, or support",2026-05-25T11:33:21Z +issue,180206,DISABLED test_opcheck_opinfo_NumpyMulScalarCustomOp_cpu_float32 (__main__.TestCustomOpTestingCPU),https://github.com/pytorch/pytorch/issues/180206,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: flaky-tests, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-25T08:07:11Z +issue,180437,DISABLED test_find_learning_rate_ensembling_FeatureAlphaDropout_mechanism_functional_call_cpu (__main__.TestExamplesCorrectnessCPU),https://github.com/pytorch/pytorch/issues/180437,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-25T08:07:02Z +issue,112827,Multi Scale Deformable Attention Support,https://github.com/pytorch/pytorch/issues/112827,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: sdpa","decide whether this is bug, use case, docs, or support",2026-05-24T06:13:21Z +issue,14152,Failed to run 'bash ../tools/build_pytorch_libs.sh --use-cuda --use-nnpack --use-mkldnn --use-qnnpack caffe2',https://github.com/pytorch/pytorch/issues/14152,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-05-24T01:04:50Z +issue,80821,Add typing support to ModuleList and ModuleDict,https://github.com/pytorch/pytorch/issues/80821,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: typing,"decide whether this is bug, use case, docs, or support",2026-05-23T22:54:20Z +issue,150168,[ROCm] PyTorch slow on TTS,https://github.com/pytorch/pytorch/issues/150168,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: rocm,"decide whether this is bug, use case, docs, or support",2026-05-23T12:46:25Z +issue,88096,Add nondeterministic alert to `torch.Tensor.scatter()`,https://github.com/pytorch/pytorch/issues/88096,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: determinism,"decide whether this is bug, use case, docs, or support",2026-05-22T21:35:21Z +issue,184769,Optimize Metal Kernels with sincos,https://github.com/pytorch/pytorch/issues/184769,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: complex, module: mps, module: performance","decide whether this is bug, use case, docs, or support",2026-05-22T19:38:33Z +issue,178134,torch.compile (CUDA fp32) vs eager gives large max relative error for Conv2d,https://github.com/pytorch/pytorch/issues/178134,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: convolution, module: cuda, module: inductor, module: pt2 accuracy","decide whether this is bug, use case, docs, or support",2026-05-22T09:58:07Z +issue,55777,torch.load is very slow with gzip.open,https://github.com/pytorch/pytorch/issues/55777,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: serialization,"decide whether this is bug, use case, docs, or support",2026-05-22T05:22:09Z +issue,170636,[xpu] No XPU implementation of multiprocessing tensor reduction.,https://github.com/pytorch/pytorch/issues/170636,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xpu,"decide whether this is bug, use case, docs, or support",2026-05-22T01:54:20Z +issue,177277,Conv3D does not preserve dim_order for some inputs,https://github.com/pytorch/pytorch/issues/177277,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: convolution, module: faketensor, module: memory format, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-05-21T17:31:43Z +issue,103222,compilation fails `error: invalid argument '-std=c++17' not allowed with 'C'`,https://github.com/pytorch/pytorch/issues/103222,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-05-21T13:20:38Z +issue,160333,Tensor subclasses don't work with torch.library.triton_op,https://github.com/pytorch/pytorch/issues/160333,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-21T09:44:28Z +issue,164057,The device (Intel(R) Arc(TM) A380 Graphics) doesn't support querying the available free memory,https://github.com/pytorch/pytorch/issues/164057,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xpu,"decide whether this is bug, use case, docs, or support",2026-05-21T08:49:54Z +issue,149953,`RuntimeError: UR error` with XPU,https://github.com/pytorch/pytorch/issues/149953,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: xpu","decide whether this is bug, use case, docs, or support",2026-05-21T08:43:11Z +issue,87961,crash in `torch.package.PackageExporter`,https://github.com/pytorch/pytorch/issues/87961,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: edge cases,"decide whether this is bug, use case, docs, or support",2026-05-20T18:32:06Z +issue,87556,The behavior of cast `NaN` is different on cpu and cuda,https://github.com/pytorch/pytorch/issues/87556,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: nans and infs","decide whether this is bug, use case, docs, or support",2026-05-20T18:31:59Z +issue,87172,Unstable results in sin/arcsin/arccos calls,https://github.com/pytorch/pytorch/issues/87172,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: numerical-stability","decide whether this is bug, use case, docs, or support",2026-05-20T18:31:57Z +issue,83494,`torch.pinverse` produces wrong output!,https://github.com/pytorch/pytorch/issues/83494,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-05-20T18:31:55Z +issue,165501,Inductor custom pass mechanism that maintains config constraints,https://github.com/pytorch/pytorch/issues/165501,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: vllm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-20T01:21:01Z +issue,140970,`torch._inductor.cpu_vec_isa.pick_vec_isa` takes ~9 seconds to run,https://github.com/pytorch/pytorch/issues/140970,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: inductor, oncall: cpu inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-20T01:14:28Z +issue,182490,[Docathon] Convert `notes/hip.rst` from rST to MyST Markdown,https://github.com/pytorch/pytorch/issues/182490,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-05-19T20:17:09Z +issue,161609,Bucketize lowering crash when using a sliced computed tensor as boundaries,https://github.com/pytorch/pytorch/issues/161609,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T19:21:49Z +issue,168042,TorchInductor compile breaks when called in a multithreaded context with GIL off,https://github.com/pytorch/pytorch/issues/168042,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T18:05:58Z +issue,138980,compile_fx inplace modifly the input graph,https://github.com/pytorch/pytorch/issues/138980,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: vllm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T18:04:25Z +issue,106006,[torch.compile] assertion sometimes ignored with inductor backend,https://github.com/pytorch/pytorch/issues/106006,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T18:03:57Z +issue,184397,[ROCm][Windows] Track proper fix for HIP per-config flag injection bypass (follow-up to #183856),https://github.com/pytorch/pytorch/issues/184397,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: rocm, module: windows","decide whether this is bug, use case, docs, or support",2026-05-19T16:21:10Z +issue,179143,[RFC] AdamTR: Adam variant for deterministic Token-Routed architectures,https://github.com/pytorch/pytorch/issues/179143,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: optimizer,"decide whether this is bug, use case, docs, or support",2026-05-19T15:36:22Z +issue,130015,Improve the Inductor generated kernel for the pattern of `output1 = pointwise(intput); output2 = transpose(output1)`,https://github.com/pytorch/pytorch/issues/130015,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T15:25:17Z +issue,143337,An idea to improve Welford implementation in Inductor,https://github.com/pytorch/pytorch/issues/143337,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: pt2,"decide whether this is bug, use case, docs, or support",2026-05-19T15:24:27Z +issue,165785,Harden CuTeDSL Inductor Path,https://github.com/pytorch/pytorch/issues/165785,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T15:21:31Z +issue,177578,[Inductor][NVGEMM] Extend nvMatmulHeuristics config key tuple for stages/split_k,https://github.com/pytorch/pytorch/issues/177578,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T15:20:15Z +issue,179652,`torch.library.custom.register_autograd` incompatible with `jacrev`,https://github.com/pytorch/pytorch/issues/179652,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: functorch","decide whether this is bug, use case, docs, or support",2026-05-19T08:32:30Z +issue,130039,`min_cut_rematerialization_partition` is not optimal memory usage.,https://github.com/pytorch/pytorch/issues/130039,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-19T05:09:16Z +issue,145401,torch.compile has different numerics for var_mean,https://github.com/pytorch/pytorch/issues/145401,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T21:43:30Z +issue,144362,Some operators miss dtype check when using `torch.compile`,https://github.com/pytorch/pytorch/issues/144362,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: error checking, module: inductor, module: structured kernels, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T21:43:25Z +issue,120902,RuntimeError: cuFFT error: CUFFT_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/120902,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: fft, module: sparse","decide whether this is bug, use case, docs, or support",2026-05-18T21:00:04Z +issue,183391,"torch.fmin with CUDA out=int16 casts inputs before computation, producing incorrect result",https://github.com/pytorch/pytorch/issues/183391,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: edge cases, module: type promotion","decide whether this is bug, use case, docs, or support",2026-05-18T17:33:01Z +issue,113287,Provide a way to AOT torch.compile and serialize a model,https://github.com/pytorch/pytorch/issues/113287,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile-time, module: dynamo, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T16:11:18Z +issue,130614,"Significant Memory Increase with Conv2d using padding_mode 'replicate', 'reflect', or 'circular'",https://github.com/pytorch/pytorch/issues/130614,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: convolution, module: memory usage, module: nn","decide whether this is bug, use case, docs, or support",2026-05-18T14:26:35Z +issue,176611,[cudagraph][torch.compile] `cudagraph_or_error` can be silently ignored even when the cudagraph is not captured.,https://github.com/pytorch/pytorch/issues/176611,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: error checking, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T06:40:31Z +issue,170762,[BUG][Inconsistency] aten.cudnn_batch_norm_backward produces numerical mismatch between eager and inductor,https://github.com/pytorch/pytorch/issues/170762,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cudnn, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T06:39:45Z +issue,165280,[inductor] avoid triton kernel when creating zero tensor,https://github.com/pytorch/pytorch/issues/165280,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T06:39:28Z +issue,163932,[inductor] a potential numerical instability issue inside a softmax model.,https://github.com/pytorch/pytorch/issues/163932,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: nans and infs, module: numerical-stability, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T06:39:23Z +issue,161842,Exhaustive Autotuning causes `cudaErrorIllegalAddress`,https://github.com/pytorch/pytorch/issues/161842,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T06:39:15Z +issue,159413,Inductor Assertion Error with extremely large input,https://github.com/pytorch/pytorch/issues/159413,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-18T06:39:00Z +issue,134606,VMAP over GRU: Batching rule not implemented for aten::gru.input,https://github.com/pytorch/pytorch/issues/134606,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functorch, module: vmap","decide whether this is bug, use case, docs, or support",2026-05-17T15:15:34Z +issue,181094,Optimization: pin memory management optimization based on pre-allocation and buddy system,https://github.com/pytorch/pytorch/issues/181094,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: performance","decide whether this is bug, use case, docs, or support",2026-05-17T08:22:19Z +issue,87131,torch.clamp does not clamp out of -0 from 0 when ran on the CPU,https://github.com/pytorch/pytorch/issues/87131,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: numpy","decide whether this is bug, use case, docs, or support",2026-05-17T03:28:49Z +issue,183768,torch.compile returns inf for torch.acosh on large finite float32 inputs while eager returns finite values,https://github.com/pytorch/pytorch/issues/183768,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-17T02:51:31Z +issue,183607,torch.compile changes fp16 overflow behavior for cast-to-float16 multiplication,https://github.com/pytorch/pytorch/issues/183607,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: half, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-17T02:51:09Z +issue,180811,"[AOTInductor] max_autotune non-deterministically selects cuBLAS over fused Triton kernels for small GEMM, causing significant performance regression",https://github.com/pytorch/pytorch/issues/180811,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, module: cuda, module: determinism, module: inductor","decide whether this is bug, use case, docs, or support",2026-05-17T02:48:31Z +issue,155679,Torch compile CUDA graphs leads to a large number of CUDA streams,https://github.com/pytorch/pytorch/issues/155679,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: dependency bug, module: inductor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-05-16T01:05:57Z +issue,182272,[Windows] RTX 5070 Ti (Blackwell sm_120) - cuInit error 100 on hybrid graphics laptops,https://github.com/pytorch/pytorch/issues/182272,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: cuda, module: docs, module: windows","decide whether this is bug, use case, docs, or support",2026-05-15T22:40:31Z +issue,177719,[Discussion] Disable size_asserts by default in inductor for vllm x compile inference,https://github.com/pytorch/pytorch/issues/177719,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: inductor, module: performance, module: vllm","decide whether this is bug, use case, docs, or support",2026-05-15T22:16:23Z +issue,37410,Reset a `torch.optim.Optimizer`,https://github.com/pytorch/pytorch/issues/37410,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: optimizer,"decide whether this is bug, use case, docs, or support",2026-05-15T07:08:39Z +issue,152270,Question about that support of torch.compile for a custom CUDA operator?,https://github.com/pytorch/pytorch/issues/152270,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: docs, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-15T03:38:02Z +issue,91469,torch.compile with aotautograd does not support double backwards,https://github.com/pytorch/pytorch/issues/91469,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: autograd, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-15T03:37:00Z +issue,125977,[compile time] AOT Autograd is taking long time in tracing,https://github.com/pytorch/pytorch/issues/125977,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-15T03:07:08Z +issue,150877,DISABLED test_multiple_module (__main__.TestInvokeSubgraphExportStrict),https://github.com/pytorch/pytorch/issues/150877,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-15T03:06:40Z +issue,127445,PyTorch version agnostic C++ extensions,https://github.com/pytorch/pytorch/issues/127445,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpp-extensions, module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-15T03:06:32Z +issue,150062,Support HOPs in fx_graph_runnable,https://github.com/pytorch/pytorch/issues/150062,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-15T02:04:05Z +issue,157456,Deprecation of CUTLASS Python interface,https://github.com/pytorch/pytorch/issues/157456,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-05-14T23:37:56Z +issue,139257,Error compiling the torch.library.custom_op with input mutations with set_,https://github.com/pytorch/pytorch/issues/139257,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T22:34:17Z +issue,136446,Allow Inductor to Compose with FakeTensorMode to Estimate Memory Usage,https://github.com/pytorch/pytorch/issues/136446,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: inductor, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T22:21:42Z +issue,133721,[HOP] Add BC tests for hop in export,https://github.com/pytorch/pytorch/issues/133721,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T22:21:42Z +issue,183795,Clean up XNNPACK_NO_CODE_CACHE,https://github.com/pytorch/pytorch/issues/183795,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xnnpack,"decide whether this is bug, use case, docs, or support",2026-05-14T21:54:54Z +issue,131974,Missing grad_fn information while torch.compile with customized gradient function,https://github.com/pytorch/pytorch/issues/131974,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T19:05:37Z +issue,129430,OperatorBase.py_impl should support with_keyset=True extra argument.,https://github.com/pytorch/pytorch/issues/129430,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: pt2-dispatcher, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T19:05:16Z +issue,113089,Add docs for __tensor_flatten__ / __tensor_unflatten__,https://github.com/pytorch/pytorch/issues/113089,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_dispatch__, module: docs, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T17:41:45Z +issue,142388,User-defined way to hook into Inductor autotuning (+ custom op),https://github.com/pytorch/pytorch/issues/142388,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: inductor, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T17:41:42Z +issue,145522,use_const_ref_for_mutable_tensors doesn't work with out= overloads,https://github.com/pytorch/pytorch/issues/145522,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: pt2-dispatcher,"decide whether this is bug, use case, docs, or support",2026-05-14T15:35:50Z +issue,121837,Revive libtorch-shared-without-deps flavor,https://github.com/pytorch/pytorch/issues/121837,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: abi, module: binaries, module: cpp","decide whether this is bug, use case, docs, or support",2026-05-14T10:01:18Z +issue,183686,[triton-cpu] All HuggingFace models of the dynamo benchmark suites failed with the latest triton-cpu backend,https://github.com/pytorch/pytorch/issues/183686,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T08:54:26Z +issue,127621,custom_op API: better type anntation for Tuple,https://github.com/pytorch/pytorch/issues/127621,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T04:03:04Z +issue,129395,[custom_op] Figure out what to do with random custom ops,https://github.com/pytorch/pytorch/issues/129395,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T04:02:55Z +issue,147210,Consider relaxing the in-place mutation restriction for torch.cond and torch.while_loop,https://github.com/pytorch/pytorch/issues/147210,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T03:44:31Z +issue,146941,Make HOPs more debuggable,https://github.com/pytorch/pytorch/issues/146941,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T03:44:31Z +issue,146860,[BE] Put all the HOP tests in one location,https://github.com/pytorch/pytorch/issues/146860,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T02:27:56Z +issue,146936,flexible custom operators: custom operators that accept arbitrary input/output type,https://github.com/pytorch/pytorch/issues/146936,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T02:27:55Z +issue,146861,[BE] the majority of HOP OpInfo tests should be in one place,https://github.com/pytorch/pytorch/issues/146861,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-14T02:27:54Z +issue,146111,[scan] Support lowering and lifted arguments for inductor,https://github.com/pytorch/pytorch/issues/146111,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-13T23:37:38Z +issue,182817,[MPS] Switch test_output_match / test_output_grad_match to magnitude-floored relative tolerance,https://github.com/pytorch/pytorch/issues/182817,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: numerical-stability, module: tests","decide whether this is bug, use case, docs, or support",2026-05-13T20:54:44Z +issue,111824,GroupNorm & InstanceNorm does not handle channels_last correctly,https://github.com/pytorch/pytorch/issues/111824,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: memory format, module: nn","decide whether this is bug, use case, docs, or support",2026-05-13T18:29:11Z +issue,144805,"""GenericHOPVariable"" / abstract out Dynamo support for HOPs",https://github.com/pytorch/pytorch/issues/144805,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-13T04:03:13Z +issue,183036,Adadelta uses SGD in its examples,https://github.com/pytorch/pytorch/issues/183036,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: optimizer","decide whether this is bug, use case, docs, or support",2026-05-11T18:45:21Z +issue,182560,import torch fails with concurrent nested imports despite GIL ('torch._C' is not a package),https://github.com/pytorch/pytorch/issues/182560,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: python frontend","decide whether this is bug, use case, docs, or support",2026-05-11T17:15:02Z +issue,182960,"`torch.cuda.ExternalStream(0, device=...)` silently returns a fresh pooled stream instead of wrapping handle `0x0`",https://github.com/pytorch/pytorch/issues/182960,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda, module: edge cases, module: python frontend","decide whether this is bug, use case, docs, or support",2026-05-11T17:07:32Z +issue,142135,Device assert throws a runtime error in cuda backend and results in a crash in xpu backend,https://github.com/pytorch/pytorch/issues/142135,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xpu,"decide whether this is bug, use case, docs, or support",2026-05-11T07:56:53Z +issue,143239,"xpu: torch.nn.DataParallel fails on multi-XPU environment with ""module 'torch._C' has no attribute '_scatter'""",https://github.com/pytorch/pytorch/issues/143239,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: xpu, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-05-10T08:19:10Z +issue,179707,Improve static type hints for dataloader with known `collate_fn`,https://github.com/pytorch/pytorch/issues/179707,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: typing","decide whether this is bug, use case, docs, or support",2026-05-09T17:48:15Z +issue,7887,[Caffe2] convert ONNX to caffe2,https://github.com/pytorch/pytorch/issues/7887,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:32:45Z +issue,9867,How can I sure that I run caffe2 with GPU,https://github.com/pytorch/pytorch/issues/9867,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:32:41Z +issue,11869,[caffe2] Unaligned AVX instruction operands in LayerNorm implementation in OSS build,https://github.com/pytorch/pytorch/issues/11869,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:32:25Z +issue,10818,[Caffe2] Can't load pretrained model,https://github.com/pytorch/pytorch/issues/10818,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:32:21Z +issue,10464,caffe2 CI does not test header install,https://github.com/pytorch/pytorch/issues/10464,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:32:09Z +issue,10440,"[caffe2] enforce fail at context_gpu.cu:285, cannot get GPU memory usage statistics",https://github.com/pytorch/pytorch/issues/10440,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:32:05Z +issue,9707,"[caffe2]mpirun multi-node multi GPU in Distributed mode ,run resnet50_trainer.py get RuntimeError",https://github.com/pytorch/pytorch/issues/9707,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:57Z +issue,9349,[Caffe2] Can't find Caffe2<->ONNX conversion tools,https://github.com/pytorch/pytorch/issues/9349,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:52Z +issue,8595,[Caffe2] How to build caffe2/mobile/ulp2/ulp_test?,https://github.com/pytorch/pytorch/issues/8595,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:48Z +issue,8533,[Caffe2] Runtime error while using a pre-trained style_transfer model,https://github.com/pytorch/pytorch/issues/8533,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:44Z +issue,8442,[caffe2] Why the GPU memory consumption has no change after enable optimize_gradient_memory?,https://github.com/pytorch/pytorch/issues/8442,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:40Z +issue,7835,[caffe2] how to using the Mul operator to mul the input vetor,https://github.com/pytorch/pytorch/issues/7835,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:36Z +issue,7614,[Caffe2] modify op in the net from init and predict files,https://github.com/pytorch/pytorch/issues/7614,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:32Z +issue,7569,[Caffe2]How to set lr_mult and decay_mult in Conv layer?,https://github.com/pytorch/pytorch/issues/7569,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:28Z +issue,7491,[caffe2] How to know the input shape for certain pre-trained network,https://github.com/pytorch/pytorch/issues/7491,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:24Z +issue,7490,"[Caffe2]: caffe2.python.caffe_translator.py script doesn't convert ""Split"" and ""Slice"" Layer",https://github.com/pytorch/pytorch/issues/7490,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:20Z +issue,7374,[Caffe2] Can't use resnet50_trainer.py through redis.,https://github.com/pytorch/pytorch/issues/7374,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:16Z +issue,7362,[caffe2] Can I load the gpu data to cpu in .cu file?,https://github.com/pytorch/pytorch/issues/7362,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:12Z +issue,7060,[feature request] Provide Caffe2 CUDA dockerfile with USE_REDIS=ON,https://github.com/pytorch/pytorch/issues/7060,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:07Z +issue,6979,[Caffe2] Is there any way to load custom operators in c++?,https://github.com/pytorch/pytorch/issues/6979,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:31:03Z +issue,6785,[caffe2] Training and inference,https://github.com/pytorch/pytorch/issues/6785,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:30:59Z +issue,6549,caffe2/cuda_rtc can throw during the destructor,https://github.com/pytorch/pytorch/issues/6549,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-05-09T14:30:55Z +issue,179784,"torch.xlogy(0, 0) returns NaN, while in many numerical computing libraries (e.g., SciPy) and common machine learning practice, it is defined as 0.",https://github.com/pytorch/pytorch/issues/179784,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: scipy compatibility, module: special","decide whether this is bug, use case, docs, or support",2026-05-08T19:49:25Z +issue,181946,[MPS] Remove Unnecessary Tensor Memory Gathers and Contiguous Calls for Stride-Aware MPS Backend Inputs,https://github.com/pytorch/pytorch/issues/181946,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: performance","decide whether this is bug, use case, docs, or support",2026-05-08T16:26:55Z +issue,178492,native_batch_norm_backward extremely slow on MPS for 3D tensors (slower than CPU),https://github.com/pytorch/pytorch/issues/178492,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: norms and normalization, module: performance","decide whether this is bug, use case, docs, or support",2026-05-08T14:26:39Z +issue,180343,Accelerator-independent API to determine if a specific backend is available?,https://github.com/pytorch/pytorch/issues/180343,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: accelerator, module: backend, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-05-08T06:40:05Z +issue,146045,Redundant collectives are deduplicated but stick around as dead code,https://github.com/pytorch/pytorch/issues/146045,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-07T18:37:49Z +issue,182842,[v2.11.0] Cannot jvp through vmap,https://github.com/pytorch/pytorch/issues/182842,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: forward ad, module: functorch, module: vmap","decide whether this is bug, use case, docs, or support",2026-05-07T17:05:37Z +issue,180824,"`torch.compile fails with RecursionError on nn.Module with custom __getattr__, while eager execution succeeds`",https://github.com/pytorch/pytorch/issues/180824,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-07T11:08:38Z +issue,139313,BFloat16 support for `torch.fft.rfft`,https://github.com/pytorch/pytorch/issues/139313,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: complex,"decide whether this is bug, use case, docs, or support",2026-05-07T07:22:10Z +issue,167616,UNSTABLE trunk-rocm-sandbox / linux-jammy-rocm-py3.10 / test,https://github.com/pytorch/pytorch/issues/167616,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: rocm","decide whether this is bug, use case, docs, or support",2026-05-07T06:48:26Z +issue,169857,"[ROCm] MIOpen 3D convolution kernel fails with ""invalid configuration argument"" for Qwen2.5-VL vision encoder - silent fallback causes severe performance regression",https://github.com/pytorch/pytorch/issues/169857,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: rocm,"decide whether this is bug, use case, docs, or support",2026-05-06T23:02:58Z +issue,167411,Torch nightly rocm6.4 pytorch-triton-rocm requirement mismatch,https://github.com/pytorch/pytorch/issues/167411,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: rocm","decide whether this is bug, use case, docs, or support",2026-05-06T21:47:28Z +issue,182684,[`SDPA`] Checking for FA backend eligibility via SDPAParams fails under fullgraph compile,https://github.com/pytorch/pytorch/issues/182684,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: sdpa, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-06T18:57:14Z +issue,182450,Spin codegen run doesn't use self-contained environement,https://github.com/pytorch/pytorch/issues/182450,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: codegen, module: lint, module: spin","decide whether this is bug, use case, docs, or support",2026-05-06T15:42:06Z +issue,71389,Create a nested directory while saving objects using `torch.save`,https://github.com/pytorch/pytorch/issues/71389,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: serialization,"decide whether this is bug, use case, docs, or support",2026-05-06T15:39:21Z +issue,160514,BenchmarkingTest.test_benchmark_on_non_zero_device mock hit failed,https://github.com/pytorch/pytorch/issues/160514,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: tests,"decide whether this is bug, use case, docs, or support",2026-05-06T15:33:27Z +issue,180284,Unused jvp function on autograd function causes inscrutable shape error,https://github.com/pytorch/pytorch/issues/180284,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: compiled autograd, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-05-06T15:21:40Z +issue,181718,bf16 matmul throughput drops on RTX 4090 for shapes where N%16==8,https://github.com/pytorch/pytorch/issues/181718,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: bfloat16, module: cublas, module: cuda, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-05-06T12:43:17Z +issue,149639,Pylint error: ` torch.linalg.vector_norm is not callable`,https://github.com/pytorch/pytorch/issues/149639,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: lint, module: typing","decide whether this is bug, use case, docs, or support",2026-05-06T10:53:55Z +issue,168871,TestModule: test/test_transformers.py::TestSDPACudaOnly,https://github.com/pytorch/pytorch/issues/168871,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rocm, module: sdpa","decide whether this is bug, use case, docs, or support",2026-05-04T22:37:10Z +issue,182381,[HOP] Backward pass for scan HOP can result in invalid tensor metadata,https://github.com/pytorch/pytorch/issues/182381,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: faketensor, module: higher order operators, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-05-04T22:25:38Z +issue,144459,"In AOTI Runtime, acquiring a write lock on model_exec_mutex_ may cause starvation.",https://github.com/pytorch/pytorch/issues/144459,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, oncall: export","decide whether this is bug, use case, docs, or support",2026-05-04T21:42:34Z +issue,182336,Adding new lint passes PR CI but not trunk CI,https://github.com/pytorch/pytorch/issues/182336,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: devx, module: lint, module: spin","decide whether this is bug, use case, docs, or support",2026-05-04T20:03:28Z +issue,181723,torch.cuda.make_graphed_callables can silently corrupt parameter gradient accumulation by returning static grad buffers,https://github.com/pytorch/pytorch/issues/181723,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: correctness (silent), module: cuda, module: cuda graphs","decide whether this is bug, use case, docs, or support",2026-05-04T17:17:56Z +issue,181801,Inconsistent Signed-Zero Handling Between CPU and CUDA Backends,https://github.com/pytorch/pytorch/issues/181801,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda, module: edge cases, module: nn","decide whether this is bug, use case, docs, or support",2026-05-04T17:15:15Z +issue,182286,[Bug] CUDA caching allocator unable to satisfy 3-18 GiB contiguous allocation despite 50+ GiB free on RTX PRO 6000 Blackwell + WSL2 (16 GiB hidden CUDA overhead),https://github.com/pytorch/pytorch/issues/182286,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cudacachingallocator, module: memory usage, module: wsl","decide whether this is bug, use case, docs, or support",2026-05-04T13:50:31Z +issue,113228,dataloader use lmdb stuck,https://github.com/pytorch/pytorch/issues/113228,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: deadlock,"decide whether this is bug, use case, docs, or support",2026-05-04T08:16:53Z +issue,180968,Initializing weights after parametrization silently fails,https://github.com/pytorch/pytorch/issues/180968,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: initialization, module: nn, module: nn.utils.parametrize","decide whether this is bug, use case, docs, or support",2026-05-03T20:40:30Z +issue,172807,"[CUDA] Auto-detected arch flags omit accelerated suffix (sm_120a → sm_120), breaking Blackwell NVFP4/block-scaled MMA",https://github.com/pytorch/pytorch/issues/172807,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cuda","decide whether this is bug, use case, docs, or support",2026-05-03T13:13:57Z +issue,74839,Feature request: Root-finding functionality (like scipy.optimize.root_scalar),https://github.com/pytorch/pytorch/issues/74839,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: scientific computing,"decide whether this is bug, use case, docs, or support",2026-05-03T12:15:55Z +issue,182156,Add torch.utils.nan_guard for forward-pass NaN/Inf debugging,https://github.com/pytorch/pytorch/issues/182156,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nans and infs, module: nn","decide whether this is bug, use case, docs, or support",2026-05-01T18:23:22Z +issue,112903,Support for Bazel workspace function or Bazel module,https://github.com/pytorch/pytorch/issues/112903,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: bazel, module: build","decide whether this is bug, use case, docs, or support",2026-05-01T16:13:37Z +issue,181737,image diffusion model output changes for the worse after 530ff92655,https://github.com/pytorch/pytorch/issues/181737,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cublas, module: cuda, module: numerical-stability","decide whether this is bug, use case, docs, or support",2026-05-01T14:49:33Z +issue,182105,Adding a UserWarning when pickling / deepcopying tensor views,https://github.com/pytorch/pytorch/issues/182105,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: molly-guard, module: pickle, module: serialization, module: ux","decide whether this is bug, use case, docs, or support",2026-05-01T10:35:50Z +issue,182008,Windows Application Control (WDAC) blocks loading unsigned torch DLL,https://github.com/pytorch/pytorch/issues/182008,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: windows","decide whether this is bug, use case, docs, or support",2026-04-30T17:48:24Z +issue,145555,need to document `FlopCounterMode`,https://github.com/pytorch/pytorch/issues/145555,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-04-30T17:28:06Z +issue,153546,Make implicit packages (PEP420) explicit PyTorch,https://github.com/pytorch/pytorch/issues/153546,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: lint","decide whether this is bug, use case, docs, or support",2026-04-30T17:28:05Z +issue,75225,One of the backends for `lu_factor` fails on windows.,https://github.com/pytorch/pytorch/issues/75225,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: windows","decide whether this is bug, use case, docs, or support",2026-04-30T07:30:24Z +issue,180839,[Dynamo] Attempted to call function marked as skipped: torch.sparse_csr_tensor cannot be traced,https://github.com/pytorch/pytorch/issues/180839,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: sparse, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-30T00:48:31Z +issue,141265,rope perf improvement,https://github.com/pytorch/pytorch/issues/141265,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-29T23:22:13Z +issue,181494,Add support for sparse block-diagonal matrices,https://github.com/pytorch/pytorch/issues/181494,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sparse,"decide whether this is bug, use case, docs, or support",2026-04-29T18:54:21Z +issue,103841,[question] [docs] Short/mid/long-term status of TorchScript / JIT / torch.jit.trace / FX / symbolic tracing and its replacement by Dynamo,https://github.com/pytorch/pytorch/issues/103841,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, oncall: jit","decide whether this is bug, use case, docs, or support",2026-04-28T14:37:09Z +issue,152355,[source.wheel] Pin setuptools runtime dependency,https://github.com/pytorch/pytorch/issues/152355,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: build, module: cpp-extensions","decide whether this is bug, use case, docs, or support",2026-04-28T06:58:10Z +issue,181649,UNSTABLE rocm-nightly / linux-noble-rocm-nightly-py3.12-gfx942 / test (distributed),https://github.com/pytorch/pytorch/issues/181649,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: rocm","decide whether this is bug, use case, docs, or support",2026-04-27T20:45:25Z +issue,181648,UNSTABLE rocm-nightly / linux-noble-rocm-nightly-py3.12-gfx942 / test (inductor),https://github.com/pytorch/pytorch/issues/181648,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-27T20:45:13Z +issue,181647,UNSTABLE rocm-nightly / linux-noble-rocm-nightly-py3.12-gfx942 / test (default),https://github.com/pytorch/pytorch/issues/181647,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: rocm","decide whether this is bug, use case, docs, or support",2026-04-27T20:45:10Z +issue,152710,"[FSDP2] NO_SHARD as fully_shard(mesh=(Replicate, Shard)) with shard of world size 1",https://github.com/pytorch/pytorch/issues/152710,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: distributed,"decide whether this is bug, use case, docs, or support",2026-04-27T13:40:13Z +issue,179008,[privateuse1] Update the privateuse1 tutorial to mention registering HooksInterface,https://github.com/pytorch/pytorch/issues/179008,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: privateuse1,"decide whether this is bug, use case, docs, or support",2026-04-27T10:54:18Z +issue,179010,[privateuse1] Update privateuse1 tutorial to show the Python backend approach,https://github.com/pytorch/pytorch/issues/179010,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: privateuse1,"decide whether this is bug, use case, docs, or support",2026-04-27T10:52:58Z +issue,154318,[BUG] DataLoader low GPU utilization and extremely slow compared to manual batching,https://github.com/pytorch/pytorch/issues/154318,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data, module: dataloader, module: performance","decide whether this is bug, use case, docs, or support",2026-04-27T09:46:06Z +issue,158714,Feature request: `DataLoader` using multithreading instead of multiprocessing,https://github.com/pytorch/pytorch/issues/158714,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: data, module: dataloader, module: multithreading","decide whether this is bug, use case, docs, or support",2026-04-27T06:18:53Z +issue,154363,`scaled_dot_product_attention` broadcasting (GQA) is a memory footgun,https://github.com/pytorch/pytorch/issues/154363,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: memory usage, module: sdpa","decide whether this is bug, use case, docs, or support",2026-04-26T22:24:54Z +issue,181525,explore using gmpy2 to accelerate expression eval in sympy,https://github.com/pytorch/pytorch/issues/181525,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-26T20:20:56Z +issue,179294,[MPS] `scaled_dot_product_attention` (SDPA) improvements,https://github.com/pytorch/pytorch/issues/179294,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: performance, module: sdpa","decide whether this is bug, use case, docs, or support",2026-04-26T01:54:18Z +issue,179430,Add Frechet/Inverse Weibull distribution,https://github.com/pytorch/pytorch/issues/179430,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: distributions,"decide whether this is bug, use case, docs, or support",2026-04-25T18:55:22Z +issue,96602,"[MPS] softmax returns NaN attention probabilities for large tensors, in float16 and float32.",https://github.com/pytorch/pytorch/issues/96602,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: mps,"decide whether this is bug, use case, docs, or support",2026-04-25T15:36:08Z +issue,181488,[TensorIterator] reorder_dimensions is sensitive to operand order for commutative ops,https://github.com/pytorch/pytorch/issues/181488,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, module: tensoriterator","decide whether this is bug, use case, docs, or support",2026-04-25T11:51:14Z +issue,181379,"[cuDNN][SDPA] head_dim limit stuck at 128 on sm_120 (RTX 5090), blocking head_dim=256 models",https://github.com/pytorch/pytorch/issues/181379,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cudnn, module: sdpa","decide whether this is bug, use case, docs, or support",2026-04-24T17:33:44Z +issue,28648,Can't successfully install pytorch with python3.6 on my pi 4,https://github.com/pytorch/pytorch/issues/28648,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-04-23T17:41:42Z +issue,180620,"@custom_op doesn't support ""maybe_out semantics""",https://github.com/pytorch/pytorch/issues/180620,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-23T14:47:32Z +issue,180960,[cudagraphs] torch.cond does not work with cudagraphs when nccl collectives are used,https://github.com/pytorch/pytorch/issues/180960,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: higher order operators, module: nccl, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-04-23T13:46:02Z +issue,178600,RuntimeError of use_libuv and ReduceOp.AVG with DDP + Gloo on Windows multi-gpu train,https://github.com/pytorch/pytorch/issues/178600,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: ddp, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-04-23T05:33:41Z +issue,171022,In-place real-domain FFT operators for memory-efficient training (bf16 support),https://github.com/pytorch/pytorch/issues/171022,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: fft,"decide whether this is bug, use case, docs, or support",2026-04-23T02:45:24Z +issue,181108,Use of dynamic shapes impedes Inductor fusion for XSA projection,https://github.com/pytorch/pytorch/issues/181108,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: inductor, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-22T17:24:08Z +issue,113149,Compilation error on loongarch64,https://github.com/pytorch/pytorch/issues/113149,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: third_party","decide whether this is bug, use case, docs, or support",2026-04-22T15:07:22Z +issue,181076,[Inductor] Improve compile time of `build_fusion_regions`,https://github.com/pytorch/pytorch/issues/181076,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-22T07:49:54Z +issue,180975,[Tracking] RISC-V PyTorch enablement,https://github.com/pytorch/pytorch/issues/180975,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: ci, module: cpu, module: risc-v","decide whether this is bug, use case, docs, or support",2026-04-21T13:12:07Z +issue,177654,`import functorch.dim` silently breaks `torch.Tensor.expand(size=...)`,https://github.com/pytorch/pytorch/issues/177654,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: functorch","decide whether this is bug, use case, docs, or support",2026-04-21T12:34:52Z +issue,52241,CTCLoss gradient is incorrect,https://github.com/pytorch/pytorch/issues/52241,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: docs, module: loss","decide whether this is bug, use case, docs, or support",2026-04-20T22:17:26Z +issue,180915,accelerate sympy computations using gmpy2,https://github.com/pytorch/pytorch/issues/180915,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-20T21:11:06Z +issue,179710,AArch64 unit test failure TestFX.test_shape_prop_layout_3d,https://github.com/pytorch/pytorch/issues/179710,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: tests","decide whether this is bug, use case, docs, or support",2026-04-20T10:09:37Z +issue,174970,[RFC] Hooks to trigger recompilation,https://github.com/pytorch/pytorch/issues/174970,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: dynamo, module: fx, module: inductor","decide whether this is bug, use case, docs, or support",2026-04-20T08:23:22Z +issue,158917,[Feat] Tracking for OpenReg Improvements,https://github.com/pytorch/pytorch/issues/158917,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: openreg, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-04-20T06:41:25Z +issue,180650,`torch.linalg._powsum` is slow on CPU for ord != 2,https://github.com/pytorch/pytorch/issues/180650,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: linear algebra, module: performance","decide whether this is bug, use case, docs, or support",2026-04-20T03:00:29Z +issue,119613,init_device_mesh doesn't have a timeout kwarg,https://github.com/pytorch/pytorch/issues/119613,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: devicemesh, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-04-19T07:56:23Z +issue,116697,Tracing GPU usage of `torch.vmap` vectorized operation using `torch.profiler` cause CUDA OOM,https://github.com/pytorch/pytorch/issues/116697,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: vmap, oncall: profiler","decide whether this is bug, use case, docs, or support",2026-04-19T01:34:35Z +issue,131972,backward of adaptive max pool (adaptive_max_pool2d_backward_cuda) doesn't have a deterministic implementation,https://github.com/pytorch/pytorch/issues/131972,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: determinism","decide whether this is bug, use case, docs, or support",2026-04-18T22:09:51Z +issue,158551,CUDAGraphs RuntimeError: Accessing Overwritten Tensor Output Despite Clone and cudagraph_mark_step_begin in PyTorch 2.3.0,https://github.com/pytorch/pytorch/issues/158551,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: ddp, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-18T15:22:15Z +issue,151204,[Dynamo] Dynamo fails to trace reduce_scatter_v,https://github.com/pytorch/pytorch/issues/151204,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: dynamo, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-18T15:09:31Z +issue,172231,[cuBLAS] setting `cublas/cublaslt` backend seems to be a no-op for `torch.addmm` when dispatched to cuBLASLt,https://github.com/pytorch/pytorch/issues/172231,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cublas,"decide whether this is bug, use case, docs, or support",2026-04-18T00:05:37Z +issue,158343,Upgrade to latest XNNPACK,https://github.com/pytorch/pytorch/issues/158343,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xnnpack,"decide whether this is bug, use case, docs, or support",2026-04-17T21:17:16Z +issue,180712,Add a lazy import hook for autoloading?,https://github.com/pytorch/pytorch/issues/180712,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: backend, module: ux","decide whether this is bug, use case, docs, or support",2026-04-17T20:10:18Z +issue,157807,[RFC] Replace setuptools build backend with scikit-build-core,https://github.com/pytorch/pytorch/issues/157807,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-04-16T23:16:52Z +issue,180596,capture_dynamic_output_shape_ops=True crashes 8/50 HuggingFace models with GuardOnDataDependentSymNode,https://github.com/pytorch/pytorch/issues/180596,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: guards, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-16T19:42:40Z +issue,180595,capture_scalar_outputs=True crashes 7/50 HuggingFace models with PendingUnbackedSymbolNotFound,https://github.com/pytorch/pytorch/issues/180595,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-16T19:42:31Z +issue,179520,Different results for with and without torch.compile on flux.1-dev model,https://github.com/pytorch/pytorch/issues/179520,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-16T17:50:16Z +issue,173897,XPU dist.all_reduce is async: torch.xpu.synchronize does not wait for collective completion,https://github.com/pytorch/pytorch/issues/173897,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: xpu, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-04-16T14:52:44Z +issue,134956,Pytorch XPU Windows build failed in cmake rerun loop due to the source code deep path,https://github.com/pytorch/pytorch/issues/134956,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xpu,"decide whether this is bug, use case, docs, or support",2026-04-16T08:38:48Z +issue,179858,torch.compile doesn't use async all_to_all,https://github.com/pytorch/pytorch/issues/179858,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, module: dynamo, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-15T19:10:38Z +issue,178883,`torch.compile` produces different output for model with `amax`/`amin` + partial reuse reduction pattern,https://github.com/pytorch/pytorch/issues/178883,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: correctness (silent), module: cuda, module: decompositions","decide whether this is bug, use case, docs, or support",2026-04-15T14:43:28Z +issue,177225,mps test got Exception: mean(): could not infer output dtype.,https://github.com/pytorch/pytorch/issues/177225,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: primtorch, module: tests","decide whether this is bug, use case, docs, or support",2026-04-15T11:00:01Z +issue,148402,Generate two reduction loops for vectorization,https://github.com/pytorch/pytorch/issues/148402,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-15T00:56:02Z +issue,154356,"Q.size(-1) == m INTERNAL ASSERT FAILED at ""/pytorch/aten/src/ATen/native/BatchLinearAlgebra.cpp",https://github.com/pytorch/pytorch/issues/154356,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: error checking, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-04-14T19:23:25Z +issue,180362,RFC: Power (ppc64le) CI and Build Infrastructure Plan,https://github.com/pytorch/pytorch/issues/180362,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: ci, module: power","decide whether this is bug, use case, docs, or support",2026-04-14T18:22:31Z +issue,170753,[CI][Windows][Infrastructure] Support for RTX GPU workflows on Windows,https://github.com/pytorch/pytorch/issues/170753,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: windows","decide whether this is bug, use case, docs, or support",2026-04-14T16:12:33Z +issue,169277,[Accelerator] Hardware native support versus Software simulated support for Device Capabilties,https://github.com/pytorch/pytorch/issues/169277,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: accelerator,"decide whether this is bug, use case, docs, or support",2026-04-14T11:10:33Z +issue,179709,Cannot export model with CenterCrop to ONNX,https://github.com/pytorch/pytorch/issues/179709,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: onnx,"decide whether this is bug, use case, docs, or support",2026-04-13T19:12:22Z +issue,147478,Blackwell Only Bugs,https://github.com/pytorch/pytorch/issues/147478,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: inductor, module: tests, oncall: export","decide whether this is bug, use case, docs, or support",2026-04-13T17:23:04Z +issue,150725,Continuous calls to nn.Linear in fp32 on the 5090D cause severe performance degradation,https://github.com/pytorch/pytorch/issues/150725,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: nn, module: performance","decide whether this is bug, use case, docs, or support",2026-04-13T17:23:03Z +issue,153555,Updated Scaled_mm to support more scaling formats via CuBlas,https://github.com/pytorch/pytorch/issues/153555,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: floatx (formerly float8)","decide whether this is bug, use case, docs, or support",2026-04-13T17:23:02Z +issue,169015,[CI][CUDA][B200] Test_weight_norm_conv2d_cuda Failure,https://github.com/pytorch/pytorch/issues/169015,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-04-13T17:22:55Z +issue,178839,[ROCm] MIOpen Gemm solver receives workspace_size=0 via legacy Find1 API,https://github.com/pytorch/pytorch/issues/178839,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: convolution, module: rocm","decide whether this is bug, use case, docs, or support",2026-04-13T17:12:02Z +issue,146542,DeepSeek: mixed precision optimizers (BF16AdamW),https://github.com/pytorch/pytorch/issues/146542,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: optimizer,"decide whether this is bug, use case, docs, or support",2026-04-13T15:43:07Z +issue,159770,extract statistics from attention weights in FlexAttention,https://github.com/pytorch/pytorch/issues/159770,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-13T14:50:21Z +issue,48972,non-negative least squares solver feature request,https://github.com/pytorch/pytorch/issues/48972,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-04-13T08:50:09Z +issue,127176,"RuntimeError: ""_amp_foreach_non_finite_check_and_unscale_cuda"" not implemented for 'BFloat16'",https://github.com/pytorch/pytorch/issues/127176,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: amp (automated mixed precision), module: bfloat16, module: mta","decide whether this is bug, use case, docs, or support",2026-04-13T05:08:08Z +issue,121122,Evaluate defaulting `TORCHINDUCTOR_CACHE_DIR` to `~/.cache` on Linux systems,https://github.com/pytorch/pytorch/issues/121122,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-11T23:49:42Z +issue,25174,BCEWithLogitsLoss expects wrong shape of weight (#classes instead of batch size),https://github.com/pytorch/pytorch/issues/25174,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: loss, module: nn","decide whether this is bug, use case, docs, or support",2026-04-11T17:08:12Z +issue,157345,nn.rmsnorm is super slower than nn.layernorm,https://github.com/pytorch/pytorch/issues/157345,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: norms and normalization, module: performance","decide whether this is bug, use case, docs, or support",2026-04-11T07:46:04Z +issue,179510,RuntimeError: Argument 'primals_out' of Node 'results' was used before it has been defined! Please check that Nodes in the graph are topologically ordered,https://github.com/pytorch/pytorch/issues/179510,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functorch, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-10T21:10:22Z +issue,59194,Add numerically-stable function for angle between vectors,https://github.com/pytorch/pytorch/issues/59194,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distance functions, module: numerical-stability","decide whether this is bug, use case, docs, or support",2026-04-09T18:22:45Z +issue,128121,2nd compile of deepcopy(model) fails on multiple ubuntu-pc (fatal error: Python.h: file not found),https://github.com/pytorch/pytorch/issues/128121,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-09T18:09:18Z +issue,145374,Memory Leak in MPS Backend During LSTM Iterations (Out of Memory Error),https://github.com/pytorch/pytorch/issues/145374,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: memory usage, module: mps, module: rnn","decide whether this is bug, use case, docs, or support",2026-04-09T02:11:29Z +issue,146691,MPS support `torch.linalg.norm` on complex numbers,https://github.com/pytorch/pytorch/issues/146691,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: complex, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-04-09T00:13:29Z +issue,179643,Call Hierarchy Metadata,https://github.com/pytorch/pytorch/issues/179643,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: fx, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-08T18:00:48Z +issue,127605,RuntimeError: CUDA error: an illegal memory access was encountered with specific input shape and Conv1d,https://github.com/pytorch/pytorch/issues/127605,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cudnn, module: nn","decide whether this is bug, use case, docs, or support",2026-04-08T14:51:38Z +issue,127078,Wrong Flag defined for ppc64le arch,https://github.com/pytorch/pytorch/issues/127078,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cpu,"decide whether this is bug, use case, docs, or support",2026-04-07T19:15:04Z +issue,150655,Create better alerting for binary size validations and time it takes to build the binary,https://github.com/pytorch/pytorch/issues/150655,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: binaries,"decide whether this is bug, use case, docs, or support",2026-04-07T18:58:17Z +issue,169670,"Energy normalization error on torch.fft.fft2 and torch.fft.ifft2 when the matrix size = [2048, 2048]",https://github.com/pytorch/pytorch/issues/169670,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: intel, module: linear algebra, module: mkl","decide whether this is bug, use case, docs, or support",2026-04-07T13:05:23Z +issue,55937,test_stream_event_nogil: Is the test making a wrong assumption?,https://github.com/pytorch/pytorch/issues/55937,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rocm, module: tests","decide whether this is bug, use case, docs, or support",2026-04-07T11:42:56Z +issue,174913,Test: TestLinalg.test_tensorinv,https://github.com/pytorch/pytorch/issues/174913,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: rocm","decide whether this is bug, use case, docs, or support",2026-04-06T21:56:35Z +issue,113045,[Feature][DTensor] Manage additional `_padded_local_tensor` attribute,https://github.com/pytorch/pytorch/issues/113045,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-04-06T17:57:20Z +issue,179448,[DTensor] Inconsistent output placements of sharded DTensor indexing,https://github.com/pytorch/pytorch/issues/179448,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-04-06T17:46:20Z +issue,179382,Support dropout in varlen_attn,https://github.com/pytorch/pytorch/issues/179382,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sdpa,"decide whether this is bug, use case, docs, or support",2026-04-05T14:37:43Z +issue,178749,[FSDP] Multi-GPU distributed tests crash with NCCL 2.29.7 NVLS multicast slot exhaustion on H200,https://github.com/pytorch/pytorch/issues/178749,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: fsdp, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-04-05T09:12:37Z +issue,133168,The more detailed GELU formula should be added to the doc of `GELU()`:,https://github.com/pytorch/pytorch/issues/133168,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-04-05T06:23:39Z +issue,160875,Add `uv` installation commands to pytorch.org,https://github.com/pytorch/pytorch/issues/160875,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, oncall: releng","decide whether this is bug, use case, docs, or support",2026-04-03T15:05:30Z +issue,171158,`torch.func.grad` to allow some inplace ops,https://github.com/pytorch/pytorch/issues/171158,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: functorch","decide whether this is bug, use case, docs, or support",2026-04-03T12:16:41Z +issue,167540,"[Dtensor]:change the test_mm shape from (12,8) * (8,16) to (512, 512) * (512, 512), throw assert error",https://github.com/pytorch/pytorch/issues/167540,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-04-02T23:19:19Z +issue,178482,torch.compile(dynamic=True) ignores incompatible `out` tensor for `torch.empty` and returns wrong shape,https://github.com/pytorch/pytorch/issues/178482,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: tensor creation","decide whether this is bug, use case, docs, or support",2026-04-02T22:40:43Z +issue,178930,[torch.compile] Better way to override the default backend,https://github.com/pytorch/pytorch/issues/178930,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: backend, module: compile ux, module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-02T21:23:58Z +issue,158710,[feature request] Inplace downcast dtype conversion,https://github.com/pytorch/pytorch/issues/158710,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: memory usage,"decide whether this is bug, use case, docs, or support",2026-04-02T18:34:56Z +issue,175695,Multi Tensor Apply ops have low occupancy due to a maximum of 320 thread blocks per kernel launch,https://github.com/pytorch/pytorch/issues/175695,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: mta, module: performance","decide whether this is bug, use case, docs, or support",2026-04-02T14:35:49Z +issue,179124,grid_sampler_2d_backward_cuda is not deterministic when torch.use_deterministic_algorithms(True),https://github.com/pytorch/pytorch/issues/179124,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: determinism, module: interpolation","decide whether this is bug, use case, docs, or support",2026-04-02T06:44:57Z +issue,178864,torch.multinomial on CUDA with zero-sum probabilities triggers device-side assert instead of invalid multinomial distribution error,https://github.com/pytorch/pytorch/issues/178864,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: edge cases, module: error checking","decide whether this is bug, use case, docs, or support",2026-04-01T22:23:35Z +issue,178988,Cannot build documentation (`RecursionError`),https://github.com/pytorch/pytorch/issues/178988,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: docs","decide whether this is bug, use case, docs, or support",2026-04-01T20:24:49Z +issue,113063,Rework Dynamic Benchmarks To Actually Vary Shapes,https://github.com/pytorch/pytorch/issues/113063,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: benchmark, module: dynamic shapes, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-01T18:58:08Z +issue,151997,torch.compile doesnot support index with tensor,https://github.com/pytorch/pytorch/issues/151997,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-01T15:40:20Z +issue,140463,Ban all operations on sympy expressions in Inductor,https://github.com/pytorch/pytorch/issues/140463,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-04-01T15:37:42Z +issue,90900,Check that SymPy semantics match Python semantics,https://github.com/pytorch/pytorch/issues/90900,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes,"decide whether this is bug, use case, docs, or support",2026-03-31T23:20:19Z +issue,44282,PyTorch wheel's own OpenMP library clashing with system-wide OpenMP library at runtime,https://github.com/pytorch/pytorch/issues/44282,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: openmp,"decide whether this is bug, use case, docs, or support",2026-03-31T14:48:09Z +issue,178820,Add Dynamo support for dist.record_comm (capture_profiler_record_comm),https://github.com/pytorch/pytorch/issues/178820,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, module: dynamo, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-31T02:38:06Z +issue,178616,[DTensor] reshape is dispatched to aten.view instead of aten.reshape,https://github.com/pytorch/pytorch/issues/178616,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-03-30T20:21:12Z +issue,165070,Dynamically load CUDSS library using pypi wheel to save wheel space,https://github.com/pytorch/pytorch/issues/165070,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cuda","decide whether this is bug, use case, docs, or support",2026-03-30T18:39:43Z +issue,141827,Compiled and masked MHA running with DDP and `no_grad()` leads to Graph break due to unsupported builtin torch._C._distributed_c10d.PyCapsule._broadcast_coalesced,https://github.com/pytorch/pytorch/issues/141827,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-03-30T18:21:46Z +issue,166563,[RFC] Modifying Getting started page for Experimental Wheel Variant Support,https://github.com/pytorch/pytorch/issues/166563,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-03-30T17:49:33Z +issue,178593,torch.gcd returns incorrect results due to automatic dtype casting between signed and unsigned integers,https://github.com/pytorch/pytorch/issues/178593,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: type promotion, module: unsigned int","decide whether this is bug, use case, docs, or support",2026-03-30T17:30:40Z +issue,178614,bag boundry is not calculated correctly in EmbeddingBag when setting include_last_offset = True,https://github.com/pytorch/pytorch/issues/178614,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda, module: embedding, module: nn","decide whether this is bug, use case, docs, or support",2026-03-30T17:28:57Z +issue,177599,arange CPU kernel has different precisions for scalar and vectorized executions,https://github.com/pytorch/pytorch/issues/177599,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: half, module: numerical-stability","decide whether this is bug, use case, docs, or support",2026-03-30T17:09:11Z +issue,178345,[build env combo] elainewy,https://github.com/pytorch/pytorch/issues/178345,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-03-30T17:07:13Z +issue,177548,Update Internal fbsource CK,https://github.com/pytorch/pytorch/issues/177548,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rocm, module: third_party","decide whether this is bug, use case, docs, or support",2026-03-30T15:53:36Z +issue,162872,[MPS] dead lock when calling `Event.synchronize()` before `Event.elapsed_time()`,https://github.com/pytorch/pytorch/issues/162872,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: deadlock, module: mps","decide whether this is bug, use case, docs, or support",2026-03-30T07:16:53Z +issue,178472,conv1d on meta device path skips groups validation and returns an invalid output shape,https://github.com/pytorch/pytorch/issues/178472,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: convolution, module: error checking, module: meta tensors","decide whether this is bug, use case, docs, or support",2026-03-27T02:39:11Z +issue,175022,[RFC] Cross-Repository CI Relay for PyTorch Out-of-Tree Backends,https://github.com/pytorch/pytorch/issues/175022,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-03-26T20:33:42Z +issue,175409,Choose rounding mode when casting to dtype,https://github.com/pytorch/pytorch/issues/175409,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: floatx (formerly float8), module: inductor, module: type promotion","decide whether this is bug, use case, docs, or support",2026-03-26T18:01:13Z +issue,178511,[torch.compile] allow_in_graph called inside compiled function causes Unsupported error,https://github.com/pytorch/pytorch/issues/178511,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-26T15:11:33Z +issue,174045,Suboptimal `affine_grid` implementation,https://github.com/pytorch/pytorch/issues/174045,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: performance","decide whether this is bug, use case, docs, or support",2026-03-26T14:50:48Z +issue,154299,ImportError: libcudnn.so.9: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/154299,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: cuda","decide whether this is bug, use case, docs, or support",2026-03-26T07:46:36Z +issue,154052,Most requested ops for the MPS backend,https://github.com/pytorch/pytorch/issues/154052,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: mps,"decide whether this is bug, use case, docs, or support",2026-03-26T04:58:41Z +issue,178091,"propagate_single_input_strategy copies input tensor_meta to output, producing wrong dtype for ops like convert_element_type",https://github.com/pytorch/pytorch/issues/178091,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-03-25T21:10:29Z +issue,176065,Errors in Activation Checkpointing tutorial,https://github.com/pytorch/pytorch/issues/176065,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-03-25T20:48:47Z +issue,170837,MPS backend - inconsistent results for batched inference on BERT/RoBERTa,https://github.com/pytorch/pytorch/issues/170837,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: mps","decide whether this is bug, use case, docs, or support",2026-03-25T20:48:04Z +issue,71381,Cannot import PyTorch in Alpine Docker Container,https://github.com/pytorch/pytorch/issues/71381,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: build","decide whether this is bug, use case, docs, or support",2026-03-25T20:18:38Z +issue,44027,Immutable (read-only) tensors,https://github.com/pytorch/pytorch/issues/44027,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-03-25T13:02:47Z +issue,166291,"""FSDP does not support non-contiguous parameters yet"" prevents using channels-last convolutions",https://github.com/pytorch/pytorch/issues/166291,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: distributed,"decide whether this is bug, use case, docs, or support",2026-03-25T04:57:00Z +issue,177935,[RFC] Add A Cross-Backend Device Name API: `torch.accelerator.current_device_name()`,https://github.com/pytorch/pytorch/issues/177935,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: accelerator,"decide whether this is bug, use case, docs, or support",2026-03-25T03:25:24Z +issue,177021,linear attention / mamba2 OP,https://github.com/pytorch/pytorch/issues/177021,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-03-24T20:20:38Z +issue,177014,`cudaErrorInvalidDeviceFunction` error due to use of `__CUDA_ARCH__` in host code,https://github.com/pytorch/pytorch/issues/177014,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cuda","decide whether this is bug, use case, docs, or support",2026-03-24T19:08:09Z +issue,178052,Deterministic implementation of CTC Loss,https://github.com/pytorch/pytorch/issues/178052,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: determinism, module: loss","decide whether this is bug, use case, docs, or support",2026-03-24T17:50:12Z +issue,177910,nvshmem symm memory backend causes stuck sometimes for helion distributed kernel autotuning,https://github.com/pytorch/pytorch/issues/177910,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: deadlock, module: helion, module: symm_mem","decide whether this is bug, use case, docs, or support",2026-03-24T15:11:32Z +issue,99037,AttributeError: type object 'torch._C._profiler.ProfilerActivity' has no attribute 'MPS',https://github.com/pytorch/pytorch/issues/99037,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, oncall: profiler","decide whether this is bug, use case, docs, or support",2026-03-24T08:22:46Z +issue,177771,Accelerators: torch.Stream.record_event doesn't handle torch.cuda.Events even though torch.cuda.Event is a torch.Event subclass,https://github.com/pytorch/pytorch/issues/177771,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: accelerator, module: cuda","decide whether this is bug, use case, docs, or support",2026-03-23T22:17:19Z +issue,177477,Misleading error message (INTERNAL ASSERT FAILED) in `jiterator._create_multi_output_jit_fn` when passing CPU tensors,https://github.com/pytorch/pytorch/issues/177477,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: error checking, module: jiterator","decide whether this is bug, use case, docs, or support",2026-03-23T22:17:19Z +issue,177322,Support TORCH_SHOW_CPP_STACKTRACES=1 on Windows,https://github.com/pytorch/pytorch/issues/177322,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: devx, module: windows","decide whether this is bug, use case, docs, or support",2026-03-23T22:17:15Z +issue,165506,[torch.compile] dynamic shape kernel mismatch,https://github.com/pytorch/pytorch/issues/165506,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-23T22:17:12Z +issue,170361,[Dynamo] Use VariableBuilder/SourcelessBuilder consistently,https://github.com/pytorch/pytorch/issues/170361,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-23T19:34:33Z +issue,177663,[DTensor] greedy redistribution cost favors all_gather over all_to_all,https://github.com/pytorch/pytorch/issues/177663,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-03-23T18:11:23Z +issue,177969,test utilities should intersect `torch.accelerator.get_device_capability` when resolving dtypes used for a test,https://github.com/pytorch/pytorch/issues/177969,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: accelerator, module: tests","decide whether this is bug, use case, docs, or support",2026-03-23T17:17:44Z +issue,74746,Change the type hint for nn.Module.__call__ to be friendly to overrides.,https://github.com/pytorch/pytorch/issues/74746,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: typing","decide whether this is bug, use case, docs, or support",2026-03-23T16:22:00Z +issue,178132,torch.compile streams support in AOTI,https://github.com/pytorch/pytorch/issues/178132,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compiled streams, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-23T08:29:51Z +issue,177957,api shows unexpected difference for flatten(): numpy vs. pytorch,https://github.com/pytorch/pytorch/issues/177957,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-03-22T14:45:46Z +issue,178072,[PERFORMANCE] [NN] Change `torch._C._get_tracing_state()` to `torch._C._is_tracing()` for around 15% improved performance of `module.py` `_call_impl()`,https://github.com/pytorch/pytorch/issues/178072,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: performance","decide whether this is bug, use case, docs, or support",2026-03-21T20:07:52Z +issue,53721,PyTorch on Embedded Hardware,https://github.com/pytorch/pytorch/issues/53721,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-03-21T12:46:32Z +issue,177260,Standardize device-agnostic test cases via instantiate_device_type_tests,https://github.com/pytorch/pytorch/issues/177260,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: tests,"decide whether this is bug, use case, docs, or support",2026-03-20T22:16:51Z +issue,177830,Why register_replacement fails to support fuzzy matching for optional operator parameters in search_fn/search_fn_pattern,https://github.com/pytorch/pytorch/issues/177830,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-20T21:32:20Z +issue,177970,`TEST_DEVICE_BIAS` linter scope should be expanded,https://github.com/pytorch/pytorch/issues/177970,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: tests","decide whether this is bug, use case, docs, or support",2026-03-20T16:02:33Z +issue,159635,[BUG] DTensor (PyTorch native TP) + dynamic shapes + torch.compile failed!,https://github.com/pytorch/pytorch/issues/159635,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-20T06:58:25Z +issue,177896,Bot to notify PoC if ci pin updater fails,https://github.com/pytorch/pytorch/issues/177896,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: devx","decide whether this is bug, use case, docs, or support",2026-03-19T20:30:03Z +issue,155720,[rocm] HIP Graph capture raises segmentation fault on AMD GPU but CUDA Graph capture succeeds on Nvidia GPU,https://github.com/pytorch/pytorch/issues/155720,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: rocm,"decide whether this is bug, use case, docs, or support",2026-03-19T16:12:54Z +issue,177859,Combining `bfloat16` with complex scalars raises `NotImplementedError`,https://github.com/pytorch/pytorch/issues/177859,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: bfloat16, module: complex","decide whether this is bug, use case, docs, or support",2026-03-19T15:39:14Z +issue,176274,PyTorch + Math libraries: redundant `copy_` calls when working with buffer views,https://github.com/pytorch/pytorch/issues/176274,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: cuda, module: linear algebra, module: memory usage","decide whether this is bug, use case, docs, or support",2026-03-19T13:11:42Z +issue,177833,StepLR accepts step_size=0 and crashes with ZeroDivisionError on first step(),https://github.com/pytorch/pytorch/issues/177833,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: error checking, module: lrscheduler, module: optimizer","decide whether this is bug, use case, docs, or support",2026-03-19T10:38:49Z +issue,134767,"Setting a negative, `int` and `bool` value to `label_smoothing` argument of `nn.CrossEntropyLoss()` works against the error messages",https://github.com/pytorch/pytorch/issues/134767,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: loss,"decide whether this is bug, use case, docs, or support",2026-03-19T10:24:29Z +issue,164429,Add support for Intel GPU Memory Information,https://github.com/pytorch/pytorch/issues/164429,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: xpu,"decide whether this is bug, use case, docs, or support",2026-03-19T02:07:19Z +issue,136495,[dynamo] enable TorchDispatchMode for eager part when graph breaks,https://github.com/pytorch/pytorch/issues/136495,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-18T23:59:16Z +issue,137749,Graph break on wait() of Torchrec's Awaitable instance,https://github.com/pytorch/pytorch/issues/137749,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-18T23:58:16Z +issue,94792,Compiling libtorch from Source on Mac Beyond v1.11.0,https://github.com/pytorch/pytorch/issues/94792,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cpp, module: macos","decide whether this is bug, use case, docs, or support",2026-03-18T19:56:02Z +issue,123148,GRU is super slow on MPS backend,https://github.com/pytorch/pytorch/issues/123148,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: performance, module: rnn","decide whether this is bug, use case, docs, or support",2026-03-18T18:34:28Z +issue,167297,"DISABLED test_max_autotune_addmm_max_autotune_gemm_backends_ATen,Triton,CK_x_shape1 (__main__.TestCKBackend)",https://github.com/pytorch/pytorch/issues/167297,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-18T13:37:08Z +issue,167272,"DISABLED test_max_autotune_addmm_max_autotune_gemm_backends_ATen,Triton,CK_x_shape2 (__main__.TestCKBackend)",https://github.com/pytorch/pytorch/issues/167272,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-18T13:36:22Z +issue,97668,sparse_csr_tensor matmul wrong output in bfloat16,https://github.com/pytorch/pytorch/issues/97668,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sparse,"decide whether this is bug, use case, docs, or support",2026-03-18T11:49:23Z +issue,157989,[XPU] torch.xpu.mem_get_info() query failed or reports Out-Of-Memory even when there are enough free memory,https://github.com/pytorch/pytorch/issues/157989,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: windows, module: xpu","decide whether this is bug, use case, docs, or support",2026-03-18T06:14:16Z +issue,164966,XPU OOM when allocate tensor according to its reported available memory,https://github.com/pytorch/pytorch/issues/164966,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: memory usage, module: xpu","decide whether this is bug, use case, docs, or support",2026-03-18T05:51:30Z +issue,76324,Bessel and Related Functions,https://github.com/pytorch/pytorch/issues/76324,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: special,"decide whether this is bug, use case, docs, or support",2026-03-18T00:42:34Z +issue,28224,Allow to disable polling for CUDA synchronization,https://github.com/pytorch/pytorch/issues/28224,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: performance","decide whether this is bug, use case, docs, or support",2026-03-17T20:00:22Z +issue,164593,[CI] Make sure B200 benchmark testing covers regression such as https://github.com/triton-lang/triton/issues/8328,https://github.com/pytorch/pytorch/issues/164593,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: benchmark, module: ci","decide whether this is bug, use case, docs, or support",2026-03-17T19:22:55Z +issue,112344,[test/dynamo] BE: cleanup `test_misc.py`,https://github.com/pytorch/pytorch/issues/112344,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-16T19:50:06Z +issue,177393,[release 2.11][vllm] Language Models Test (Extended Generation) test_models[True-False-5-32-bigcode/starcoder2-3b],https://github.com/pytorch/pytorch/issues/177393,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: tests, module: vllm","decide whether this is bug, use case, docs, or support",2026-03-16T17:08:52Z +issue,177053,[inductor] realize_reads_threshold could be tuned differently with auto-chunking,https://github.com/pytorch/pytorch/issues/177053,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: pt2,"decide whether this is bug, use case, docs, or support",2026-03-16T15:34:24Z +issue,177520,"Memory management, memory not returned after CPU -> GPU",https://github.com/pytorch/pytorch/issues/177520,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpp, module: cuda, module: memory usage, module: windows","decide whether this is bug, use case, docs, or support",2026-03-16T14:22:57Z +issue,174481,Modular Dynamo FX Graph in torch 2.10,https://github.com/pytorch/pytorch/issues/174481,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: fx, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-16T10:47:57Z +issue,177425,binary_populate_env.sh: PYTORCH_ROOT optional check is inconsistent with later required use,https://github.com/pytorch/pytorch/issues/177425,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-03-13T21:59:10Z +issue,163974,C++ build dependencies,https://github.com/pytorch/pytorch/issues/163974,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpp, module: doc infra, module: docs","decide whether this is bug, use case, docs, or support",2026-03-13T21:35:30Z +issue,119455,Users confusing torch.distributed.nn.functional with functional collectives,https://github.com/pytorch/pytorch/issues/119455,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: c10d, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-13T19:23:05Z +issue,177386,`CatBackward0` rejects valid gradient after padded round-trip on jagged nested tensors,https://github.com/pytorch/pytorch/issues/177386,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: nestedtensor","decide whether this is bug, use case, docs, or support",2026-03-13T13:23:52Z +issue,125678,Unnecessary warning when numpy not installed,https://github.com/pytorch/pytorch/issues/125678,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: numpy","decide whether this is bug, use case, docs, or support",2026-03-13T13:11:05Z +issue,177372,Add torch.narrow_scatter operator to complete the *_copy/*_scatter operator family.,https://github.com/pytorch/pytorch/issues/177372,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functionalization, module: functorch, module: pt2-dispatcher, module: scatter & gather ops","decide whether this is bug, use case, docs, or support",2026-03-13T08:20:09Z +issue,177050,PyTorch 2.10-2.11 Linux x86 Binary size increase investigation,https://github.com/pytorch/pytorch/issues/177050,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: binaries,"decide whether this is bug, use case, docs, or support",2026-03-12T20:05:37Z +issue,177268,[OpenReg] Add testing operator framework usage examples,https://github.com/pytorch/pytorch/issues/177268,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: openreg, module: tests","decide whether this is bug, use case, docs, or support",2026-03-12T15:07:49Z +issue,136748,Inconsistent behavior of cdist with half-precision inputs,https://github.com/pytorch/pytorch/issues/136748,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: norms and normalization","decide whether this is bug, use case, docs, or support",2026-03-12T14:08:58Z +issue,122806,Deprecate accessing FakeTensor.data_ptr outside of torch.compile,https://github.com/pytorch/pytorch/issues/122806,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: faketensor,"decide whether this is bug, use case, docs, or support",2026-03-11T21:19:09Z +issue,118492,`torch.compile(dynamic=True)` compiles forever,https://github.com/pytorch/pytorch/issues/118492,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile-time, module: dynamic shapes, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-11T21:18:22Z +issue,118521,Improve constant propagation in the dynamic shape case,https://github.com/pytorch/pytorch/issues/118521,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: fx.passes","decide whether this is bug, use case, docs, or support",2026-03-11T20:56:45Z +issue,120105,Weird Artifact name: 'not_implemented' not registered being raised,https://github.com/pytorch/pytorch/issues/120105,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: logging","decide whether this is bug, use case, docs, or support",2026-03-11T20:49:01Z +issue,118211,"Dynamo codegen for mapping outputs back to user variables should contain safety checks (no fake tensors, no variable trackers)",https://github.com/pytorch/pytorch/issues/118211,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dynamo,"decide whether this is bug, use case, docs, or support",2026-03-11T20:40:11Z +issue,121818,torch._check but trust me,https://github.com/pytorch/pytorch/issues/121818,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes,"decide whether this is bug, use case, docs, or support",2026-03-11T20:39:54Z +issue,107631,torch.fx.Interpreter modules don't get compiled,https://github.com/pytorch/pytorch/issues/107631,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: graph breaks","decide whether this is bug, use case, docs, or support",2026-03-11T20:34:53Z +issue,119407,"convolution_backward has different size results on ""CPU"", FakeTensor",https://github.com/pytorch/pytorch/issues/119407,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: faketensor,"decide whether this is bug, use case, docs, or support",2026-03-11T20:33:53Z +issue,113589,Fake tensor produces incorrect values w/ is_coalesced and sparse_coo,https://github.com/pytorch/pytorch/issues/113589,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: sparse","decide whether this is bug, use case, docs, or support",2026-03-11T20:23:41Z +issue,88109,Python Dispatcher registrations beyond BackendSelect do nothing,https://github.com/pytorch/pytorch/issues/88109,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dispatch, module: python dispatcher","decide whether this is bug, use case, docs, or support",2026-03-11T20:17:22Z +issue,113643,Distinguish immutable/mutable fake tensor mode,https://github.com/pytorch/pytorch/issues/113643,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: faketensor","decide whether this is bug, use case, docs, or support",2026-03-11T20:15:04Z +issue,127772,Run existing eager DTensor tests under torch.compile,https://github.com/pytorch/pytorch/issues/127772,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-11T18:25:28Z +issue,170241,Allow for multiple stream captures to a cuda graph cudagraph at once.,https://github.com/pytorch/pytorch/issues/170241,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs,"decide whether this is bug, use case, docs, or support",2026-03-11T17:44:10Z +issue,104194,View ops on fake tensors can dispatch `detach`es to backend kernels,https://github.com/pytorch/pytorch/issues/104194,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: faketensor,"decide whether this is bug, use case, docs, or support",2026-03-11T16:13:41Z +issue,107582,[FakeTensor] `to` doesn't error with `allow_non_fake_inputs=False`,https://github.com/pytorch/pytorch/issues/107582,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: faketensor,"decide whether this is bug, use case, docs, or support",2026-03-11T15:59:16Z +issue,102187,Extend fake fast path to more situations,https://github.com/pytorch/pytorch/issues/102187,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: faketensor,"decide whether this is bug, use case, docs, or support",2026-03-11T15:47:36Z +issue,100985,"native_batch_norm has different size results on ""CPU"" vs ""META"" device",https://github.com/pytorch/pytorch/issues/100985,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: meta tensors","decide whether this is bug, use case, docs, or support",2026-03-11T15:33:58Z +issue,91655,FakeTensors not moving between device properly on Module.cuda(),https://github.com/pytorch/pytorch/issues/91655,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: faketensor,"decide whether this is bug, use case, docs, or support",2026-03-11T15:27:43Z +issue,82616,FakeTensor Support For Pickling,https://github.com/pytorch/pytorch/issues/82616,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: faketensor,"decide whether this is bug, use case, docs, or support",2026-03-11T15:17:25Z +issue,84925,`is_pinned()` support in PrimTorch and FakeTensor.,https://github.com/pytorch/pytorch/issues/84925,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: primtorch","decide whether this is bug, use case, docs, or support",2026-03-11T15:12:31Z +issue,177135,release notes commit script should respect user-specified labels,https://github.com/pytorch/pytorch/issues/177135,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-03-11T14:01:42Z +issue,176625,The `padding` type should be set as Scaler for torch.nested.to_padded_tensor (currently set to double),https://github.com/pytorch/pytorch/issues/176625,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: edge cases, module: nestedtensor","decide whether this is bug, use case, docs, or support",2026-03-11T00:05:22Z +issue,167766,Fix cudagraph flaky tests,https://github.com/pytorch/pytorch/issues/167766,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T22:12:36Z +issue,170535,Graphsafe RNG clone_state,https://github.com/pytorch/pytorch/issues/170535,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cuda graphs","decide whether this is bug, use case, docs, or support",2026-03-10T22:10:38Z +issue,175557,torch.compile crashes during autoregressive decode with growing KV cache in cudagraph_trees,https://github.com/pytorch/pytorch/issues/175557,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda graphs, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T22:06:16Z +issue,168820,Test: TestFP8Matmul.test_scaled_mm_vs_emulated_row_wise,https://github.com/pytorch/pytorch/issues/168820,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: linear algebra, module: rocm","decide whether this is bug, use case, docs, or support",2026-03-10T17:48:40Z +issue,176592,custom_op : custom_op overhead is much higher than autograd.Function (in eager),https://github.com/pytorch/pytorch/issues/176592,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: performance, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T17:18:50Z +issue,168279,Add flag to error on any Inductor specialization,https://github.com/pytorch/pytorch/issues/168279,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T16:07:13Z +issue,156805,[RFC] Support the unified memory API GetAllocator and getDeviceAllocator for in-tree and out-of-tree backend,https://github.com/pytorch/pytorch/issues/156805,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: accelerator,"decide whether this is bug, use case, docs, or support",2026-03-10T07:48:48Z +issue,113669,document functional_collectives,https://github.com/pytorch/pytorch/issues/113669,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T06:02:22Z +issue,155562,[Misc] test_foreach_add_different_mesh cannot work on machines with less than 4 GPUs,https://github.com/pytorch/pytorch/issues/155562,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-03-10T05:57:51Z +issue,108971,Really slow compilation times for torch.compile causing distributed training errors,https://github.com/pytorch/pytorch/issues/108971,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile-time, module: ddp, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T05:56:23Z +issue,145726,Add correct `__repr__` for parallel distributed modules,https://github.com/pytorch/pytorch/issues/145726,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: distributed,"decide whether this is bug, use case, docs, or support",2026-03-10T05:45:45Z +issue,152442,`torch.compile` causes assertion error in distributed checkpoint wrapper test,https://github.com/pytorch/pytorch/issues/152442,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T04:58:34Z +issue,158519,`get_model_state_dict` output inconsistent before and after `set_model_state_dict` when using `torch.compile`,https://github.com/pytorch/pytorch/issues/158519,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T04:48:32Z +issue,155632,"Functional all_gather_into_tensor does not support stacking, fails when compiled",https://github.com/pytorch/pytorch/issues/155632,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-10T04:48:30Z +issue,168823,Test: TestFP8Matmul.test_blockwise_mxfp8_nvfp4_mxfp4_numerics,https://github.com/pytorch/pytorch/issues/168823,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: linear algebra, module: rocm","decide whether this is bug, use case, docs, or support",2026-03-10T01:27:44Z +issue,169182,Kineto library not installed,https://github.com/pytorch/pytorch/issues/169182,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotinductor, module: build, module: cpp-extensions","decide whether this is bug, use case, docs, or support",2026-03-09T23:25:35Z +issue,163716,torch.compile bug because aot autograd trace ignores requires_grad,https://github.com/pytorch/pytorch/issues/163716,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-09T23:23:20Z +issue,54881,Often the documentation does not match the actual signature of the function,https://github.com/pytorch/pytorch/issues/54881,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_function__, module: docs","decide whether this is bug, use case, docs, or support",2026-03-09T23:23:12Z +issue,62982,Don't use try-catch to handle overload resolution handling in torch.ops,https://github.com/pytorch/pytorch/issues/62982,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_dispatch__, module: bootcamp","decide whether this is bug, use case, docs, or support",2026-03-09T23:23:04Z +issue,174711,add an alternative to compile time benchmarking in inductor for determinism that is reasonably performant,https://github.com/pytorch/pytorch/issues/174711,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: determinism, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-09T23:09:21Z +issue,168099,Unify pointwise DTensor and NestedTensor OP Coverage. Adds over 100 op overloads to DTensor and about to 10 to NestedTensor,https://github.com/pytorch/pytorch/issues/168099,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-03-09T20:16:34Z +issue,137505,RuntimeError: Backend nccl does not support allgather_into_tensor_coalesced,https://github.com/pytorch/pytorch/issues/137505,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: distributed,"decide whether this is bug, use case, docs, or support",2026-03-09T18:13:31Z +issue,176692,Dynamo does not know how to trace the builtin `.TensorBase._make_wrapper_subclass.`,https://github.com/pytorch/pytorch/issues/176692,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-09T16:19:17Z +issue,26207,Dispatch Tracing/Debugging,https://github.com/pytorch/pytorch/issues/26207,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dispatch,"decide whether this is bug, use case, docs, or support",2026-03-08T04:04:10Z +issue,152213,Outdated install commands,https://github.com/pytorch/pytorch/issues/152213,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: docs,"decide whether this is bug, use case, docs, or support",2026-03-07T19:28:18Z +issue,165987,`triu_/tril_` ignores memory overlaps which can lead to silent correctness issues,https://github.com/pytorch/pytorch/issues/165987,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: error checking, module: memory format","decide whether this is bug, use case, docs, or support",2026-03-07T07:07:44Z +issue,171620,Feedback about torch.autograd.graph.Node.next_functions,https://github.com/pytorch/pytorch/issues/171620,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: docs","decide whether this is bug, use case, docs, or support",2026-03-07T04:21:25Z +issue,163997,`F.scaled_dot_product_attention` returns all-`NaN` rows during evaluation but not training,https://github.com/pytorch/pytorch/issues/163997,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nans and infs, module: sdpa","decide whether this is bug, use case, docs, or support",2026-03-07T00:37:30Z +issue,133511,Do dynamo/inductor support custom classes like TorchScript.,https://github.com/pytorch/pytorch/issues/133511,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: dynamo, module: vllm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-06T23:40:51Z +issue,174949,[vllm] CUBLAS_STATUS_INVALID_VALUE in cublasGemmEx after upgrading to PyTorch 2.10,https://github.com/pytorch/pytorch/issues/174949,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cublas, module: cuda, module: vllm","decide whether this is bug, use case, docs, or support",2026-03-06T22:11:49Z +issue,176730,scaled_dot_product_attention_math usage and perf implications of returning the second value,https://github.com/pytorch/pytorch/issues/176730,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: nn, module: performance, module: sdpa","decide whether this is bug, use case, docs, or support",2026-03-06T19:03:47Z +issue,176188,test_structured_trace.py's test_ddp_graphs and test_graph_breaks flaky on all platforms,https://github.com/pytorch/pytorch/issues/176188,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: rocm,"decide whether this is bug, use case, docs, or support",2026-03-06T14:56:17Z +issue,176687,Proposal: Pytorch agent harness for external contributors,https://github.com/pytorch/pytorch/issues/176687,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-03-06T07:22:50Z +issue,176628,"[NN] Performance improvement in `nn.Module._call_impl()` by around 9% by optimizing fast-path for `forward_call()`, hook iterations and checks plus some minor changes",https://github.com/pytorch/pytorch/issues/176628,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: performance","decide whether this is bug, use case, docs, or support",2026-03-06T02:01:35Z +issue,175538,[Inductor] Need a CustomIRPass Abstraction for Loop‑Level IR Transformations,https://github.com/pytorch/pytorch/issues/175538,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-05T17:02:56Z +issue,176166,DataLoader: collate_fn leads to high RAM usage,https://github.com/pytorch/pytorch/issues/176166,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: performance, module: windows","decide whether this is bug, use case, docs, or support",2026-03-05T14:50:31Z +issue,136003,[MPS] Inconsistent performance issues,https://github.com/pytorch/pytorch/issues/136003,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: performance","decide whether this is bug, use case, docs, or support",2026-03-05T05:53:00Z +issue,101850,Unexpected modification to CPU affinity of Dataloader workers,https://github.com/pytorch/pytorch/issues/101850,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: multiprocessing, module: openmp","decide whether this is bug, use case, docs, or support",2026-03-05T04:01:15Z +issue,176507,Security hardening: Add HMAC integrity verification and authentication support to torch.compile Redis remote cache,https://github.com/pytorch/pytorch/issues/176507,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-05T00:22:39Z +issue,149809,LoadHIP.cmake should find_package(composable_kernel),https://github.com/pytorch/pytorch/issues/149809,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: rocm","decide whether this is bug, use case, docs, or support",2026-03-04T19:16:07Z +issue,176415,Output difference anomaly under equivalent migration in logdet operator,https://github.com/pytorch/pytorch/issues/176415,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: linear algebra, module: nans and infs, module: windows","decide whether this is bug, use case, docs, or support",2026-03-04T09:54:18Z +issue,176413,Output difference anomaly under equivalent migration in logdet operator,https://github.com/pytorch/pytorch/issues/176413,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: windows","decide whether this is bug, use case, docs, or support",2026-03-04T09:52:26Z +issue,176412,Output difference on 2D float32 tensors with negative out-of-domain values,https://github.com/pytorch/pytorch/issues/176412,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: nans and infs, module: windows","decide whether this is bug, use case, docs, or support",2026-03-04T09:51:34Z +issue,173885,[Inductor][CPU][float16] LayerNorm outputs NaN when input contains Inf,https://github.com/pytorch/pytorch/issues/173885,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: inductor, module: nans and infs, module: nn","decide whether this is bug, use case, docs, or support",2026-03-04T06:38:36Z +issue,169926,Failed to build PyTorch with CUDA by using clang with undefined reference of const_data_ptr in TensorBase,https://github.com/pytorch/pytorch/issues/169926,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: build, module: cuda","decide whether this is bug, use case, docs, or support",2026-03-04T06:34:33Z +issue,176385,Inconsistent cross-framework behavior (nan vs +/-inf) for float64 out-of-domain inputs,https://github.com/pytorch/pytorch/issues/176385,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: nans and infs, module: special","decide whether this is bug, use case, docs, or support",2026-03-04T05:07:51Z +issue,176384,Output difference anomaly under equivalent migration in erfinv operator,https://github.com/pytorch/pytorch/issues/176384,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: python frontend","decide whether this is bug, use case, docs, or support",2026-03-04T05:06:26Z +issue,175311,[RFC][Inductor] PyTorch Inductor cuTile Backend,https://github.com/pytorch/pytorch/issues/175311,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: helion, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-04T03:24:08Z +issue,159155,Audit all usages of constrain_to_fx_strides. Many of them might want constrain_to_fake_tensor,https://github.com/pytorch/pytorch/issues/159155,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-03T19:57:48Z +issue,172987,Leveraging Unified Memory for MPS Tensors on Apple Silicone,https://github.com/pytorch/pytorch/issues/172987,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: memory usage, module: mps","decide whether this is bug, use case, docs, or support",2026-03-03T18:08:14Z +issue,175479,Feedback about LSTM,https://github.com/pytorch/pytorch/issues/175479,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: rnn","decide whether this is bug, use case, docs, or support",2026-03-03T17:49:59Z +issue,173793,Numerical stability discrepancy in torch.compile: LayerNorm produces NaN for large inputs (~1e37) while Eager mode is stable,https://github.com/pytorch/pytorch/issues/173793,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: nans and infs, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-03T14:36:33Z +issue,154077,"accept 0-d tensor for `fill_value` parameter of `full`, `full_like`",https://github.com/pytorch/pytorch/issues/154077,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: python frontend","decide whether this is bug, use case, docs, or support",2026-03-03T14:02:36Z +issue,158371,Transition Adam and maybe other optimizers to fused code path by default to avoid `foreach=True`-specific VRAM peak due to temp TensorList for bias-corrected moments,https://github.com/pytorch/pytorch/issues/158371,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: optimizer, module: performance","decide whether this is bug, use case, docs, or support",2026-03-03T09:34:33Z +issue,175269,[dynamo] Refactor variables/builtin.py,https://github.com/pytorch/pytorch/issues/175269,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-03-02T21:48:35Z +issue,175738,"torch.export + torch.onnx.export(dynamo=True) gives incorrect results for scatter_reduce_(reduce=""mean"")",https://github.com/pytorch/pytorch/issues/175738,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: onnx,"decide whether this is bug, use case, docs, or support",2026-03-02T18:17:51Z +issue,176091,Windows CPU: SDPA FlashAttention gives wrong results with multiple intra-op threads (mixed OpenMP runtimes),https://github.com/pytorch/pytorch/issues/176091,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cpu, module: openmp, module: sdpa","decide whether this is bug, use case, docs, or support",2026-03-02T18:12:16Z +issue,176153,Libtorch - make call reset_parameters during init optional,https://github.com/pytorch/pytorch/issues/176153,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpp, module: nn","decide whether this is bug, use case, docs, or support",2026-03-02T14:10:51Z +issue,166123,"Can't compile torch-cublas-hgemm project in Windows on a torch newer than 2.8.0, I get complains involving compiled_autograd.h",https://github.com/pytorch/pytorch/issues/166123,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpp-extensions, module: windows","decide whether this is bug, use case, docs, or support",2026-03-02T10:18:48Z +issue,109592,test_memory_timeline fails on PPC due to extra temopraries,https://github.com/pytorch/pytorch/issues/109592,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: power, oncall: profiler","decide whether this is bug, use case, docs, or support",2026-03-01T20:03:11Z +issue,176078,custom_op with mutable optional arguments,https://github.com/pytorch/pytorch/issues/176078,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: python frontend,"decide whether this is bug, use case, docs, or support",2026-03-01T00:06:22Z +issue,72388,ArgMax for Multi Dimensional Tensor,https://github.com/pytorch/pytorch/issues/72388,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: reductions, module: ux","decide whether this is bug, use case, docs, or support",2026-02-28T06:21:42Z +issue,126804,"Torch fails to build with cub from CCCL v2.3.2 or newer, due to -D__CUDA_NO_HALF_OPERATORS__",https://github.com/pytorch/pytorch/issues/126804,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cuda","decide whether this is bug, use case, docs, or support",2026-02-27T06:47:53Z +issue,175899,[CI][VLLM] Track disabled vLLM tests in PyTorch CI,https://github.com/pytorch/pytorch/issues/175899,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-02-26T22:09:32Z +issue,175873,[MPS][Performance] SDPA op implementation is wasting resources by returning attention weights,https://github.com/pytorch/pytorch/issues/175873,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mps, module: performance","decide whether this is bug, use case, docs, or support",2026-02-26T18:24:58Z +issue,50335,Function Request: scipy.interpolate.interp1d,https://github.com/pytorch/pytorch/issues/50335,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: interpolation, module: numpy","decide whether this is bug, use case, docs, or support",2026-02-26T16:30:18Z +issue,175850,RFC: Vmap Extension Points for Custom Container Types,https://github.com/pytorch/pytorch/issues/175850,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functorch, module: vmap","decide whether this is bug, use case, docs, or support",2026-02-26T11:19:32Z +issue,170117,PyTorch Nightly's Torch Compile Feature Does Not Work on THOR due to Triton 3.6.x regression.,https://github.com/pytorch/pytorch/issues/170117,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: binaries, module: cuda","decide whether this is bug, use case, docs, or support",2026-02-26T08:45:49Z +issue,175829,Why the same test is running twice in regular and slow shard,https://github.com/pytorch/pytorch/issues/175829,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: tests","decide whether this is bug, use case, docs, or support",2026-02-26T04:44:08Z +issue,175737,GPUDirectStorage Status?,https://github.com/pytorch/pytorch/issues/175737,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: docs, module: serialization","decide whether this is bug, use case, docs, or support",2026-02-25T18:23:44Z +issue,174417,[release 2.11] [triton] test/inductor/test_loop_ordering.py::LoopOrderingTest::test_fp8_cast_and_t,https://github.com/pytorch/pytorch/issues/174417,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: floatx (formerly float8), module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-25T16:22:31Z +issue,174165,Fallback to Composite Implicit Autograd for Purely Functional Composite Custom Ops,https://github.com/pytorch/pytorch/issues/174165,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: custom-operators, module: dispatch, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-02-25T05:10:54Z +issue,164696,Support torch._inductor.config.inplace_buffers for custom_op whenever possible,https://github.com/pytorch/pytorch/issues/164696,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: inductor, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-25T02:43:09Z +issue,173525,[MPS] layer_norm backward numerical issues,https://github.com/pytorch/pytorch/issues/173525,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: edge cases, module: mps","decide whether this is bug, use case, docs, or support",2026-02-25T00:33:10Z +issue,146129,torch.compile on Mamba2 model produces NaNs,https://github.com/pytorch/pytorch/issues/146129,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-24T18:38:46Z +issue,174066,How to prevent operator fusion at specific points in TorchInductor?,https://github.com/pytorch/pytorch/issues/174066,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-24T16:09:26Z +issue,170558,[dynamo] Remove all Unsupported subclasses,https://github.com/pytorch/pytorch/issues/170558,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-24T04:13:42Z +issue,127882,tensor.size does not accept `dim=None` argument despite documentation because of some confusion with named dimensions,https://github.com/pytorch/pytorch/issues/127882,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-02-24T00:10:37Z +issue,174602,[nondeterminism][inconsistency] ctc_loss sometimes return incorrect results,https://github.com/pytorch/pytorch/issues/174602,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: error checking,"decide whether this is bug, use case, docs, or support",2026-02-23T18:38:08Z +issue,175193,ZLib Reference outdated in riscv ci dockerfile,https://github.com/pytorch/pytorch/issues/175193,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: infra, module: risc-v","decide whether this is bug, use case, docs, or support",2026-02-23T18:25:01Z +issue,141259,Consider relaxing torch.cond's requirement that the outputs of each branch have matching requires_grad metadata,https://github.com/pytorch/pytorch/issues/141259,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-23T05:10:33Z +issue,141784,[`FlexAttention`] Request for the Support of Dropout,https://github.com/pytorch/pytorch/issues/141784,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flex attention, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-23T04:22:40Z +issue,139518,OpOverloads are slow?,https://github.com/pytorch/pytorch/issues/139518,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-23T03:35:01Z +issue,139152,Dynamo capture of tensor.data assignment doesn't identical to eager call of tensor.data assignment,https://github.com/pytorch/pytorch/issues/139152,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T20:51:09Z +issue,138355,tlparse's fx_graph_runnable should include custom op info,https://github.com/pytorch/pytorch/issues/138355,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: fx, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T20:31:48Z +issue,138220,Update auto_functionalized_v2 to pass view subgraphs,https://github.com/pytorch/pytorch/issues/138220,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T20:25:59Z +issue,153221,FlexAttention: Support for post_mod,https://github.com/pytorch/pytorch/issues/153221,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T20:09:13Z +issue,136272,AOTDispatcher debug mode,https://github.com/pytorch/pytorch/issues/136272,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T18:33:11Z +issue,136149,Dynamo: warning when we incorrectly inline autograd.Function.backward due to subclass gradients,https://github.com/pytorch/pytorch/issues/136149,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T18:21:44Z +issue,119589,Free-floating fake tensor without the mode (FakeTensorMode),https://github.com/pytorch/pytorch/issues/119589,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: pt2-dispatcher, module: serialization, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T18:17:33Z +issue,115626,Alternatives to compiling a for loop,https://github.com/pytorch/pytorch/issues/115626,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T05:42:08Z +issue,175484,Unable to disable warning about tf32 precision,https://github.com/pytorch/pytorch/issues/175484,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T05:17:11Z +issue,122540,[RFC] Solving Warm Start in torch.compile,https://github.com/pytorch/pytorch/issues/122540,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: dynamo, module: faketensor, module: inductor","decide whether this is bug, use case, docs, or support",2026-02-22T04:56:00Z +issue,132865,aotautograd CSE interferes with Inductor reinplacing,https://github.com/pytorch/pytorch/issues/132865,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: pt2-dispatcher, module: reinplacing, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T04:53:01Z +issue,174496,Improve performance of torch._foreach_copy_ on CUDA tensors,https://github.com/pytorch/pytorch/issues/174496,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: mta, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-22T04:26:56Z +issue,164299,MPS Memory Leaks across core MPS files,https://github.com/pytorch/pytorch/issues/164299,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: mps,"decide whether this is bug, use case, docs, or support",2026-02-21T21:13:42Z +issue,175478,torch.distributions.Gamma().sample() should take a torch.Generator,https://github.com/pytorch/pytorch/issues/175478,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: determinism, module: distributions, module: random","decide whether this is bug, use case, docs, or support",2026-02-21T19:40:55Z +issue,89757,third-order gradient of torch.pow with tensor args and certain input returns NaN,https://github.com/pytorch/pytorch/issues/89757,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: nans and infs","decide whether this is bug, use case, docs, or support",2026-02-21T18:00:29Z +issue,129964,run dynamo CI shard with aot_eager_decomp_partition backend to see fallout,https://github.com/pytorch/pytorch/issues/129964,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-21T05:24:38Z +issue,117115,request: torch.compile support for tensor subclass subgraph boundaries,https://github.com/pytorch/pytorch/issues/117115,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_dispatch__, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-21T05:07:00Z +issue,123252,FakeTensor support of pin_memory,https://github.com/pytorch/pytorch/issues/123252,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: faketensor, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-21T05:02:37Z +issue,125030,"Allow operators with no returns to be ""out="" operations",https://github.com/pytorch/pytorch/issues/125030,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: custom-operators, module: pt2-dispatcher, module: reinplacing, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-21T04:53:19Z +issue,151761,Inconsistent `sum`/`dot`/`norm` behavior,https://github.com/pytorch/pytorch/issues/151761,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: linear algebra,"decide whether this is bug, use case, docs, or support",2026-02-21T04:08:27Z +issue,173452,Feature request: Support forward hooks with torch.compile fullgraph=True for capturing intermediate activations,https://github.com/pytorch/pytorch/issues/173452,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-21T00:13:40Z +issue,173255,Torch.compile is decreasing training speed instead increasing it,https://github.com/pytorch/pytorch/issues/173255,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, module: performance, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T23:43:37Z +issue,173621,Feasibility of Persistent Caching for Dynamo Frontend,https://github.com/pytorch/pytorch/issues/173621,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T23:34:58Z +issue,40770,Incremental version of pca_lowrank,https://github.com/pytorch/pytorch/issues/40770,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-02-20T22:13:52Z +issue,173160,Feature Request: load_state_dict should optionally support partial tensor filling in non-strict mode,https://github.com/pytorch/pytorch/issues/173160,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-02-20T21:03:20Z +issue,103359,Libtorch compile error when defining D_GLIBCXX_DEBUG,https://github.com/pytorch/pytorch/issues/103359,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: abi, module: build","decide whether this is bug, use case, docs, or support",2026-02-20T20:01:18Z +issue,97894,Dropout traces poorly with AotAutograd/make_fx,https://github.com/pytorch/pytorch/issues/97894,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T20:01:02Z +issue,113914,Register a torch op for `functorch.experimental.control_flow.map` to lower,https://github.com/pytorch/pytorch/issues/113914,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functorch, module: pt2-dispatcher, module: xla, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T19:56:55Z +issue,114411,higher order ops + torch.compile + __torch_dispatch__ subclasses,https://github.com/pytorch/pytorch/issues/114411,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_dispatch__, module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T19:53:44Z +issue,114413,Support sending __torch_dispatch__ subclasses in torch.compile that do not desugar,https://github.com/pytorch/pytorch/issues/114413,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_dispatch__, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T19:50:04Z +issue,114397,__torch_dispatch__ + compile: run functionalization before the subclass,https://github.com/pytorch/pytorch/issues/114397,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_dispatch__, module: functionalization, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T19:48:22Z +issue,115741,torch_function mode explodes with torch.compile,https://github.com/pytorch/pytorch/issues/115741,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_function__, module: dynamo, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T19:47:14Z +issue,106732,Hugging Face safetensor does not work with FakeTensorMode,https://github.com/pytorch/pytorch/issues/106732,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T19:40:20Z +issue,104146,make_fx: torch.where scalar promotion burns in device,https://github.com/pytorch/pytorch/issues/104146,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T18:54:38Z +issue,103650,AOT autograd: Avoid dependency on strides for manual regeneration of outputs that are aliased to inputs,https://github.com/pytorch/pytorch/issues/103650,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T18:52:46Z +issue,114412,torch.compile <> __torch_dispatch__ support tracker issue,https://github.com/pytorch/pytorch/issues/114412,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: __torch_dispatch__, module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T18:39:00Z +issue,109237,"AOTAutograd should track view chains so it can replay them, instead of using as_strided.",https://github.com/pytorch/pytorch/issues/109237,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T16:34:17Z +issue,130028,Support python attribute mutations of tensor subclass graph inputs from within __torch_dispatch__,https://github.com/pytorch/pytorch/issues/130028,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: distributed, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T16:16:49Z +issue,141349,Burn down test/functorch/test_aotdispatch.py OpInfo test failures,https://github.com/pytorch/pytorch/issues/141349,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-20T16:01:59Z +issue,175372,python property setter ignored when assigning nn.Parameter() to nn.Module,https://github.com/pytorch/pytorch/issues/175372,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-02-20T12:49:24Z +issue,23430,[Feature request] Let DistributedSampler take a Sampler as input,https://github.com/pytorch/pytorch/issues/23430,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-02-20T12:18:58Z +issue,170556,[dynamo] Add linter rule to disallow `raise Unsupported` in torch/_dynamo,https://github.com/pytorch/pytorch/issues/170556,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamo, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-19T10:09:18Z +issue,161757,[RFC] Flash decoding optimization based on FlexAttention on X86 CPU,https://github.com/pytorch/pytorch/issues/161757,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sdpa,"decide whether this is bug, use case, docs, or support",2026-02-19T08:07:21Z +issue,175306,SAC should treat inductor_compiled_code and other HOPs specially (pull out a more unique name) to prevent mixups,https://github.com/pytorch/pytorch/issues/175306,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: activation checkpointing, module: compile ux, module: higher order operators, module: pt2-dispatcher","decide whether this is bug, use case, docs, or support",2026-02-19T02:22:28Z +issue,129279,[ONNX] Create unit tests for the new export path by adapting all existing tests,https://github.com/pytorch/pytorch/issues/129279,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: onnx,"decide whether this is bug, use case, docs, or support",2026-02-18T20:01:25Z +issue,150296,[RFC] zentorch Integration,https://github.com/pytorch/pytorch/issues/150296,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-18T18:45:35Z +issue,174227,[RFC] Support physical memory accounting for VMM-based pluggable allocators,https://github.com/pytorch/pytorch/issues/174227,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: cudacachingallocator, module: memory usage, module: vllm","decide whether this is bug, use case, docs, or support",2026-02-18T16:59:51Z +issue,119930,[dtensor] ops coverage tracker,https://github.com/pytorch/pytorch/issues/119930,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dtensor, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-02-18T00:31:09Z +issue,151782,JVP: Option to Disable Gradient Caching for Tangents,https://github.com/pytorch/pytorch/issues/151782,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: functorch,"decide whether this is bug, use case, docs, or support",2026-02-17T20:49:45Z +issue,136092,User API For Donated Buffers,https://github.com/pytorch/pytorch/issues/136092,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: inductor, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-17T19:27:12Z +issue,173802,Batching rule of aten::_weight_norm,https://github.com/pytorch/pytorch/issues/173802,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: batching,"decide whether this is bug, use case, docs, or support",2026-02-16T19:21:44Z +issue,174763,"torch._grouped_mm is not autocast-compatible, causes dtype mismatch",https://github.com/pytorch/pytorch/issues/174763,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: amp (automated mixed precision), module: docs, module: linear algebra","decide whether this is bug, use case, docs, or support",2026-02-16T10:04:43Z +issue,175032,Dead code: reduction of post AI tech debt,https://github.com/pytorch/pytorch/issues/175032,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-02-14T21:25:51Z +issue,174983,Make `export CUDA_LOG_FILE=stdout` less noisy with spurious errors.,https://github.com/pytorch/pytorch/issues/174983,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: ux","decide whether this is bug, use case, docs, or support",2026-02-13T22:36:47Z +issue,166856,CPU `normal_()` Non-Deterministic Behavior with Non-Contiguous Tensors,https://github.com/pytorch/pytorch/issues/166856,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: determinism, module: random","decide whether this is bug, use case, docs, or support",2026-02-13T22:13:21Z +issue,45728,Low shared memory,https://github.com/pytorch/pytorch/issues/45728,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dataloader,"decide whether this is bug, use case, docs, or support",2026-02-13T15:55:48Z +issue,174930,UNSTABLE inductor / inductor-test-cuda13 / test (inductor_torchbench),https://github.com/pytorch/pytorch/issues/174930,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-02-13T00:50:19Z +issue,174929,UNSTABLE inductor / inductor-cpu-test / test (cpu_inductor_torchbench),https://github.com/pytorch/pytorch/issues/174929,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-02-13T00:49:37Z +issue,171568,[ROCm][Inductor] Flash/efficient attention meta kernel shape mismatch on CDNA (MI200/MI300),https://github.com/pytorch/pytorch/issues/171568,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: rocm,"decide whether this is bug, use case, docs, or support",2026-02-12T23:36:37Z +issue,173904,Unexpected cuda memory usage when doing 4d+ matmul,https://github.com/pytorch/pytorch/issues/173904,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: edge cases, module: linear algebra, module: memory usage","decide whether this is bug, use case, docs, or support",2026-02-12T21:00:20Z +issue,141958,Support SDPA flash attention/ memory efficient attn on ROCm gfx908,https://github.com/pytorch/pytorch/issues/141958,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rocm, module: sdpa","decide whether this is bug, use case, docs, or support",2026-02-12T00:45:34Z +issue,170758,bf16 layernorm has wrong answer,https://github.com/pytorch/pytorch/issues/170758,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: bfloat16, module: cpu, module: edge cases","decide whether this is bug, use case, docs, or support",2026-02-11T18:43:21Z +issue,174792,Provide a way to systematically run all the autotune configs provided by torch.compile,https://github.com/pytorch/pytorch/issues/174792,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: tests, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-11T18:36:26Z +issue,141550,RISC-V CI support,https://github.com/pytorch/pytorch/issues/141550,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: risc-v","decide whether this is bug, use case, docs, or support",2026-02-11T10:28:48Z +issue,90574,Bfloat16 tensor .numpy() support,https://github.com/pytorch/pytorch/issues/90574,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: bfloat16, module: numpy","decide whether this is bug, use case, docs, or support",2026-02-11T06:58:23Z +issue,132989,Different random seeds can cause differences in arithmetic precision,https://github.com/pytorch/pytorch/issues/132989,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: numerical-stability,"decide whether this is bug, use case, docs, or support",2026-02-11T01:52:09Z +issue,165628,[Meta] Type Hints in Pytorch,https://github.com/pytorch/pytorch/issues/165628,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: typing,"decide whether this is bug, use case, docs, or support",2026-02-10T19:16:22Z +issue,174313,[release 2.11][triton] trunk / linux-jammy-rocm-py3.10 / test failures,https://github.com/pytorch/pytorch/issues/174313,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-10T18:39:17Z +issue,122588,[ONNX] Applying consecutive conv2d layers in a for loop,https://github.com/pytorch/pytorch/issues/122588,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: onnx, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-10T13:32:02Z +issue,82098,[feature request] Discover actually loaded shared libraries at runtime,https://github.com/pytorch/pytorch/issues/82098,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: build,"decide whether this is bug, use case, docs, or support",2026-02-10T12:03:55Z +issue,135508,"BUG: `torch.cuda.is_available()` returns `False` in certain torch, CUDA and driver version",https://github.com/pytorch/pytorch/issues/135508,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: cuda, module: docker","decide whether this is bug, use case, docs, or support",2026-02-09T19:23:10Z +issue,174359,The OpenReg runtime is stuck,https://github.com/pytorch/pytorch/issues/174359,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: deadlock, module: multithreading, module: openreg","decide whether this is bug, use case, docs, or support",2026-02-09T18:18:09Z +issue,61686,[discussion] Remove the need of mandatory super() nn.Module's call,https://github.com/pytorch/pytorch/issues/61686,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-02-09T13:58:11Z +issue,174355,Forward AD not implemented for GRU cell when using CUDA,https://github.com/pytorch/pytorch/issues/174355,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: autograd, module: cuda, module: nn","decide whether this is bug, use case, docs, or support",2026-02-09T01:31:34Z +issue,126158,[torch.arange] Output mismatch with Numpy / TensorFlow,https://github.com/pytorch/pytorch/issues/126158,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: numerical-stability,"decide whether this is bug, use case, docs, or support",2026-02-07T15:34:57Z +issue,108521,"resutl of (torch.mm(a,b) does not match result of (a[:part,:], b)",https://github.com/pytorch/pytorch/issues/108521,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: numerical-stability,"decide whether this is bug, use case, docs, or support",2026-02-07T15:13:45Z +issue,139964,Report issue for torch.nn.Linear when forwarding a 3-dim tensor.,https://github.com/pytorch/pytorch/issues/139964,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: numerical-stability,"decide whether this is bug, use case, docs, or support",2026-02-07T15:13:04Z +issue,76528,Discrepancy in einsum when done in batch vs non-batch,https://github.com/pytorch/pytorch/issues/76528,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: linear algebra, module: tf32","decide whether this is bug, use case, docs, or support",2026-02-07T07:59:57Z +issue,76632,RestrictPtrTraits in CUDA potentially has no effect.,https://github.com/pytorch/pytorch/issues/76632,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-02-07T07:57:09Z +issue,171398,ISAAC inference,https://github.com/pytorch/pytorch/issues/171398,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: rocm","decide whether this is bug, use case, docs, or support",2026-02-05T23:27:40Z +issue,124275,Performance Impact of `torch.autograd.profiler.record_function` in DataLoader's `_BaseDataLoaderIter` Iteration,https://github.com/pytorch/pytorch/issues/124275,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: performance","decide whether this is bug, use case, docs, or support",2026-02-05T20:55:13Z +issue,143887,`torch.accelerator` cross-device utilities and properties,https://github.com/pytorch/pytorch/issues/143887,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: accelerator,"decide whether this is bug, use case, docs, or support",2026-02-05T20:11:09Z +issue,159520,[RFC] Enable native Windows CI/CD on ROCm,https://github.com/pytorch/pytorch/issues/159520,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: rocm","decide whether this is bug, use case, docs, or support",2026-02-05T17:44:34Z +issue,174266,BatchNorm1d CPU outputs differ across devices/threads with identical inputs,https://github.com/pytorch/pytorch/issues/174266,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: edge cases, module: nn","decide whether this is bug, use case, docs, or support",2026-02-05T00:26:46Z +issue,172320,"[OpenReg] Extend device infrastructure tests (device, event, streams)",https://github.com/pytorch/pytorch/issues/172320,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: openreg,"decide whether this is bug, use case, docs, or support",2026-02-04T15:14:00Z +issue,150974,feature: tlparse summary,https://github.com/pytorch/pytorch/issues/150974,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: logging, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-04T12:56:50Z +issue,159516,Validate PyPI Wheel Binary Size failing for >3months,https://github.com/pytorch/pytorch/issues/159516,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: releng,"decide whether this is bug, use case, docs, or support",2026-02-04T04:31:01Z +issue,174220,Dr CI produces confusing issue attribution report,https://github.com/pytorch/pytorch/issues/174220,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: devx","decide whether this is bug, use case, docs, or support",2026-02-03T18:36:52Z +issue,119461,Promote pytorch nightly docker image,https://github.com/pytorch/pytorch/issues/119461,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: docker","decide whether this is bug, use case, docs, or support",2026-02-03T16:12:29Z +issue,79197,[feature request] Support dataclass derivations of nn.Module,https://github.com/pytorch/pytorch/issues/79197,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-02-03T15:26:07Z +issue,109204,Pytorch ROCM windows builds,https://github.com/pytorch/pytorch/issues/109204,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rocm, module: windows","decide whether this is bug, use case, docs, or support",2026-02-03T14:49:49Z +issue,137574,[feature request] Provide FlexAttention as a new available/selectable backend for SDPA,https://github.com/pytorch/pytorch/issues/137574,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sdpa,"decide whether this is bug, use case, docs, or support",2026-02-03T13:35:39Z +issue,119196,Possible NCCL race condition `destroy_process_group` followed by `init_process_group`.,https://github.com/pytorch/pytorch/issues/119196,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nccl, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-02-03T01:04:49Z +issue,154444,[RFC] Removing ideep git submodule dependency from PyTorch & add oneDNN git submodule,https://github.com/pytorch/pytorch/issues/154444,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: intel, module: third_party","decide whether this is bug, use case, docs, or support",2026-02-03T00:11:59Z +issue,174081,"[vllm] [release 2.10] V1 e2e + engine test_ngram_and_suffix_correctness[speculative_config1], test_draft_model_realistic_example",https://github.com/pytorch/pytorch/issues/174081,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: vllm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-02-02T19:21:45Z +issue,174088,Gold doesn't work with GB200,https://github.com/pytorch/pytorch/issues/174088,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: build, module: cuda","decide whether this is bug, use case, docs, or support",2026-02-02T17:38:51Z +issue,163408,RFC: Inherit some modules from torchtune.,https://github.com/pytorch/pytorch/issues/163408,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-02-02T09:28:12Z +issue,133559,Support for variable number of dimensions in functorch.dim dims() and autonaming induced inconsistencies,https://github.com/pytorch/pytorch/issues/133559,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: first class dims, module: functorch","decide whether this is bug, use case, docs, or support",2026-02-01T16:20:34Z +issue,137093,Use SymInt instead of sympy.Expr uniformly in Inductor,https://github.com/pytorch/pytorch/issues/137093,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-31T18:23:40Z +issue,125254,AssertionError: Device mps not supported,https://github.com/pytorch/pytorch/issues/125254,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: mps, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-30T23:32:56Z +issue,171558,`torch.sparse.spdiags` produces incorrect results,https://github.com/pytorch/pytorch/issues/171558,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: sparse,"decide whether this is bug, use case, docs, or support",2026-01-30T21:54:46Z +issue,173950,[CI] enabling CUDA 13.0 periodic test,https://github.com/pytorch/pytorch/issues/173950,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: ci,"decide whether this is bug, use case, docs, or support",2026-01-30T20:30:22Z +issue,163327,[MPS] scatter_add_ is incorrect on MPS with non-zero offset slices,https://github.com/pytorch/pytorch/issues/163327,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: mps","decide whether this is bug, use case, docs, or support",2026-01-30T15:50:21Z +issue,173638,`torch.linalg.slogdet` does not propagate `NaN` in CUDA,https://github.com/pytorch/pytorch/issues/173638,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: correctness (silent), module: cuda, module: linear algebra, module: nans and infs","decide whether this is bug, use case, docs, or support",2026-01-29T22:32:11Z +issue,168393,"`register_autograd` broken for `torch.func`, fails to register `setup_context`",https://github.com/pytorch/pytorch/issues/168393,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functorch, module: vmap","decide whether this is bug, use case, docs, or support",2026-01-29T21:24:42Z +issue,173796,`torch.nn.functional.PoissonNLLLoss` returns `NaN` imaginary part on CPU (vs 0 on GPU) for overflowing complex inputs,https://github.com/pytorch/pytorch/issues/173796,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nans and infs, module: nn","decide whether this is bug, use case, docs, or support",2026-01-29T20:53:16Z +issue,173661,[Perfomance issue] Dataloader has unexpected overhead,https://github.com/pytorch/pytorch/issues/173661,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: dataloader, module: performance","decide whether this is bug, use case, docs, or support",2026-01-29T17:41:00Z +issue,34242,RuntimeError with torch.unique: radix_sort: failed on 2nd step: invalid argument,https://github.com/pytorch/pytorch/issues/34242,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cuda, module: sorting and selection","decide whether this is bug, use case, docs, or support",2026-01-29T09:17:46Z +issue,133702,"Add `beta`, `betainc`, etc. to `torch.special`",https://github.com/pytorch/pytorch/issues/133702,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: distributions,"decide whether this is bug, use case, docs, or support",2026-01-29T05:41:00Z +issue,173642,[Feature Request][torch.compile] Improve UX when autotuning repeated layer,https://github.com/pytorch/pytorch/issues/173642,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: compile ux, module: dynamo, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-29T00:51:49Z +issue,173729,Way to request a non-aliased torch.cuda.Stream,https://github.com/pytorch/pytorch/issues/173729,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-01-28T23:01:06Z +issue,173514,Improve documentation of assumptions for symmetric memory,https://github.com/pytorch/pytorch/issues/173514,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: symm_mem","decide whether this is bug, use case, docs, or support",2026-01-28T22:37:30Z +issue,112398,[Tracker] Move nested tensors to beta,https://github.com/pytorch/pytorch/issues/112398,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nestedtensor,"decide whether this is bug, use case, docs, or support",2026-01-28T22:28:59Z +issue,108794,[C++ Frontend] Simple Changes for Cleaner Options,https://github.com/pytorch/pytorch/issues/108794,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cpp,"decide whether this is bug, use case, docs, or support",2026-01-28T14:54:34Z +issue,173516,[CUDA] MaxPool2d CUDA kernel lacks int64 support → one-side error,https://github.com/pytorch/pytorch/issues/173516,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-01-27T23:37:05Z +issue,76344,"[ONNX] Use topk to export max(dim,keepdim) to onnx",https://github.com/pytorch/pytorch/issues/76344,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: onnx,"decide whether this is bug, use case, docs, or support",2026-01-27T16:21:25Z +issue,170419,Cannot set version_counter for inference tensor,https://github.com/pytorch/pytorch/issues/170419,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: python frontend,"decide whether this is bug, use case, docs, or support",2026-01-27T13:35:05Z +issue,95487,Torch RPC on multiple nodes with GPU returns a EOF error,https://github.com/pytorch/pytorch/issues/95487,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: rpc, module: tensorpipe, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-01-27T13:06:21Z +issue,171375,aten.lift triggers internal assert failure in Inductor functionalization pass,https://github.com/pytorch/pytorch/issues/171375,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-27T09:23:46Z +issue,173104,Crash with `CUDACachingAllocator.cpp` when running Nvidia `nemotron-parse` at load,https://github.com/pytorch/pytorch/issues/173104,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cudacachingallocator,"decide whether this is bug, use case, docs, or support",2026-01-26T16:33:25Z +issue,173290,Potential issue with lt_two in DistanceKernel.cu,https://github.com/pytorch/pytorch/issues/173290,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-01-26T16:30:30Z +issue,172886,libtorch_cpu.so have unresolved symbols,https://github.com/pytorch/pytorch/issues/172886,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: mpi","decide whether this is bug, use case, docs, or support",2026-01-25T19:02:42Z +issue,141288,Enhancement Request: improve this error message,https://github.com/pytorch/pytorch/issues/141288,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-01-24T20:42:54Z +issue,133367,Adafactor foreach impl performance tracker,https://github.com/pytorch/pytorch/issues/133367,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: mta, module: optimizer, module: performance","decide whether this is bug, use case, docs, or support",2026-01-23T17:00:57Z +issue,173059,pandas/numpy binary incompatibility breaks test_dataloader and test_datapipe,https://github.com/pytorch/pytorch/issues/173059,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-01-22T23:25:56Z +issue,119328,[pytree] Handling of `None` in torch.utils._pytree is inconsistent with JAX.,https://github.com/pytorch/pytorch/issues/119328,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: pytree,"decide whether this is bug, use case, docs, or support",2026-01-22T21:28:30Z +issue,171283,aten.uniform cannot be called with keyword argument from,https://github.com/pytorch/pytorch/issues/171283,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributions, module: docs, module: python frontend","decide whether this is bug, use case, docs, or support",2026-01-22T04:09:22Z +issue,172854,[test] Add unittest for nanmean output dtype validation (follow-up to #131043),https://github.com/pytorch/pytorch/issues/172854,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: reductions, module: tests","decide whether this is bug, use case, docs, or support",2026-01-21T23:01:54Z +issue,151705,Enable TorchInductor to Generate Matmuls Natively via `tl.dot`,https://github.com/pytorch/pytorch/issues/151705,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-21T19:59:07Z +issue,172797,autograd.Function.apply yields surprising strides,https://github.com/pytorch/pytorch/issues/172797,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: autograd,"decide whether this is bug, use case, docs, or support",2026-01-21T18:51:06Z +issue,85366,More windows for filtering and spectral analysis,https://github.com/pytorch/pytorch/issues/85366,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: scipy compatibility,"decide whether this is bug, use case, docs, or support",2026-01-21T17:33:29Z +issue,159444,#define CHECK causes issues with a bunch of other libraries,https://github.com/pytorch/pytorch/issues/159444,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: build, module: cpp","decide whether this is bug, use case, docs, or support",2026-01-21T16:18:46Z +issue,92250,torch.cuda.is_available() returns True even if the CUDA hardware can't run pytorch,https://github.com/pytorch/pytorch/issues/92250,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: cuda,"decide whether this is bug, use case, docs, or support",2026-01-21T08:50:37Z +issue,171804,[On Demand Profiling] Early CUDA initialization caused by `import torch` when `KINETO_USE_DAEMON=1`,https://github.com/pytorch/pytorch/issues/171804,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: oncall: distributed, oncall: profiler","decide whether this is bug, use case, docs, or support",2026-01-21T03:42:29Z +issue,162780,Prebuilt Wheels supporting T4G,https://github.com/pytorch/pytorch/issues/162780,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: binaries, module: cuda","decide whether this is bug, use case, docs, or support",2026-01-20T23:12:34Z +issue,147662,GPUDirect Storage dcp.save/dcp.load torch.save/torch.load,https://github.com/pytorch/pytorch/issues/147662,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: serialization, oncall: distributed checkpointing","decide whether this is bug, use case, docs, or support",2026-01-20T18:28:08Z +issue,74282,RuntimeError: cuDNN error: CUDNN_STATUS_VERSION_MISMATCH for torchvision models,https://github.com/pytorch/pytorch/issues/74282,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: cudnn, module: vision","decide whether this is bug, use case, docs, or support",2026-01-20T18:27:16Z +issue,168573,TestModule: test/inductor/test_codecache.py::TestFxGraphCache,https://github.com/pytorch/pytorch/issues/168573,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: rocm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-20T15:22:03Z +issue,144634,[MPS] `torch.mps.synchronize` hangs on error,https://github.com/pytorch/pytorch/issues/144634,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: deadlock, module: mps, module: third_party","decide whether this is bug, use case, docs, or support",2026-01-20T15:19:44Z +issue,68110,zeta(1/2) returns nan,https://github.com/pytorch/pytorch/issues/68110,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: special,"decide whether this is bug, use case, docs, or support",2026-01-20T13:56:44Z +issue,172803,[OpenReg] PrivateUse1 backend missing distributed backend registration causes KeyError,https://github.com/pytorch/pytorch/issues/172803,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: openreg, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-01-20T09:50:34Z +issue,140458,Making nn.Module generic,https://github.com/pytorch/pytorch/issues/140458,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nn, module: typing","decide whether this is bug, use case, docs, or support",2026-01-19T14:34:25Z +issue,170743,Reference count modifications led to performance drop,https://github.com/pytorch/pytorch/issues/170743,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: cpu, module: performance","decide whether this is bug, use case, docs, or support",2026-01-19T02:51:00Z +issue,76580,Random Generator for Dropout,https://github.com/pytorch/pytorch/issues/76580,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: random,"decide whether this is bug, use case, docs, or support",2026-01-19T01:30:06Z +issue,149605,torch.export does not support nested tensor operations,https://github.com/pytorch/pytorch/issues/149605,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: onnx, oncall: export, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-18T22:39:19Z +issue,172623,Nested Tensors should have their own type,https://github.com/pytorch/pytorch/issues/172623,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: nestedtensor, module: typing","decide whether this is bug, use case, docs, or support",2026-01-18T08:17:35Z +issue,172475,[Inductor] Track deterministic `alloc_id` assignment across cached compilation artifacts,https://github.com/pytorch/pytorch/issues/172475,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: determinism, module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-18T04:02:06Z +issue,172213,Eager buffer mutation after `torch.compile` + backward + `torch.export.export` hits `_map_assigned_buffer_to_proxy` `FakeTensor` assertion,https://github.com/pytorch/pytorch/issues/172213,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-18T04:01:24Z +issue,172632,AArch64 GCC14 - inductor/test_mkldnn_pattern_matcher.py FAIL - internal compiler error: Segmentation fault,https://github.com/pytorch/pytorch/issues/172632,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: build, module: third_party","decide whether this is bug, use case, docs, or support",2026-01-16T18:51:37Z +issue,172445,AArch64 GCC14 Issues Tracker,https://github.com/pytorch/pytorch/issues/172445,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: build, module: third_party","decide whether this is bug, use case, docs, or support",2026-01-16T12:38:15Z +issue,164281,Document Symmetric Memory,https://github.com/pytorch/pytorch/issues/164281,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-01-16T05:41:21Z +issue,172351,[RFC][Release-2.11] Dropping Volta support from CUDA-12.8,https://github.com/pytorch/pytorch/issues/172351,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: cuda, module: cudnn","decide whether this is bug, use case, docs, or support",2026-01-15T23:44:05Z +issue,172549,"AOTAutograd error when compiled graph has two DTensor outputs that alias each other, one differentiable one not",https://github.com/pytorch/pytorch/issues/172549,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: pt2-dispatcher,"decide whether this is bug, use case, docs, or support",2026-01-15T18:52:50Z +issue,169672,torch while_loop does not avoid CPU synchronization,https://github.com/pytorch/pytorch/issues/169672,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-15T18:18:52Z +issue,99082,Look into test coverage for `UntypedStorage`,https://github.com/pytorch/pytorch/issues/99082,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: tests, module: typing","decide whether this is bug, use case, docs, or support",2026-01-15T15:41:48Z +issue,165398,Skipping combo kernel for large pointwise nodes,https://github.com/pytorch/pytorch/issues/165398,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, module: vllm, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-15T15:12:19Z +issue,97892,A parameterized fill value for triu and tril functions,https://github.com/pytorch/pytorch/issues/97892,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: viewing and reshaping,"decide whether this is bug, use case, docs, or support",2026-01-15T06:38:24Z +issue,172522,Support out-of-graph mutations with side streams,https://github.com/pytorch/pytorch/issues/172522,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: oncall: pt2,"decide whether this is bug, use case, docs, or support",2026-01-15T01:05:18Z +issue,167881,[RFC] Modifying Getting Started Page for XPU support,https://github.com/pytorch/pytorch/issues/167881,needs maintainer decision,28,open issue has no stronger automated bucket,"decide whether this is bug, use case, docs, or support",2026-01-14T21:32:13Z +issue,172495,Add T5Gemma2 to TorchInductor benchmark HF suite,https://github.com/pytorch/pytorch/issues/172495,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: benchmark, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-14T21:13:59Z +issue,136532,Wrong results when sampling from the beta distribution for small alpha=beta,https://github.com/pytorch/pytorch/issues/136532,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: distributions, module: numerical-stability","decide whether this is bug, use case, docs, or support",2026-01-14T19:53:08Z +issue,146780,TypeError when using torch.compile with RegionViT under torch.inference_mode(),https://github.com/pytorch/pytorch/issues/146780,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-14T19:52:53Z +issue,172427,[CI] Permission denied while trying to connect to the docker API,https://github.com/pytorch/pytorch/issues/172427,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: ci, module: infra","decide whether this is bug, use case, docs, or support",2026-01-14T16:16:54Z +issue,172273,Numeric difference between div and foreach_div,https://github.com/pytorch/pytorch/issues/172273,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: mta,"decide whether this is bug, use case, docs, or support",2026-01-14T15:01:12Z +issue,172338,Fix: Support is_causal=True with need_weights=True in Multi-Head Attention,https://github.com/pytorch/pytorch/issues/172338,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-01-14T14:55:14Z +issue,172383,torch.func.linearize produces NaNs with autocast + LayerNorm (jvp works),https://github.com/pytorch/pytorch/issues/172383,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: amp (automated mixed precision), module: functorch","decide whether this is bug, use case, docs, or support",2026-01-14T14:51:58Z +issue,171376,PyTorch ThroughputBenchmark.add_input crashes when split_sizes is incompatible with input tensor shape,https://github.com/pytorch/pytorch/issues/171376,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: benchmark,"decide whether this is bug, use case, docs, or support",2026-01-14T08:00:27Z +issue,38035,Quantile Regression Loss,https://github.com/pytorch/pytorch/issues/38035,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: nn,"decide whether this is bug, use case, docs, or support",2026-01-14T03:06:31Z +issue,171759,[ONNX] Create a tutorial,https://github.com/pytorch/pytorch/issues/171759,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: onnx,"decide whether this is bug, use case, docs, or support",2026-01-13T17:36:32Z +issue,172322,[OpenReg] Extend test_rng.py and test_misc.py coverage,https://github.com/pytorch/pytorch/issues/172322,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: openreg,"decide whether this is bug, use case, docs, or support",2026-01-13T15:08:06Z +issue,172319,[OpenReg] Add OpInfo-based operator tests using op_db,https://github.com/pytorch/pytorch/issues/172319,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: openreg,"decide whether this is bug, use case, docs, or support",2026-01-13T15:07:44Z +issue,172318,[OpenReg] Add DeviceTypeTestBase example with instantiate_device_type_tests,https://github.com/pytorch/pytorch/issues/172318,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: openreg,"decide whether this is bug, use case, docs, or support",2026-01-13T15:07:28Z +issue,169597,Standardize Testing in OpenReg,https://github.com/pytorch/pytorch/issues/169597,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: openreg, module: privateuse1","decide whether this is bug, use case, docs, or support",2026-01-13T08:44:19Z +issue,172059,Add Tweedie distribution to torch.distributions,https://github.com/pytorch/pytorch/issues/172059,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: distributions,"decide whether this is bug, use case, docs, or support",2026-01-12T22:30:18Z +issue,143471,NFS errors during DataLoader shutdown when num_workers > 1 when temporary directory is on NFS,https://github.com/pytorch/pytorch/issues/143471,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: data,"decide whether this is bug, use case, docs, or support",2026-01-12T21:42:08Z +issue,171194,CUDA Smoke tests: Extend smoke tests to include profiler coverage,https://github.com/pytorch/pytorch/issues/171194,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: binaries, module: ci, module: cuda","decide whether this is bug, use case, docs, or support",2026-01-12T20:07:09Z +issue,17023,Download speed issues with the pytorch conda channel,https://github.com/pytorch/pytorch/issues/17023,needs maintainer decision,28,open issue has no stronger automated bucket; component/oncall labels: module: dependency bug,"decide whether this is bug, use case, docs, or support",2026-01-12T14:09:06Z +issue,171501,Several Windows-related GitHub Actions not running — are they intentionally disabled?,https://github.com/pytorch/pytorch/issues/171501,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: arm, module: ci, module: windows","decide whether this is bug, use case, docs, or support",2026-01-12T07:14:05Z +issue,164455,AMP + Torch Compile bug,https://github.com/pytorch/pytorch/issues/164455,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: amp (automated mixed precision), module: inductor, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-11T18:29:11Z +issue,171866,Why are there three run_repro functions implemented in pytorch?,https://github.com/pytorch/pytorch/issues/171866,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: docs, module: minifier, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-10T19:34:32Z +issue,167062,How to use torch.compile on Windows GPU?,https://github.com/pytorch/pytorch/issues/167062,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: windows, oncall: pt2","decide whether this is bug, use case, docs, or support",2026-01-10T18:07:57Z +issue,171796,Exception in subprocess causes infinite hang and zombie processes,https://github.com/pytorch/pytorch/issues/171796,needs maintainer decision,28,"open issue has no stronger automated bucket; component/oncall labels: module: multiprocessing, oncall: distributed","decide whether this is bug, use case, docs, or support",2026-01-10T11:50:51Z +pr,178513,Fix interpolate nearest mode with non-integer scale_factor,https://github.com/pytorch/pytorch/pull/178513,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2026-05-24T17:45:53Z +pr,181073,Validate norm_type in lp_pool1d/2d/3d,https://github.com/pytorch/pytorch/pull/181073,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-20T21:44:52Z +pr,176893,Add save_state_dict and load_state_dict functions (fixes #150857),https://github.com/pytorch/pytorch/pull/176893,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-20T13:58:06Z +pr,175537,Fix: Exclude third-party headers from PyTorch wheels,https://github.com/pytorch/pytorch/pull/175537,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-20T13:56:32Z +pr,174882,Fix torch.histc float16 precision inconsistency,https://github.com/pytorch/pytorch/pull/174882,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-20T13:55:21Z +pr,179936,[MPS] Fix avg_pool2d precision by using Metal kernel instead of MPSGraph,https://github.com/pytorch/pytorch/pull/179936,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/mps,review when ready,2026-05-20T08:15:02Z +pr,184349,Improve complex input error for BatchNorm,https://github.com/pytorch/pytorch/pull/184349,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-19T17:21:33Z +pr,184058,nn.functional: raise ValueError for tau <= 0 in gumbel_softmax,https://github.com/pytorch/pytorch/pull/184058,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-18T21:44:52Z +pr,183522,Fix using `torch.cuda.amp. autocast` with JIT on Python 3.13,https://github.com/pytorch/pytorch/pull/183522,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-18T21:32:58Z +pr,180420,linalg.qr: reject overlapping or aliased Q and R in out=,https://github.com/pytorch/pytorch/pull/180420,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-18T17:34:49Z +pr,182348,[SymmMem] Implement StoreExchange::broadcast for multicast handle exchange,https://github.com/pytorch/pytorch/pull/182348,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/h100-symm-mem,review when ready,2026-05-15T07:08:50Z +pr,183367,Avoid initializing CUDA in mem_get_info,https://github.com/pytorch/pytorch/pull/183367,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-14T16:06:37Z +pr,182803,Fix torch.trapz documentation signature,https://github.com/pytorch/pytorch/pull/182803,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-14T16:03:57Z +pr,182609,Fix inconsistent CPU/CUDA casting for fmax/fmin/maximum/minimum with narrower out dtype,https://github.com/pytorch/pytorch/pull/182609,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-14T16:02:23Z +pr,180748,Fix NumPy 2 deprecation warning in __array_wrap__,https://github.com/pytorch/pytorch/pull/180748,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-13T18:32:12Z +pr,183052,Fix OOM error suggesting expandable_segments when already enabled,https://github.com/pytorch/pytorch/pull/183052,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-12T05:42:23Z +pr,181052,Add docs for torch version information,https://github.com/pytorch/pytorch/pull/181052,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-11T21:39:44Z +pr,181367,[CPU] Fix fp16 matmul ~100x slower than fp32 on CPUs without AVX512-FP16,https://github.com/pytorch/pytorch/pull/181367,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-11T20:54:39Z +pr,183039,[c10d] Remove false-positive DeviceMesh pickle warning in async checkpoint,https://github.com/pytorch/pytorch/pull/183039,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,review when ready,2026-05-11T18:56:36Z +pr,159862,Fix skipIfXpu and skipIfHpu and similar skip decorators,https://github.com/pytorch/pytorch/pull/159862,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-11T11:21:05Z +pr,167306,Fix integer overflow in (fake) quantization,https://github.com/pytorch/pytorch/pull/167306,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-11T08:45:02Z +pr,152828,[MSVC] Remove redundant MEM_EFF_ATTENTION_CUDA_SOURCES definition as /Zc:lambda is set globally for C++20 by default,https://github.com/pytorch/pytorch/pull/152828,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-05-10T18:50:12Z +pr,173644,[TORCH] [JIT] Change emitDefautlArgsWithOutArgs -> emitDefaultArgsWithOutArgs in export_module.cpp,https://github.com/pytorch/pytorch/pull/173644,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-04-12T17:26:12Z +pr,127702,Define PY_SSIZE_T_CLEAN before #include ,https://github.com/pytorch/pytorch/pull/127702,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2026-04-07T17:02:40Z +pr,154395,Updated padding validation in max_pool functions to account for dilation,https://github.com/pytorch/pytorch/pull/154395,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2025-12-05T16:55:38Z +pr,162130,Raise error when no record on `extra_files`,https://github.com/pytorch/pytorch/pull/162130,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2025-11-17T15:00:50Z +pr,152104,Update _torch_docs.py to Fix torch.bernoulli(),https://github.com/pytorch/pytorch/pull/152104,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2025-09-26T04:01:01Z +pr,142160,Introduce int32 index_fill and index_copy indices,https://github.com/pytorch/pytorch/pull/142160,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2025-09-04T19:12:52Z +pr,102148,Add mps support for maxpool3d,https://github.com/pytorch/pytorch/pull/102148,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/mps,review when ready,2025-07-30T18:42:18Z +pr,147365,Replace `fw_metadata` info with trace log hint in hint message,https://github.com/pytorch/pytorch/pull/147365,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days,review when ready,2025-07-28T17:34:49Z +pr,88998,Fix fake tensor propagation for nvprims,https://github.com/pytorch/pytorch/pull/88998,has linked issue,27,PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2023-04-17T00:36:00Z +issue,158007,"`torch.compile` cannot compile a model with a basic `LSTM`, even on latest main",https://github.com/pytorch/pytorch/issues/158007,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: rnn, oncall: pt2",review linked PR state before touching issue,2026-06-15T17:05:44Z +issue,185650,[Bug] torch.kron fails with RuntimeError on non-contiguous inputs,https://github.com/pytorch/pytorch/issues/185650,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: viewing and reshaping",review linked PR state before touching issue,2026-06-06T16:22:24Z +issue,137440,"`torch.pow` throws with `RuntimeError: ""reciprocal_cuda"" not implemented for 'Char'`",https://github.com/pytorch/pytorch/issues/137440,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: core aten, module: cuda",review linked PR state before touching issue,2026-06-01T06:48:43Z +issue,184280,GroupNorm with num_groups=0 raises ZeroDivisionError instead of ValueError,https://github.com/pytorch/pytorch/issues/184280,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn, module: norms and normalization",review linked PR state before touching issue,2026-05-22T15:56:07Z +issue,177827,Internal Assert Failure in `torch.cuda.memory.caching_allocator_alloc` with negative allocation size,https://github.com/pytorch/pytorch/issues/177827,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",review linked PR state before touching issue,2026-03-23T17:31:25Z +issue,127111,[dynamo] Detect mkldnn_max_pool2d/mkldnn view errors at tracing time,https://github.com/pytorch/pytorch/issues/127111,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: faketensor, oncall: pt2",review linked PR state before touching issue,2026-03-11T21:57:17Z +issue,140171,Compiling `as_strided` call on a view input errors.,https://github.com/pytorch/pytorch/issues/140171,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2026-02-23T04:02:32Z +issue,171981,torch.fx.Tracer.trace fails on wrapped forward methods with arbitrary arguments. ValueError: code: co_varnames is too small,https://github.com/pytorch/pytorch/issues/171981,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: fx,review linked PR state before touching issue,2026-01-08T17:09:11Z +issue,120375,Dynamo doesn't generate resume calls after graph breaking on log calls,https://github.com/pytorch/pytorch/issues/120375,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-01-07T08:15:39Z +issue,158384,[dynamo] full internal stack trace being shown with `torch._dynamo.exc.BackendCompilerFailed` with suggestion to show more,https://github.com/pytorch/pytorch/issues/158384,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2026-01-06T19:02:18Z +issue,171356,`torch.clip` has checks for float16 scalar overflow on CPU but not on GPU,https://github.com/pytorch/pytorch/issues/171356,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking",review linked PR state before touching issue,2026-01-05T17:23:02Z +issue,170781,meta_model.to_empty(...) equivalent for FakeTensor,https://github.com/pytorch/pytorch/issues/170781,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-29T05:02:38Z +issue,168974,Fuzzy mode for DebugMode.check_hash_mismatches,https://github.com/pytorch/pytorch/issues/168974,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-28T22:07:02Z +issue,170770,FakeTensorMode induced Cannot swap t1 because it has weakref associated with it,https://github.com/pytorch/pytorch/issues/170770,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-28T04:48:47Z +issue,137068,UserWarning: Graph break due to unsupported builtin time.time,https://github.com/pytorch/pytorch/issues/137068,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-12-28T04:34:13Z +issue,170834,Generated torch.Library autograd functions seem not have `setup_context` which breaks `torch.func.grad` usage,https://github.com/pytorch/pytorch/issues/170834,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-23T16:07:09Z +issue,169769,pytorch while_loop / cond overly strict about memory aliasing,https://github.com/pytorch/pytorch/issues/169769,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-23T12:59:16Z +issue,170955,`grouped_mm` does grouping on CPU on 5060 Ti (Windows),https://github.com/pytorch/pytorch/issues/170955,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cuda,review linked PR state before touching issue,2025-12-22T18:17:05Z +issue,170370,[MPS][CUDA] EmbeddingBag on CUDA/MPS does not check offsets[0] like CPU,https://github.com/pytorch/pytorch/issues/170370,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: docs, module: edge cases",review linked PR state before touching issue,2025-12-20T00:52:46Z +issue,169779,[Inductor] Inconsistent handling of negative indices in index_select: CPU/CUDA/MPS,https://github.com/pytorch/pytorch/issues/169779,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: decompositions, module: python frontend, oncall: pt2",review linked PR state before touching issue,2025-12-19T11:21:35Z +issue,32864,[docs] Missing docs for torch.__version__ and torch.version,https://github.com/pytorch/pytorch/issues/32864,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs,review linked PR state before touching issue,2025-12-18T19:49:28Z +issue,170298,"`resize_` updates tensor shape metadata even when storage resize fails, creating corrupted ""Zombie"" tensors",https://github.com/pytorch/pytorch/issues/170298,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,review linked PR state before touching issue,2025-12-18T15:38:37Z +issue,136586,support FakeTensor input for torch.compile,https://github.com/pytorch/pytorch/issues/136586,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-18T15:12:09Z +issue,141790,AOTAutograd cache always misses even on simple program,https://github.com/pytorch/pytorch/issues/141790,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, oncall: pt2",review linked PR state before touching issue,2025-12-09T18:57:48Z +issue,160739,ConvTranspose 3D with BF16 or FP16 types is not supported on MPS,https://github.com/pytorch/pytorch/issues/160739,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: mps,review linked PR state before touching issue,2025-12-08T06:13:37Z +issue,169099,[RFC] Implement StoreExchange.broadcast,https://github.com/pytorch/pytorch/issues/169099,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: symm_mem, oncall: distributed",review linked PR state before touching issue,2025-12-05T07:03:40Z +issue,130182,autograd.Function HOP backward input names could be better,https://github.com/pytorch/pytorch/issues/130182,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, oncall: pt2",review linked PR state before touching issue,2025-12-05T01:39:54Z +issue,165385,[dynamo] Invalid setattr's are not handled properly in general,https://github.com/pytorch/pytorch/issues/165385,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-12-03T01:53:30Z +issue,168307,torch.compile error when creating nested tensors dynamically in forward pass,https://github.com/pytorch/pytorch/issues/168307,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: nestedtensor, oncall: pt2",review linked PR state before touching issue,2025-12-02T23:57:02Z +issue,162415,[dynamo] Weird interaction between fullgraph=True and skipped frames,https://github.com/pytorch/pytorch/issues/162415,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-12-02T23:48:04Z +issue,169208,Segmentation fault in torch.native_batch_norm with empty running_var tensor,https://github.com/pytorch/pytorch/issues/169208,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: empty tensor, module: error checking, module: norms and normalization",review linked PR state before touching issue,2025-12-02T14:42:24Z +issue,169150,[debug mode] do not hash collectives,https://github.com/pytorch/pytorch/issues/169150,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-01T22:47:49Z +issue,135061,[export] non-strict export error on torch.distributions.Normal,https://github.com/pytorch/pytorch/issues/135061,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-12-01T19:36:49Z +issue,168975,No way to dump DebugMode hashes to file for further processing / diffing,https://github.com/pytorch/pytorch/issues/168975,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-12-01T18:22:16Z +issue,169236,"[MPS] nn.ConvTranspose{1,2,3}d ignore validation check for output_padding < stride (Silent Failure)",https://github.com/pytorch/pytorch/issues/169236,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",review linked PR state before touching issue,2025-11-30T03:51:39Z +issue,136642,Linespace does not support SymInt step. (but int),https://github.com/pytorch/pytorch/issues/136642,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: faketensor, module: pt2-dispatcher, oncall: export",review linked PR state before touching issue,2025-11-26T16:12:21Z +issue,168566,Test: AOTInductorTestsTemplate.test_autotune_int64_user_defined_triton_kernel,https://github.com/pytorch/pytorch/issues/168566,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-11-25T19:46:07Z +issue,168872,Test: TestSDPACudaOnly.test_cudnn_attention_broken_166211,https://github.com/pytorch/pytorch/issues/168872,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sdpa",review linked PR state before touching issue,2025-11-25T19:13:07Z +issue,168868,Test: TestTorchDeviceType.test_pdist_norm_large,https://github.com/pytorch/pytorch/issues/168868,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: rocm,review linked PR state before touching issue,2025-11-25T00:51:52Z +issue,168136,tracing torch _ops.py like viztracer and cpython settrace will case stuck in an infinite loop.,https://github.com/pytorch/pytorch/issues/168136,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: library, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-24T14:17:52Z +issue,150319,[dynamo] Bad error message when trying to compile async functions due to bad bytecode jump targets,https://github.com/pytorch/pytorch/issues/150319,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-11-21T20:58:40Z +issue,167596,[dynamo][feature] Guard on constants only if graph is specialized and not bytecode,https://github.com/pytorch/pytorch/issues/167596,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-11-20T17:44:45Z +issue,167906,Avoid Exception Refcycle Problems,https://github.com/pytorch/pytorch/issues/167906,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: python frontend",review linked PR state before touching issue,2025-11-18T22:15:10Z +issue,139092,`tensor` not a `FakeTensor` under `FakeTensorMode` and `device('meta')`,https://github.com/pytorch/pytorch/issues/139092,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-18T20:11:28Z +issue,167719,KeyError when exporting PyTorch model with dynamic string keys in ModuleDict using torch.export,https://github.com/pytorch/pytorch/issues/167719,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-11-18T19:13:04Z +issue,129327,[export/dynamo] torch._check fails at compile time when the condition evaluates to False,https://github.com/pytorch/pytorch/issues/129327,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-11-17T21:14:35Z +issue,167732,Dynamo should support backward() capture,https://github.com/pytorch/pytorch/issues/167732,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-11-17T21:10:06Z +issue,167191,[dynamo] [dynamic] `nn.LazyLinear` reject SymInt when direct using `torch.compile()`,https://github.com/pytorch/pytorch/issues/167191,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-11-17T07:16:46Z +issue,158975,Handle matmuls gracefully in foreach_map,https://github.com/pytorch/pytorch/issues/158975,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-15T04:47:06Z +issue,126024,Can not use torch.compile with dynamic=Ture when using multi-threads,https://github.com/pytorch/pytorch/issues/126024,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: multithreading, oncall: pt2",review linked PR state before touching issue,2025-11-13T06:41:30Z +issue,167026,Support torch.cuda.use_mem_pool in torch.compile,https://github.com/pytorch/pytorch/issues/167026,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-11-11T18:40:53Z +issue,167117,`fx.GraphModule` serialization fails when type annotations are `str`s,https://github.com/pytorch/pytorch/issues/167117,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-11-11T18:19:48Z +issue,131794,[torch.compile] Unused differentiable outputs lead to unnecessary gradient computation when compared to eager,https://github.com/pytorch/pytorch/issues/131794,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-10T19:31:21Z +issue,123663,Improve Foreach OpInfo test infra to enable meta coverage,https://github.com/pytorch/pytorch/issues/123663,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: mta, module: testing, oncall: pt2",review linked PR state before touching issue,2025-11-07T18:00:34Z +issue,122200,"Dynamo compilation should always run with empty functorch interpreter stack, even if ambient context has functorch",https://github.com/pytorch/pytorch/issues/122200,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: functorch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-07T15:53:41Z +issue,128554,aot_export_module produces unserializable GraphModules,https://github.com/pytorch/pytorch/issues/128554,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-11-06T22:18:19Z +issue,139077,Add support for binding tensor kwargs in max-autotune templates,https://github.com/pytorch/pytorch/issues/139077,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-11-06T17:59:03Z +issue,137520,Support torch.cond with differing (but same dimensionality) sizes on true/false branches,https://github.com/pytorch/pytorch/issues/137520,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-11-06T02:45:10Z +issue,157926,torch.compile() disables torch.distributions parameter validation globally,https://github.com/pytorch/pytorch/issues/157926,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-11-05T22:09:23Z +issue,129845,vmap + autograd.Function [generate_vmap_rule=True] + torch.compile don't work together,https://github.com/pytorch/pytorch/issues/129845,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-05T21:06:24Z +issue,128830,TORCHDYNAMO_REPRO_AFTER=aot produces invalid repro code,https://github.com/pytorch/pytorch/issues/128830,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: minifier, oncall: pt2",review linked PR state before touching issue,2025-11-05T18:58:55Z +issue,137875,TORCH_LOGS=+inductor spams too much information about cache,https://github.com/pytorch/pytorch/issues/137875,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-11-05T18:53:19Z +issue,166516,Performance issue of torch._higher_order_ops.scan,https://github.com/pytorch/pytorch/issues/166516,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-04T21:32:28Z +issue,166905,Type Annotation for `named_modules()`,https://github.com/pytorch/pytorch/issues/166905,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: typing,review linked PR state before touching issue,2025-11-04T19:25:57Z +issue,166525,Inductor Fails with `AttributeError: 'int' object has no attribute 'meta'` for `torch.while_loop` Containing Scalar Values in Loop State,https://github.com/pytorch/pytorch/issues/166525,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-11-04T10:06:01Z +issue,161650,config.activation_memory_budget,https://github.com/pytorch/pytorch/issues/161650,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: functorch, oncall: pt2",review linked PR state before touching issue,2025-10-26T15:11:45Z +issue,133592,[torchbind x compile] Can't register a torchbind operator that mutates a tensor,https://github.com/pytorch/pytorch/issues/133592,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, module: torchbind, module: vllm, oncall: pt2",review linked PR state before touching issue,2025-10-24T16:39:56Z +issue,136404,Profiling Torch.compile (with_stack=True) leads to frozen program with massive memory.,https://github.com/pytorch/pytorch/issues/136404,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: vllm, oncall: profiler, oncall: pt2",review linked PR state before touching issue,2025-10-24T16:39:00Z +issue,152548,FakeTensorUpdater does not trace nodes correctly,https://github.com/pytorch/pytorch/issues/152548,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: vllm, oncall: pt2",review linked PR state before touching issue,2025-10-24T16:38:43Z +issue,137869,TORCH_COMPILE_CPROFILE does not work for python 3.12,https://github.com/pytorch/pytorch/issues/137869,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: vllm, oncall: pt2",review linked PR state before touching issue,2025-10-24T16:38:29Z +issue,165547,error message for deepcopy a cuda tensor under FakeTensorMode is pretty cryptic,https://github.com/pytorch/pytorch/issues/165547,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: pt2-dispatcher, module: vllm, oncall: pt2",review linked PR state before touching issue,2025-10-24T16:38:12Z +issue,161119,[UED][Falcon-7B x vLLM] excessive shape recompilations (static specialization),https://github.com/pytorch/pytorch/issues/161119,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: dynamo, module: vllm, oncall: pt2",review linked PR state before touching issue,2025-10-24T16:38:04Z +issue,161111,[UED][Falcon-7B x vLLM] errors tracing causal_conv_1d triton kernel grid fn,https://github.com/pytorch/pytorch/issues/161111,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: vllm, oncall: pt2",review linked PR state before touching issue,2025-10-24T16:38:02Z +issue,157776,Support for _Float16/C++23 std::float16_t,https://github.com/pytorch/pytorch/issues/157776,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cpp,review linked PR state before touching issue,2025-10-23T18:34:00Z +issue,165981,Fails to use side_effect op inside torch.cond branch,https://github.com/pytorch/pytorch/issues/165981,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-10-22T23:47:49Z +issue,166009,[Inductor] DCE breaks with fallback operators that produce aliases + mutation,https://github.com/pytorch/pytorch/issues/166009,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-10-21T18:45:41Z +issue,127153,"[ONNX] pad_sequence() is not exportable, with neither legacy onnx.export nor with dynamo_export",https://github.com/pytorch/pytorch/issues/127153,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-18T14:10:49Z +issue,129418,functionalization should run underneath python decomps,https://github.com/pytorch/pytorch/issues/129418,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-16T22:17:40Z +issue,165097,Distributed checkpoint should not expect private use 1 to be cuda compatible,https://github.com/pytorch/pytorch/issues/165097,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: privateuse1, oncall: distributed, oncall: distributed checkpointing",review linked PR state before touching issue,2025-10-16T11:57:37Z +issue,137388,some guards are not shown in the code_parts,https://github.com/pytorch/pytorch/issues/137388,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: guards, oncall: pt2",review linked PR state before touching issue,2025-10-15T18:55:55Z +issue,93367,Aot accuracy minifier with dynamic shapes doesn't work,https://github.com/pytorch/pytorch/issues/93367,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: minifier, oncall: pt2",review linked PR state before touching issue,2025-10-15T18:53:02Z +issue,106135,Registering function that takes const std::vector& to SymInt[] schema gives confusing error message,https://github.com/pytorch/pytorch/issues/106135,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-10-15T18:37:18Z +issue,164922,`torch.compile` fails to trace `datetime.now()` with Dynamo guard check failure,https://github.com/pytorch/pytorch/issues/164922,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-10-14T20:25:33Z +issue,150613,"Torch compile for `torch.searchsorted` failed when capturing scalar outputs, with scalar `values` taken from `sorted_sequence`",https://github.com/pytorch/pytorch/issues/150613,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: decompositions, module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-10-14T03:09:18Z +issue,164883,Wheels include protobuf headers (and headers for other projects),https://github.com/pytorch/pytorch/issues/164883,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",review linked PR state before touching issue,2025-10-13T17:19:00Z +issue,122773,torch.gather - Cannot call numel() on tensor with symbolic sizes/strides,https://github.com/pytorch/pytorch/issues/122773,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-10-10T16:02:55Z +issue,111385,Torch Compile Dynamic fails on sample on diffusers VAE,https://github.com/pytorch/pytorch/issues/111385,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-10-10T05:45:19Z +issue,164559,fwd_rng_state show up in the aot_export_joint grpah input,https://github.com/pytorch/pytorch/issues/164559,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-10-06T19:10:58Z +issue,136628,[torch.export.load] failed while executing `pow_by_natural`,https://github.com/pytorch/pytorch/issues/136628,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-10-05T22:48:02Z +issue,127163,BUG: `torch.pow(-0.0)` wrongly returns `-0.0` (should be +0.0),https://github.com/pytorch/pytorch/issues/127163,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: python frontend",review linked PR state before touching issue,2025-10-03T09:02:04Z +issue,164478,Out of date references in CONTRIBUTING.md and Wiki,https://github.com/pytorch/pytorch/issues/164478,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label,review linked PR state before touching issue,2025-10-02T18:22:17Z +issue,133655,PythonMod vs Mod spookiness,https://github.com/pytorch/pytorch/issues/133655,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-10-01T03:25:38Z +issue,164124,"torch.compile compiles multiple Triton autotune kernels, but uses the wrong ones",https://github.com/pytorch/pytorch/issues/164124,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: user triton, oncall: pt2",review linked PR state before touching issue,2025-09-29T16:50:17Z +issue,160388,Perf differences observed between AOT and JIT on RAFT model,https://github.com/pytorch/pytorch/issues/160388,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-09-24T22:20:39Z +issue,163143,[torch.compile] Side-Effectful Tokens doesn't work well with torch.cond,https://github.com/pytorch/pytorch/issues/163143,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-09-24T17:41:47Z +issue,163641,u0==u1 but u0//u1==1 throw DDE,https://github.com/pytorch/pytorch/issues/163641,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-09-23T19:14:12Z +issue,163037,[compile/inductor][functionalization] `prims.broadcast_in_dim` triggers alias-annotation error under `torch.compile` (eager OK),https://github.com/pytorch/pytorch/issues/163037,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-09-20T16:27:34Z +issue,162980,Adding custom function to _allowed_determinism_checks_to_fns from model code,https://github.com/pytorch/pytorch/issues/162980,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,review linked PR state before touching issue,2025-09-19T21:10:55Z +issue,163040,[torch.export.load] Passing an `ExportedProgram` raises `AttributeError: 'ExportedProgram' object has no attribute 'tell'` instead of a clear `TypeError`,https://github.com/pytorch/pytorch/issues/163040,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-18T16:29:49Z +issue,160016,Requesting improved torch.auto_grad.detect_anomaly() for NaN detection,https://github.com/pytorch/pytorch/issues/160016,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: autograd,review linked PR state before touching issue,2025-09-18T15:02:47Z +issue,152178,Raise an Error when File Not Found in `torch.jit.load()`,https://github.com/pytorch/pytorch/issues/152178,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, oncall: jit",review linked PR state before touching issue,2025-09-18T14:26:46Z +issue,162973,[dynamo] GB_REGISTRY linter taking a long time,https://github.com/pytorch/pytorch/issues/162973,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-09-15T20:08:39Z +issue,162687,custom op with torch.compile backward error when op has default arguments,https://github.com/pytorch/pytorch/issues/162687,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: custom-operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-09-15T18:48:09Z +issue,162151,`permute` + `add` + `transpose` plays wrong on tensor slice,https://github.com/pytorch/pytorch/issues/162151,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-09-15T17:43:18Z +issue,132923,"If you compile a function/module, we should definitely attempt to compile it (don't try to skip it)",https://github.com/pytorch/pytorch/issues/132923,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:20:36Z +issue,142358,`torch.compiler.disable()` on module hooks will disable `module.compile()`,https://github.com/pytorch/pytorch/issues/142358,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, module: nn, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:15:15Z +issue,143308,"[dynamo, guards] Implement FrameLocalsMapping version of check_verbose_nopybind",https://github.com/pytorch/pytorch/issues/143308,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: guards, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:14:27Z +issue,144360,Skip empty frames recursively when top-level is empty,https://github.com/pytorch/pytorch/issues/144360,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:13:43Z +issue,149565,Should (eventually) be recommending nonstrict_trace instead of allow_in_graph in error messages:,https://github.com/pytorch/pytorch/issues/149565,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:07:57Z +issue,144820,[dynamo] Do not always skip code objects unconditionally,https://github.com/pytorch/pytorch/issues/144820,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:07:06Z +issue,150915,`module.compile()` behaves differently from `torch.compile(module)`,https://github.com/pytorch/pytorch/issues/150915,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-09-12T21:06:40Z +issue,161796,[dynamo] Graph break logspam,https://github.com/pytorch/pytorch/issues/161796,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, module: logging, oncall: pt2",review linked PR state before touching issue,2025-09-12T19:57:00Z +issue,161705,`aten.slice` adds guard on backed symint which causes inconsistent behavior between eager and trace modes,https://github.com/pytorch/pytorch/issues/161705,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-09-11T03:42:39Z +issue,162287,using torch.sym_not cause Z3Exception: Symbolic expressions cannot be cast to concrete Boolean values.,https://github.com/pytorch/pytorch/issues/162287,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-09-10T06:56:17Z +issue,162350,Inplace modification error with softmax when combining forward and backward AD,https://github.com/pytorch/pytorch/issues/162350,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad",review linked PR state before touching issue,2025-09-08T23:29:26Z +issue,161764,"The ""max-autotune"" mode in torch.compile can lead to longer execution times relative to eager execution.",https://github.com/pytorch/pytorch/issues/161764,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: pt2",review linked PR state before touching issue,2025-09-08T17:49:39Z +issue,162373,Pylint E1102: torch.fft.rfft/istft/irfft and more is not callable,https://github.com/pytorch/pytorch/issues/162373,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: typing,review linked PR state before touching issue,2025-09-08T15:50:36Z +issue,161456,[dynamo][hop] speculate_subgraph doesn't work with subclass inputs,https://github.com/pytorch/pytorch/issues/161456,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-09-05T19:26:36Z +issue,155238,Dynamo Fails on pad_packed_sequence Due to Fake Tensor Allocation Issue During FX Graph Tracing,https://github.com/pytorch/pytorch/issues/155238,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: pt2-dispatcher, module: rnn, oncall: pt2",review linked PR state before touching issue,2025-08-30T18:12:42Z +issue,161088,torch.where should throw an exception when invoked with non-bool tensor,https://github.com/pytorch/pytorch/issues/161088,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-08-26T18:12:22Z +issue,132635,[Dynamo] support logging via reorderable_logging_functions,https://github.com/pytorch/pytorch/issues/132635,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-25T23:51:15Z +issue,161067,torch.compile not eliminating common quantization nodes,https://github.com/pytorch/pytorch/issues/161067,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-08-25T17:59:40Z +issue,160757,Graph break when torch.compiling an autograd.Function,https://github.com/pytorch/pytorch/issues/160757,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-23T16:47:47Z +issue,161138,Fuse 3 elementwise ops and reduction into 1 hbm reads,https://github.com/pytorch/pytorch/issues/161138,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-08-23T13:17:21Z +issue,161289,automatically updating ci benchmark expected csv fails,https://github.com/pytorch/pytorch/issues/161289,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-08-23T13:10:57Z +issue,129319,`int` type for `dims` of `tile()` without `dims=` works with a tensor against the doc,https://github.com/pytorch/pytorch/issues/129319,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs,review linked PR state before touching issue,2025-08-21T12:50:54Z +issue,117016,There is a performance drop because we have not yet implemented the batching rule for aten::_scaled_dot_product_efficient_attention_backward,https://github.com/pytorch/pytorch/issues/117016,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: sdpa, module: vmap",review linked PR state before touching issue,2025-08-21T05:42:19Z +issue,159843,[DTensor] nn.Embedding Compile Failure when Creating FX Graph,https://github.com/pytorch/pytorch/issues/159843,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-18T22:22:46Z +issue,155421,Backward fails with compiled attention on nested tensors on RTX6000,https://github.com/pytorch/pytorch/issues/155421,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, oncall: pt2",review linked PR state before touching issue,2025-08-18T17:16:29Z +issue,160544,Error with torch.ops.higher_order.scan + torch.compile [KeyError: combine_fn],https://github.com/pytorch/pytorch/issues/160544,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-08-15T22:48:59Z +issue,156075,`torch.logsumexp`: support `dim=None`,https://github.com/pytorch/pytorch/issues/156075,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,review linked PR state before touching issue,2025-08-13T22:50:29Z +issue,160247,wan2.1 vae take more gpu memory after compile,https://github.com/pytorch/pytorch/issues/160247,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: memory usage, oncall: pt2",review linked PR state before touching issue,2025-08-11T20:06:56Z +issue,160083,compile regression on forward pre hook with prepend=True,https://github.com/pytorch/pytorch/issues/160083,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-11T18:25:29Z +issue,160271,Error when using torch.compile with dynamic=True: tensor 'guidance' stride mismatch at index 0,https://github.com/pytorch/pytorch/issues/160271,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-08-11T17:20:32Z +issue,159831,ModuleDict subscription no longer works after compile().,https://github.com/pytorch/pytorch/issues/159831,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-05T21:24:40Z +issue,159550,_adaptive_avg_pool2d does not support dynamic shapes？,https://github.com/pytorch/pytorch/issues/159550,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-08-04T20:14:01Z +issue,129648,`torch.compile` mode and eager mode are different in generating tensor API behavior,https://github.com/pytorch/pytorch/issues/129648,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-04T18:58:24Z +issue,128531,test/dynamo/test_torchrec.py doesn't run in OSS CI (and won't run locally),https://github.com/pytorch/pytorch/issues/128531,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-04T18:35:30Z +issue,159594,"During CUDA graph capture, CUDACachingAllocator could potentially trigger cudaFree()",https://github.com/pytorch/pytorch/issues/159594,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs, module: cudacachingallocator",review linked PR state before touching issue,2025-08-04T17:37:41Z +issue,158970,Add option to assert if kernel is not fully fused in foreach_map,https://github.com/pytorch/pytorch/issues/158970,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-08-04T17:06:47Z +issue,159613,[Torch FX] subgraph_rewriter.replace_pattern does not handle back-to-back matches correctly.,https://github.com/pytorch/pytorch/issues/159613,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-08-04T16:19:01Z +issue,75862,LayerNorm and GroupNorm with num_groups=1 not equivalent,https://github.com/pytorch/pytorch/issues/75862,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",review linked PR state before touching issue,2025-08-03T04:11:10Z +issue,124747,TestAOTAutograd.test_mem_leak_from_save_for_bw fails locally but not on CI when run with dynamo,https://github.com/pytorch/pytorch/issues/124747,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-01T23:33:40Z +issue,126834,[dynamo] Check for invalid random_ arguments while tracing,https://github.com/pytorch/pytorch/issues/126834,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-08-01T23:14:07Z +issue,159457,`@assume_constant_result` Causes `NameError` When Method Contains Tensor Operations and Dynamic Attribute Assignment During Compilation,https://github.com/pytorch/pytorch/issues/159457,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-07-30T17:27:45Z +issue,93437,Accuracy minifier can find spurious accuracy failures involving uninitialized memory,https://github.com/pytorch/pytorch/issues/93437,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: minifier, oncall: pt2",review linked PR state before touching issue,2025-07-29T14:19:18Z +issue,118739,"Getting ""RuntimeError: Trying to backward through the graph a second time"" when calling backward on compiled unbind / split ops",https://github.com/pytorch/pytorch/issues/118739,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-07-29T14:18:43Z +issue,106704,Dataloader extremely slow on in-memory datasets,https://github.com/pytorch/pytorch/issues/106704,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",review linked PR state before touching issue,2025-07-29T10:57:00Z +issue,159259,Use aten/src/ATen/native/native_functions.yaml to automatically generate tests for type hints,https://github.com/pytorch/pytorch/issues/159259,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: typing,review linked PR state before touching issue,2025-07-28T22:02:33Z +issue,159192,Invalid meta kernel for _fused_moving_avg_obs_fq_helper_functional,https://github.com/pytorch/pytorch/issues/159192,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-07-28T14:31:39Z +issue,159160,"[dynamo, test] Logging error spam at the end of pytest",https://github.com/pytorch/pytorch/issues/159160,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-07-25T15:20:36Z +issue,158561,`torch.compile` In-place operation raises Autograd error even when gradients are not required,https://github.com/pytorch/pytorch/issues/158561,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-07-22T14:44:13Z +issue,158521,[static cuda launcher] scalar float support,https://github.com/pytorch/pytorch/issues/158521,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-07-22T14:36:30Z +issue,154282,LazyGraphModule causes graph breaks,https://github.com/pytorch/pytorch/issues/154282,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-07-21T19:05:28Z +issue,134784,Dynamo x autograd.Function x funtorch transform: intermediate sliced tensor doesn't require grad caused incorrect result,https://github.com/pytorch/pytorch/issues/134784,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: functorch, oncall: pt2",review linked PR state before touching issue,2025-07-17T23:36:14Z +issue,117584,[pt2d] module register_pre_forward_hook and register_forward_hook triggered graph break when it's root module,https://github.com/pytorch/pytorch/issues/117584,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2",review linked PR state before touching issue,2025-07-17T22:34:37Z +issue,105768,torch.compile uses more memory when using less parameters,https://github.com/pytorch/pytorch/issues/105768,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: memory usage, oncall: pt2",review linked PR state before touching issue,2025-07-17T15:57:13Z +issue,141005,OptimizedModule _parameters attr getting and setting are not working as expected,https://github.com/pytorch/pytorch/issues/141005,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-07-16T23:19:26Z +issue,138961,CompiledFxGraph.current_callable is not thread-safe,https://github.com/pytorch/pytorch/issues/138961,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: multithreading, oncall: pt2",review linked PR state before touching issue,2025-07-15T18:32:51Z +issue,158088,`torch.compile` doesn't properly raise eager fake tensor exception,https://github.com/pytorch/pytorch/issues/158088,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-07-15T17:51:31Z +issue,158087,`torch.compile` ignores nan check in `int(torch.tensor(torch.nan))`,https://github.com/pytorch/pytorch/issues/158087,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: nans and infs, oncall: pt2",review linked PR state before touching issue,2025-07-14T20:19:34Z +issue,39383,Pickling a Tensor or a Storage is not deterministic,https://github.com/pytorch/pytorch/issues/39383,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pickle,review linked PR state before touching issue,2025-07-13T23:37:04Z +issue,122512,Higher peak memory with torch.compile,https://github.com/pytorch/pytorch/issues/122512,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-07-11T18:31:58Z +issue,114296,Minifier doesn't work with dynamic shapes,https://github.com/pytorch/pytorch/issues/114296,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: minifier, oncall: pt2",review linked PR state before touching issue,2025-07-11T18:17:52Z +issue,125077,[Inductor] Generate triton block pointers for discontiguous strided tensors,https://github.com/pytorch/pytorch/issues/125077,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-07-11T18:04:25Z +issue,122840,a log_softmax kernel get much worse perf with padding,https://github.com/pytorch/pytorch/issues/122840,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-07-11T17:35:22Z +issue,123470,torch.compile x custom ops: op that accepts float also accepts Tensor in eager-mode,https://github.com/pytorch/pytorch/issues/123470,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-07-10T23:58:22Z +issue,125438,Fix broken linalg unittests on ARM platform,https://github.com/pytorch/pytorch/issues/125438,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: linear algebra, module: tests",review linked PR state before touching issue,2025-07-10T16:37:06Z +issue,157832,[FSDP2] should pass args as is instead of creating new ones,https://github.com/pytorch/pytorch/issues/157832,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",review linked PR state before touching issue,2025-07-08T19:31:15Z +issue,157610,`torch.compile` fails on `prims.broadcast_in_dim` with alias annotation error,https://github.com/pytorch/pytorch/issues/157610,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-07-07T22:51:34Z +issue,155584,[Upstream Triton] Handle user-specified triton.set_allocator function,https://github.com/pytorch/pytorch/issues/155584,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-07-07T15:12:29Z +issue,157217,[dynamo] dynamo is unable to enter `except RuntimeError` while eager can,https://github.com/pytorch/pytorch/issues/157217,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-07-04T14:18:30Z +issue,149982,request for faster inductor kernels for blockwise reduction across dim1 -> write,https://github.com/pytorch/pytorch/issues/149982,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-06-26T12:49:49Z +issue,156614,[dynamo] `torch.Tensor.random_` causes graph break,https://github.com/pytorch/pytorch/issues/156614,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-06-24T01:48:16Z +issue,156640,[dynamo] Wrap asserts (and other Dynamo errors) more clearly,https://github.com/pytorch/pytorch/issues/156640,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-06-23T21:34:56Z +issue,156162,torch2.7.1 issue for torch.compile numpy,https://github.com/pytorch/pytorch/issues/156162,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: numpy, oncall: pt2",review linked PR state before touching issue,2025-06-20T15:57:49Z +issue,156191,Dynamo does not know how to trace method `__len__` of class `` with torch.logging calls,https://github.com/pytorch/pytorch/issues/156191,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-06-17T19:27:01Z +issue,156127,[dynamo] Show carets in graph break stack traces,https://github.com/pytorch/pytorch/issues/156127,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-06-16T22:26:40Z +issue,155800,torch.distributions.kl_divergence Fails with MultivariateNormal in Dynamo Due to _infer_size Type Error,https://github.com/pytorch/pytorch/issues/155800,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-06-16T19:03:07Z +issue,156059,Dynamo trace an incorrect result on torch._C._storage_Use_Count,https://github.com/pytorch/pytorch/issues/156059,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-06-16T18:25:16Z +issue,153227,eval should handle (unhinted: (s77 > 3) | (u0 > 200)) when s77 has hint =5,https://github.com/pytorch/pytorch/issues/153227,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-06-14T02:56:33Z +issue,155856,[user triton][dynamo Fix TMA Descriptor reconstruction,https://github.com/pytorch/pytorch/issues/155856,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-06-13T21:14:48Z +issue,153204,mark_unbacked for strides.,https://github.com/pytorch/pytorch/issues/153204,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-06-13T03:55:43Z +issue,146274,torch.nn.functional.one_hot has inconsistent behavior between eager and torch.compile when num_classes=0,https://github.com/pytorch/pytorch/issues/146274,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-06-12T17:48:58Z +issue,154807,FP8 scaled mm lowering ignores scale_result argument,https://github.com/pytorch/pytorch/issues/154807,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-06-06T19:07:31Z +issue,154161,PGO errors out in dynamo main path for sparse tensors,https://github.com/pytorch/pytorch/issues/154161,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-06-06T01:40:07Z +issue,132929,Compile + torch.autograd.grad returns no gradients,https://github.com/pytorch/pytorch/issues/132929,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-06-06T01:38:51Z +issue,154851,Improve warning when specializations happen due to operator,https://github.com/pytorch/pytorch/issues/154851,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-06-04T23:04:04Z +issue,154284,Using `compile` on `hessian(hessian)`,https://github.com/pytorch/pytorch/issues/154284,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: functorch, oncall: pt2",review linked PR state before touching issue,2025-06-04T22:44:23Z +issue,154592,inductor benchmark_compiled_module's random initialization will cause crash for data dependent indexing,https://github.com/pytorch/pytorch/issues/154592,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-06-02T01:41:22Z +issue,101192,AOTAutograd export path does not support training graphs with parameters that do not receive gradients.,https://github.com/pytorch/pytorch/issues/101192,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",review linked PR state before touching issue,2025-05-30T21:45:44Z +issue,154454,[Dynamo] Confusing re-raise exception handling graph break message,https://github.com/pytorch/pytorch/issues/154454,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-05-27T21:58:10Z +issue,154262,max_pool2d padding assert incorrect with dilation,https://github.com/pytorch/pytorch/issues/154262,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: nn, module: padding",review linked PR state before touching issue,2025-05-27T16:34:34Z +issue,152750,Error on padding 0-sized tensors,https://github.com/pytorch/pytorch/issues/152750,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,review linked PR state before touching issue,2025-05-23T22:12:17Z +issue,154259,torch.compile regression blocks torchvision/torchbench pin upgrade,https://github.com/pytorch/pytorch/issues/154259,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-05-23T22:07:28Z +issue,112586,"Fix docstring errors in _guards.py, _ops.py, _jit_internal.py, functional.py, _tensor_str.py, library.py",https://github.com/pytorch/pytorch/issues/112586,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs,review linked PR state before touching issue,2025-05-22T17:10:22Z +issue,109652,InstanceNorm does not catch dim mismatch,https://github.com/pytorch/pytorch/issues/109652,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: nn,review linked PR state before touching issue,2025-05-21T00:38:35Z +issue,153149,Behavior of kernel_size parameter of torch.nn.functional.avg_pool2d does not match with documentation,https://github.com/pytorch/pytorch/issues/153149,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: pooling, oncall: export",review linked PR state before touching issue,2025-05-14T18:20:04Z +issue,128649,[Nested Tensor / subclasses] view(-1) and splitting dimensions with view() failures,https://github.com/pytorch/pytorch/issues/128649,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: nestedtensor, oncall: pt2",review linked PR state before touching issue,2025-05-14T17:37:07Z +issue,153175,prefer_deferred_runtime_asserts_over_guards does not work correctly e2e with compile. insert_deferred_runtime_asserts and inductor codegen skips them,https://github.com/pytorch/pytorch/issues/153175,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-05-08T20:31:08Z +issue,152346,compile generates inefficient code for mutations on small slices of inputs,https://github.com/pytorch/pytorch/issues/152346,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-05-06T15:16:04Z +issue,92600,Update quantization to make source files complient with /Zc:lambda,https://github.com/pytorch/pytorch/issues/92600,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: quantization",review linked PR state before touching issue,2025-05-05T19:54:21Z +issue,101249,`einsum` is about 40x slower on CUDA than manually multiplying and summing,https://github.com/pytorch/pytorch/issues/101249,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",review linked PR state before touching issue,2025-05-02T22:26:18Z +issue,152343,cudagraphs: `static_input_indices` incorrectly including SymInt graph args when using tensor subclasses + dynamic shapes,https://github.com/pytorch/pytorch/issues/152343,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: cuda graphs, module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-04-28T18:29:03Z +issue,152183,write a custom ViewAndMutationmeta.__repr__,https://github.com/pytorch/pytorch/issues/152183,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-04-28T13:11:29Z +issue,152095,To fix inconsistency between signature and doc on `torch.bernoulli()`,https://github.com/pytorch/pytorch/issues/152095,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs",review linked PR state before touching issue,2025-04-24T15:08:45Z +issue,149989,[dynamo][higher order ops] Make support_aliasing and support_input_mutation default to False,https://github.com/pytorch/pytorch/issues/149989,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-04-24T05:48:24Z +issue,134964,Add the pages of `torch.e` and `torch.pi` to PyTorch doc,https://github.com/pytorch/pytorch/issues/134964,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: python frontend",review linked PR state before touching issue,2025-04-23T02:51:49Z +issue,151558,2.6: SDPA disallows masked attention under vmap,https://github.com/pytorch/pytorch/issues/151558,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: sdpa, module: vmap",review linked PR state before touching issue,2025-04-17T23:53:36Z +issue,139756,Bug with torch.compile + fwAD,https://github.com/pytorch/pytorch/issues/139756,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: forward ad, oncall: pt2",review linked PR state before touching issue,2025-04-15T21:32:31Z +issue,151196,compile of vmap of jacfwd fails,https://github.com/pytorch/pytorch/issues/151196,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: functorch, oncall: pt2",review linked PR state before touching issue,2025-04-14T22:29:13Z +issue,149097,Aten arange behavior when dtype is int64 and step size is greater than range,https://github.com/pytorch/pytorch/issues/149097,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,review linked PR state before touching issue,2025-04-14T04:14:02Z +issue,150857,No way to save/load model weights between PyTorch and LibTorch (C++) in a compatible way,https://github.com/pytorch/pytorch/issues/150857,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: serialization",review linked PR state before touching issue,2025-04-09T09:11:52Z +issue,150621,"2.8.0 Nightly - ""Feature 'cvt with .bf16.f16' requires .target sm_90 or higher""",https://github.com/pytorch/pytorch/issues/150621,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2025-04-08T20:40:14Z +issue,149894,[ued][f5-tts][dynamo] `torch.compile` changes state dict,https://github.com/pytorch/pytorch/issues/149894,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-04-07T17:49:02Z +issue,142321,torch.compile 路径处理错误导致 C++ 编译失败,https://github.com/pytorch/pytorch/issues/142321,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: pt2",review linked PR state before touching issue,2025-04-07T09:34:26Z +issue,150465,`torch._dynamo.nonstrict_trace` has confusing user code stacktrace,https://github.com/pytorch/pytorch/issues/150465,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-04-01T23:30:09Z +issue,150321,Masking out Loads from mask_modded out regions,https://github.com/pytorch/pytorch/issues/150321,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: flex attention,review linked PR state before touching issue,2025-03-31T19:24:02Z +issue,149857,SDPA (`EFFICIENT_ATTENTION`) slower than torch.compile decomposition on `tf32`,https://github.com/pytorch/pytorch/issues/149857,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: sdpa, oncall: pt2",review linked PR state before touching issue,2025-03-31T16:19:09Z +issue,149921,[ued][whisper][dynamo] Graph break - setattr on class object,https://github.com/pytorch/pytorch/issues/149921,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-29T15:38:35Z +issue,149911,[dynamo] `torch.compile` doesn't respect `GradTrackingTensor`'s data attribute mutation check,https://github.com/pytorch/pytorch/issues/149911,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-29T15:38:34Z +issue,149909,[ued][kokoro] RNN/LSTMS do not work with torch.compile,https://github.com/pytorch/pytorch/issues/149909,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-29T15:38:33Z +issue,149968,[ued][qwen][dynamo] inspect.signature with non-constant function,https://github.com/pytorch/pytorch/issues/149968,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-29T15:35:23Z +issue,149895,[Dynamo] Cannot instantiate class if `__getattribute__` is defined,https://github.com/pytorch/pytorch/issues/149895,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-29T15:00:26Z +issue,137275,Device transfer for NJT within torch.compile allocates a new nested int,https://github.com/pytorch/pytorch/issues/137275,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, oncall: pt2",review linked PR state before touching issue,2025-03-28T18:54:25Z +issue,128063,[Inductor] Fusion of Tiled Point-Wise and Reduction Operators,https://github.com/pytorch/pytorch/issues/128063,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8), module: inductor, oncall: pt2",review linked PR state before touching issue,2025-03-27T15:20:10Z +issue,133734,PyTorch `torch.compile` inductor backend shows enormous compile-time regressions.,https://github.com/pytorch/pytorch/issues/133734,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-03-27T02:11:58Z +issue,133924,AOTAutograd not tracing AMP correctly when using PyTorch/XLA.,https://github.com/pytorch/pytorch/issues/133924,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: aotdispatch, module: pt2-dispatcher, module: xla",review linked PR state before touching issue,2025-03-26T12:16:01Z +issue,134533,torch._assert does not work with torch.export,https://github.com/pytorch/pytorch/issues/134533,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-03-26T00:13:49Z +issue,149556,[ued][f5-tts][dynamo] dont graph break on `torch.jit.isinstance`,https://github.com/pytorch/pytorch/issues/149556,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-25T00:29:11Z +issue,149566,User-torch._dynamo.disable annotations may be misleading,https://github.com/pytorch/pytorch/issues/149566,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-24T10:35:36Z +issue,149583,Easydict support,https://github.com/pytorch/pytorch/issues/149583,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-20T07:16:21Z +issue,149047,Unsupported: call_method NNModuleVariable() register_forward_hook [NestedUserFunctionVariable()] {},https://github.com/pytorch/pytorch/issues/149047,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-19T01:18:07Z +issue,148842,We should use max size instead of hint size when autotuning,https://github.com/pytorch/pytorch/issues/148842,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-03-18T13:18:55Z +issue,143495,"infer_size(a, b) fails when it could return a value",https://github.com/pytorch/pytorch/issues/143495,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: faketensor, oncall: pt2",review linked PR state before touching issue,2025-03-14T14:48:03Z +issue,149037,(Will PR if ok) Support generator returning values,https://github.com/pytorch/pytorch/issues/149037,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-03-13T06:30:28Z +issue,149010,[XFORMERS] torch._dynamo.exc.Unsupported,https://github.com/pytorch/pytorch/issues/149010,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-13T05:38:42Z +issue,148859,[Inductor] Output mismatch shape after compilation,https://github.com/pytorch/pytorch/issues/148859,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2 accuracy, oncall: pt2",review linked PR state before touching issue,2025-03-10T22:25:53Z +issue,148779,[Export Benchmark] non-strict export doesn't work with .numpy on Kokoro model,https://github.com/pytorch/pytorch/issues/148779,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-03-10T20:59:21Z +issue,147991,Verifier (in torch.export.export) does not make use of if-condition inside branches,https://github.com/pytorch/pytorch/issues/147991,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2025-03-10T17:27:31Z +issue,140960,Better Error Message for SymIntArrayRef expected to contain only concrete integers,https://github.com/pytorch/pytorch/issues/140960,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-03-08T08:25:28Z +issue,148475,Dynamo replaces exception by hard error in `run_node`,https://github.com/pytorch/pytorch/issues/148475,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-03-06T00:53:51Z +issue,148199,[Inductor-CPU] qlinear_binary output may have undefined strides with dynamic shape support,https://github.com/pytorch/pytorch/issues/148199,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: dynamo, oncall: cpu inductor, oncall: pt2",review linked PR state before touching issue,2025-03-04T15:38:40Z +issue,147995,Checkpoint doesn't work with torch_function if torch_function change tensor metadata,https://github.com/pytorch/pytorch/issues/147995,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: activation checkpointing",review linked PR state before touching issue,2025-02-27T16:15:48Z +issue,139603,`torch.compile` doesn't support `set` arguments,https://github.com/pytorch/pytorch/issues/139603,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-27T00:25:25Z +issue,134950,"Bert model training is not running on cpu (X86/ARM) with torch.compile() mode, generating value error.",https://github.com/pytorch/pytorch/issues/134950,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-26T22:55:23Z +issue,145445,[dynamo] `random.Random` gives wrong result on second call,https://github.com/pytorch/pytorch/issues/145445,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-26T22:16:19Z +issue,147843,The opt_output in `fx_graph_runnable.py` is inconsistent with the actual output when testing run_repro(acc=True).,https://github.com/pytorch/pytorch/issues/147843,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-02-25T14:06:40Z +issue,122729,torch.func transforms don't yet support saved tensor hooks.,https://github.com/pytorch/pytorch/issues/122729,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: vmap,review linked PR state before touching issue,2025-02-25T11:36:45Z +issue,147530,[fx][dynamo][be] Don't allow arbitrary dataclass in the graph,https://github.com/pytorch/pytorch/issues/147530,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-25T06:29:37Z +issue,127863,Potential 7.5x compilation time improvement for foreach kernel,https://github.com/pytorch/pytorch/issues/127863,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: pt2 optimizer, oncall: pt2",review linked PR state before touching issue,2025-02-24T21:44:13Z +issue,147486,remote_cache.py loggings cause a logging error after pytest exit,https://github.com/pytorch/pytorch/issues/147486,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: logging, module: testing, oncall: pt2",review linked PR state before touching issue,2025-02-24T19:00:20Z +issue,147212,dict_tag optimization leads to wrong results with relational guards,https://github.com/pytorch/pytorch/issues/147212,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-24T15:14:03Z +issue,129456,[dynamic][inline-inbuilt-nn-modules] Extra compilation - pytest test/onnx/dynamo/test_dynamo_with_onnxruntime_backend.py::TestDynamoWithONNXRuntime::test_llama_attention_with_local_backend_0,https://github.com/pytorch/pytorch/issues/129456,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: onnx, oncall: pt2",review linked PR state before touching issue,2025-02-19T17:15:15Z +issue,128556,Dynamo graph breaks when passing a constant value to a nn module initialized within a function,https://github.com/pytorch/pytorch/issues/128556,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2025-02-18T23:29:31Z +issue,133462,PT2 inference slowdown after updating Huggingface pin on CI,https://github.com/pytorch/pytorch/issues/133462,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: performance, oncall: pt2",review linked PR state before touching issue,2025-02-14T22:38:32Z +issue,146951,[dynamo] `x is x` gets incorrectly interpreted to `False`,https://github.com/pytorch/pytorch/issues/146951,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-13T06:11:05Z +issue,146673,"Dynamo should handle `class.method_descriptor(instance, *args, **kwargs)` generically",https://github.com/pytorch/pytorch/issues/146673,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-08T01:28:56Z +issue,146628,Dynamo should consider tensor mutation when reconstructing generator,https://github.com/pytorch/pytorch/issues/146628,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-02-07T02:52:01Z +issue,145596,[Dynamo] compile torch.logit with different data types,https://github.com/pytorch/pytorch/issues/145596,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2025-01-27T18:11:54Z +issue,113490,/nodefaultlib:vcomp doesn't seem to be set when compiling with Intel OpenMP on Windows,https://github.com/pytorch/pytorch/issues/113490,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: build, module: windows",review linked PR state before touching issue,2025-01-27T12:22:51Z +issue,145529,Module.to() fail in dynamo when swap_module_params_on_conversion is true,https://github.com/pytorch/pytorch/issues/145529,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2025-01-24T06:47:14Z +issue,145231,Flaky Dynamo test: TestAutograd.test_gradcheck_nondeterministic,https://github.com/pytorch/pytorch/issues/145231,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: testing, oncall: pt2",review linked PR state before touching issue,2025-01-21T22:37:42Z +issue,143463,unreasonable ConstraintViolationError when using torch dynamo to compile torch model,https://github.com/pytorch/pytorch/issues/143463,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: dynamo, oncall: export, oncall: pt2",review linked PR state before touching issue,2025-01-07T19:13:58Z +issue,132133,Adding betainc with backward operation,https://github.com/pytorch/pytorch/issues/132133,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: special,review linked PR state before touching issue,2025-01-02T13:43:34Z +issue,142489,Stop importing arbitrary third-party libraries in Dynamo,https://github.com/pytorch/pytorch/issues/142489,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-12-27T23:50:53Z +issue,135759,core_aten_decompositions table have non-functional or CIA entries.,https://github.com/pytorch/pytorch/issues/135759,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: export, oncall: pt2",review linked PR state before touching issue,2024-12-17T22:09:34Z +issue,142378,Support int step for nonfused optimizer implementations,https://github.com/pytorch/pytorch/issues/142378,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,review linked PR state before touching issue,2024-12-09T16:30:25Z +issue,141916,rms_norm forward performance analysis,https://github.com/pytorch/pytorch/issues/141916,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-12-04T04:26:20Z +issue,137996,cprofile on compiler should trigger on backwards,https://github.com/pytorch/pytorch/issues/137996,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2024-12-02T17:55:39Z +issue,141744,"""RuntimeError: Triton Error [CUDA]: device kernel image is invalid"" when running generated Triton code",https://github.com/pytorch/pytorch/issues/141744,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor,review linked PR state before touching issue,2024-12-02T16:43:56Z +issue,113809,"Error during DDP, torch.compile, and cudagraph_trees",https://github.com/pytorch/pytorch/issues/113809,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: ddp, module: inductor, oncall: distributed",review linked PR state before touching issue,2024-11-26T23:51:38Z +issue,141473,How to use torch.compile + HF model?,https://github.com/pytorch/pytorch/issues/141473,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-11-26T04:21:05Z +issue,140909,[2.5.0+] Running a compiled model under FlopCounterMode regresses its performance,https://github.com/pytorch/pytorch/issues/140909,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-11-22T22:19:48Z +issue,141162,Dynamo should reorder warnings (and other logging) by default if they don't involve tensor values,https://github.com/pytorch/pytorch/issues/141162,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-11-22T03:14:59Z +issue,141017,multilabel_margin_loss gives incorrect results in torch.compile,https://github.com/pytorch/pytorch/issues/141017,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: pt2 accuracy, oncall: pt2",review linked PR state before touching issue,2024-11-22T02:54:54Z +issue,141258,`torch.compile` fails with runtime error on Numpy scalar operations,https://github.com/pytorch/pytorch/issues/141258,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: numpy, oncall: pt2",review linked PR state before touching issue,2024-11-21T23:17:20Z +issue,123972,Dynamo export: limited support in Torch.cond,https://github.com/pytorch/pytorch/issues/123972,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: onnx, oncall: pt2",review linked PR state before touching issue,2024-11-21T19:46:00Z +issue,140845,compiled LSTM performance is worse than not compiled,https://github.com/pytorch/pytorch/issues/140845,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: decompositions, oncall: pt2",review linked PR state before touching issue,2024-11-19T14:59:07Z +issue,140884,`torch.compile` tries to invoke a torch op on numpy.ndarray in nightly release,https://github.com/pytorch/pytorch/issues/140884,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-11-18T20:16:01Z +issue,140795,Memory usage increase post compilation for torch.compile ViT-H-14-quickgelu_dfn5b,https://github.com/pytorch/pytorch/issues/140795,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-11-18T18:36:09Z +issue,139629,Inductor should run fallbacks without Autograd-related dispatch keys,https://github.com/pytorch/pytorch/issues/139629,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-11-08T19:45:53Z +issue,139232,TORCH_COMPILE_CPROFILE=1 doesn't work on python 3.12,https://github.com/pytorch/pytorch/issues/139232,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-10-31T00:18:13Z +issue,139080,[dynamo] `dynamo_expected_failures` is silent on certain tests that ended up passing,https://github.com/pytorch/pytorch/issues/139080,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: tests, oncall: pt2",review linked PR state before touching issue,2024-10-30T23:44:17Z +issue,138926,torch.cond complains two tensors have different metadata,https://github.com/pytorch/pytorch/issues/138926,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",review linked PR state before touching issue,2024-10-29T15:11:32Z +issue,41245,torch.sign is divergent from numpy.sign on NaN,https://github.com/pytorch/pytorch/issues/41245,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: numpy,review linked PR state before touching issue,2024-10-19T11:06:27Z +issue,105839,"MPS memory issue, MPS backend out of memory, but works if I empty the MPS cache",https://github.com/pytorch/pytorch/issues/105839,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: mps",review linked PR state before touching issue,2024-10-17T12:02:04Z +issue,126474,[dynamo] Handle inplace op aliasing errors,https://github.com/pytorch/pytorch/issues/126474,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-10-16T18:26:21Z +issue,137546,`torch.linspace` doesn't agree with decomposition.,https://github.com/pytorch/pytorch/issues/137546,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: decompositions, module: linear algebra",review linked PR state before touching issue,2024-10-09T13:31:24Z +issue,137468,overflow in `torch.quantize_per_tensor`,https://github.com/pytorch/pytorch/issues/137468,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, oncall: quantization",review linked PR state before touching issue,2024-10-08T14:52:01Z +issue,93501,_pack_padded_sequence fails in dynamo due to requiring a non-fake 2nd argument,https://github.com/pytorch/pytorch/issues/93501,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: performance, oncall: pt2",review linked PR state before touching issue,2024-09-28T21:11:26Z +issue,136271,[triton x pt2] Dynamo should trace data_ptr accesses,https://github.com/pytorch/pytorch/issues/136271,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: user triton, oncall: pt2",review linked PR state before touching issue,2024-09-27T23:37:37Z +issue,136258,`to()` Method Does Not Move Internal Components of Sparse Tensors,https://github.com/pytorch/pytorch/issues/136258,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: sparse,review linked PR state before touching issue,2024-09-26T09:52:58Z +issue,120381,torch.compile has significantly higher peak memory for a simple checkpointed layer of matmul + layernorm + gelu,https://github.com/pytorch/pytorch/issues/120381,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-09-13T10:22:55Z +issue,121318,torch/export/_trace.py and related non-strict entrypoints do not initialize TORCH_LOGS/TORCH_TRACE,https://github.com/pytorch/pytorch/issues/121318,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: logging, oncall: export, oncall: pt2",review linked PR state before touching issue,2024-09-13T03:28:25Z +issue,120911,Symbolic shape tests don't actually test that arguments aren't specialized,https://github.com/pytorch/pytorch/issues/120911,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: tests, oncall: pt2",review linked PR state before touching issue,2024-09-13T03:16:40Z +issue,122294,AOIInductor: Dynamic Shapes Specificiaton fails for SAM,https://github.com/pytorch/pytorch/issues/122294,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",review linked PR state before touching issue,2024-09-13T03:06:30Z +issue,118747,decorate_context results in uninformative backtraces,https://github.com/pytorch/pytorch/issues/118747,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: logging, oncall: pt2",review linked PR state before touching issue,2024-09-13T03:05:17Z +issue,118332,Further simplifying guards,https://github.com/pytorch/pytorch/issues/118332,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2024-09-12T18:17:24Z +issue,135492,inductor guards on 32 bit indexing even when generated code would be the same,https://github.com/pytorch/pytorch/issues/135492,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-09-09T19:13:16Z +issue,123651,Divisibility substitution does the wrong thing with non-trivial max that would be < 2 after division,https://github.com/pytorch/pytorch/issues/123651,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2024-09-06T22:37:10Z +issue,134841,LPPool2d lacks a check for the validity of norm_type,https://github.com/pytorch/pytorch/issues/134841,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: norms and normalization",review linked PR state before touching issue,2024-08-30T15:28:39Z +issue,134363,Improving inductor pattern matcher's replacement graph assumptions,https://github.com/pytorch/pytorch/issues/134363,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-08-26T19:04:44Z +issue,94336,`scatter` fails the gradient computation in reverse mode for `src` when `index` is empty,https://github.com/pytorch/pytorch/issues/94336,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: scatter & gather ops",review linked PR state before touching issue,2024-08-21T09:46:51Z +issue,133493,vmap + autograd.Function [generate_vmap_rule=False] + torch.compile don't work together,https://github.com/pytorch/pytorch/issues/133493,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: functorch, oncall: pt2",review linked PR state before touching issue,2024-08-19T19:15:26Z +issue,118334,Using autograd.Functions defined in torch/ cause graph breaks,https://github.com/pytorch/pytorch/issues/118334,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-08-14T19:32:59Z +issue,133078,"`torch.compile` support for `CXX=""zig c++""`",https://github.com/pytorch/pytorch/issues/133078,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2024-08-10T00:44:58Z +issue,132368,Switch to new name `oneDNN` instead of `MKLDNN`,https://github.com/pytorch/pytorch/issues/132368,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: intel, module: mkldnn",review linked PR state before touching issue,2024-08-09T07:17:09Z +issue,131280,The shape of weights in lstm seems to be wrong.,https://github.com/pytorch/pytorch/issues/131280,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rnn",review linked PR state before touching issue,2024-08-06T03:46:44Z +issue,132020,`torch.Tensor.to` ignores `memory_format` kwarg,https://github.com/pytorch/pytorch/issues/132020,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: memory format, module: python frontend",review linked PR state before touching issue,2024-07-29T18:38:25Z +issue,126673,torch.nn.AdaptiveAvgPool2d lacks checking of input dimension,https://github.com/pytorch/pytorch/issues/126673,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn",review linked PR state before touching issue,2024-07-29T07:28:29Z +issue,103417,"Torch model compile error ""/usr/bin/ld: cannot find -lcuda"" though cuda is installed via run file",https://github.com/pytorch/pytorch/issues/103417,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,review linked PR state before touching issue,2024-07-27T17:51:11Z +issue,130641,`dims` of `flip()` without `list` or `tuple` works with a tensor agaist the doc,https://github.com/pytorch/pytorch/issues/130641,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label,review linked PR state before touching issue,2024-07-13T01:44:57Z +issue,129936,[symbolic-shapes] Guard validation in Dynamo changes the shape_env,https://github.com/pytorch/pytorch/issues/129936,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2024-07-09T16:33:53Z +issue,129970,Setting the 0D tensor of only one element to `fill_value` of `full()`/`full_like()` works against error message,https://github.com/pytorch/pytorch/issues/129970,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: error checking, module: python frontend",review linked PR state before touching issue,2024-07-08T13:15:04Z +issue,111320,test_max_pool1d reliably OOMs after https://github.com/pytorch/pytorch/pull/111216/,https://github.com/pytorch/pytorch/issues/111320,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, oncall: pt2",review linked PR state before touching issue,2024-06-30T11:49:05Z +issue,112788,async_compile.triton is unable to cache generated triton kernels due to meta containing kernel names,https://github.com/pytorch/pytorch/issues/112788,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-06-28T19:01:23Z +issue,129338,`linspace()` can also use `complex` and `bool` type for `start` and `end` argument against the doc,https://github.com/pytorch/pytorch/issues/129338,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs,review linked PR state before touching issue,2024-06-24T15:51:03Z +issue,102839,Dynamo should only unroll loops by a preset factor (unless otherwise explicitly instructed),https://github.com/pytorch/pytorch/issues/102839,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-06-18T18:49:39Z +issue,117265,[dynamo][inline-inbuilt-nn-modules]torch.compile silently incorrect with full_backward_pre_hook,https://github.com/pytorch/pytorch/issues/117265,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-06-18T16:56:47Z +issue,124163,Support `dynamic=True` in torch._dynamo.explain,https://github.com/pytorch/pytorch/issues/124163,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-06-17T23:52:44Z +issue,123411,dynamo/fx doesn't honor 'non-persistent' buffers,https://github.com/pytorch/pytorch/issues/123411,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-06-17T23:44:44Z +issue,123374,[dynamo] incorrect error traceback for runtime errors when executing dynamo codegen,https://github.com/pytorch/pytorch/issues/123374,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-06-17T23:43:29Z +issue,122578,Recompiles and cache_size_limit from detectron2 CycleBatchNormList,https://github.com/pytorch/pytorch/issues/122578,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",review linked PR state before touching issue,2024-06-17T23:40:15Z +issue,126642,"[optim] add missing 'maximize' parameter to LBFGS, NAdam and RAdam optimizers",https://github.com/pytorch/pytorch/issues/126642,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: loss, module: optimizer",review linked PR state before touching issue,2024-06-03T15:17:58Z +issue,38019,"torch.cuda.nccl.init_rank does not handle ""uid"" properly, causing runtime error",https://github.com/pytorch/pytorch/issues/38019,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nccl",review linked PR state before touching issue,2024-06-02T10:21:34Z +issue,121526,Compiling RMSNorm Triton Kernal gives error,https://github.com/pytorch/pytorch/issues/121526,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: user triton, oncall: pt2",review linked PR state before touching issue,2024-05-22T06:54:40Z +issue,126514,"Weights become NaN with torch.compile optimizer capturable=True, lr=0.0, nn.Embedding",https://github.com/pytorch/pytorch/issues/126514,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: pt2",review linked PR state before touching issue,2024-05-20T22:09:44Z +issue,125277,H100 Max Autotune Compilation Significantly Slower than A100,https://github.com/pytorch/pytorch/issues/125277,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-05-14T18:43:02Z +issue,125236,torch.no_grad() is not working for dynamo inductor backend,https://github.com/pytorch/pytorch/issues/125236,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-05-06T18:36:37Z +issue,125388,torch.multinomial raises no error when sampling from zero weight and replacement=False.,https://github.com/pytorch/pytorch/issues/125388,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label,review linked PR state before touching issue,2024-05-03T00:40:42Z +issue,124110,Equality between SymBool is extremely busted,https://github.com/pytorch/pytorch/issues/124110,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",review linked PR state before touching issue,2024-04-18T16:29:35Z +issue,105560,aten._trilinear,https://github.com/pytorch/pytorch/issues/105560,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2024-03-28T23:16:52Z +issue,122191,torch.std_mean slower than separate torch.mean and torch.std calls on CPU,https://github.com/pytorch/pytorch/issues/122191,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",review linked PR state before touching issue,2024-03-20T07:52:28Z +issue,120819,Support a context manager to set torch dtype?,https://github.com/pytorch/pytorch/issues/120819,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,review linked PR state before touching issue,2024-03-06T18:36:46Z +issue,119883,ModularIndexing does not have well-defined semantics,https://github.com/pytorch/pytorch/issues/119883,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor,review linked PR state before touching issue,2024-02-14T14:19:16Z +issue,119300,Optimizer step pre/post hook tests should try harder to ascertain the hook happens before/after the step call,https://github.com/pytorch/pytorch/issues/119300,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,review linked PR state before touching issue,2024-02-06T19:16:08Z +issue,111471,test_learnable_forward_per_channel fails due to integer overflow,https://github.com/pytorch/pytorch/issues/111471,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,review linked PR state before touching issue,2023-12-21T21:09:00Z +issue,104860,torch.nn.Conv2d/Conv1d's padding mode circular cannot accept 3-dim input,https://github.com/pytorch/pytorch/issues/104860,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: nn,review linked PR state before touching issue,2023-12-11T08:53:04Z +issue,93617,[Inductor] Input Buffers Should Be Representable As Storage And Layout,https://github.com/pytorch/pytorch/issues/93617,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",review linked PR state before touching issue,2023-11-30T04:07:16Z +issue,78618,Module parameters/submodules can be shadowed by class attributes silently,https://github.com/pytorch/pytorch/issues/78618,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: nn,review linked PR state before touching issue,2023-11-07T10:33:10Z +issue,110865,BCEWithLogitsLoss: Check if labels / targets are within zero and one,https://github.com/pytorch/pytorch/issues/110865,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: nn,review linked PR state before touching issue,2023-10-26T15:55:52Z +issue,105318,[docs] torch.sigmoid to make clear equivalence relations to other sigmoid functions,https://github.com/pytorch/pytorch/issues/105318,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs,review linked PR state before touching issue,2023-07-24T18:36:15Z +issue,103584,torch.cuda.mem_get_info to return 0 if CUDA context isn't initialized,https://github.com/pytorch/pytorch/issues/103584,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cuda,review linked PR state before touching issue,2023-06-20T20:22:07Z +issue,99421,Question about GRU(RNN/LSTM) outputs shape,https://github.com/pytorch/pytorch/issues/99421,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: rnn",review linked PR state before touching issue,2023-04-21T19:13:49Z +issue,97750,irrelevant error output for Minified repro,https://github.com/pytorch/pytorch/issues/97750,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: minifier, oncall: pt2",review linked PR state before touching issue,2023-03-30T21:14:22Z +issue,97310,MPS: `unique` and `unique_consecutive` extremely slow when `return_counts=True`,https://github.com/pytorch/pytorch/issues/97310,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",review linked PR state before touching issue,2023-03-23T18:06:57Z +issue,70925,allow `dim=None` in `concat`,https://github.com/pytorch/pytorch/issues/70925,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: viewing and reshaping",review linked PR state before touching issue,2023-03-07T09:33:27Z +issue,70915,`arange` should return empty array if bounds are inconsistent with step sign,https://github.com/pytorch/pytorch/issues/70915,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: tensor creation",review linked PR state before touching issue,2023-03-07T09:32:09Z +issue,86558,Data conversion ops ignore `memory_format=torch.contiguous_format`,https://github.com/pytorch/pytorch/issues/86558,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: primtorch,review linked PR state before touching issue,2022-10-11T05:49:35Z +issue,85960,Discrepancy in output shape for batch_norm inference mode between CUDA and CPU,https://github.com/pytorch/pytorch/issues/85960,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",review linked PR state before touching issue,2022-10-04T17:13:57Z +issue,80301,"Need ""valid"" and ""same"" padding mode for convTranspose2d",https://github.com/pytorch/pytorch/issues/80301,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn, module: padding",review linked PR state before touching issue,2022-06-28T15:58:26Z +issue,51455,"[docs] Improve documentation for LayerNorm, GroupNorm, etc (+ add python reference impl)",https://github.com/pytorch/pytorch/issues/51455,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: norms and normalization",review linked PR state before touching issue,2022-05-25T18:34:20Z +issue,77170,Move the MPSGuardImpl to inherit from NoOpDeviceGuardImpl,https://github.com/pytorch/pytorch/issues/77170,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: backend, module: mps",review linked PR state before touching issue,2022-05-11T14:24:09Z +issue,71392,Error in `torch.trapz` documentation,https://github.com/pytorch/pytorch/issues/71392,has linked PR,26,open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs,review linked PR state before touching issue,2022-01-27T16:05:20Z +issue,67969,"[docs] F.multilabel_soft_margin_loss does not list arg reduction, but lists outdated size_average/reduce (true for other losses as well)",https://github.com/pytorch/pytorch/issues/67969,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",review linked PR state before touching issue,2021-11-18T17:37:11Z +issue,67887,Add more explanation on multithreaded graph building of Autograd,https://github.com/pytorch/pytorch/issues/67887,has linked PR,26,"open PR references this issue; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs, module: multiprocessing, module: multithreading",review linked PR state before touching issue,2021-11-10T10:39:30Z +pr,186435,Fix mkldnn FakeTensor error handling,https://github.com/pytorch/pytorch/pull/186435,draft/noise,23,PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-07-08T11:53:23Z +pr,189006,Ensure correct sign of input for test_laplace in test_distributions.py,https://github.com/pytorch/pytorch/pull/189006,draft/noise,23,PR is draft; PR closes or references open issue numbers,wait for author to mark ready,2026-07-06T09:20:20Z +pr,186778,Reduce fp16 batch-1 Conv2d forward divergence on XPU,https://github.com/pytorch/pytorch/pull/186778,draft/noise,23,PR is draft; PR closes or references open issue numbers,wait for author to mark ready,2026-07-06T07:48:06Z +pr,180857,Add variable-length support for Flash Attention in XPU,https://github.com/pytorch/pytorch/pull/180857,draft/noise,23,PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-30T11:20:01Z +pr,188143,Add mixed Tensor and Scalar bound support for torch.clamp and torch.clip,https://github.com/pytorch/pytorch/pull/188143,draft/noise,23,PR is draft; PR closes or references open issue numbers,wait for author to mark ready,2026-06-29T14:05:29Z +pr,187930,Fix torch.compile validation for random ops,https://github.com/pytorch/pytorch/pull/187930,draft/noise,23,PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-28T02:34:01Z +pr,187409,[distributed] Add NCCL M2N-backed symm_mem reshard,https://github.com/pytorch/pytorch/pull/187409,draft/noise,23,"PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan",wait for author to mark ready,2026-06-24T23:55:59Z +pr,186089,[XPU] Align XPU unit tests decorators and overrides.,https://github.com/pytorch/pytorch/pull/186089,draft/noise,23,PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/trunk,wait for author to mark ready,2026-06-24T11:40:04Z +pr,187896,Re-land chain-based dispatch for custom_op fast path with override detection,https://github.com/pytorch/pytorch/pull/187896,draft/noise,23,PR is draft; PR closes or references open issue numbers,wait for author to mark ready,2026-06-23T01:36:15Z +pr,187554,Fix flaky gradcheck in _gradcheck_log_prob for simplex distributions.,https://github.com/pytorch/pytorch/pull/187554,draft/noise,23,PR is draft; PR closes or references open issue numbers,wait for author to mark ready,2026-06-17T12:08:55Z +pr,183530,feat: add XPUCodeCache tests for compilation and execution,https://github.com/pytorch/pytorch/pull/183530,draft/noise,23,"PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-17T12:02:41Z +pr,183366,[FSDP] Support replicated shard placement params,https://github.com/pytorch/pytorch/pull/183366,draft/noise,23,"PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-16T05:40:51Z +pr,182147,Support XPU and other accelerators in RNG state handling,https://github.com/pytorch/pytorch/pull/182147,draft/noise,23,"PR is draft; PR closes or references open issue numbers; CI lane labels: ciflow/inductor, ciflow/trunk",wait for author to mark ready,2026-06-08T11:40:17Z +pr,186068,Add skipIfXpu decorator to tests for semi-structured sparsity support,https://github.com/pytorch/pytorch/pull/186068,draft/noise,23,PR is draft; PR closes or references open issue numbers,wait for author to mark ready,2026-06-03T10:46:23Z +pr,183875,Add claude skill for PrivateUse1 (PU1) third-party device integration,https://github.com/pytorch/pytorch/pull/183875,draft/noise,23,PR is draft; review decision requests changes; PR has not updated for at least 45 days,wait for contributor update,2026-05-20T02:07:32Z +pr,179565,[WIP][CI] Add RHEL 9.6 build and test support,https://github.com/pytorch/pytorch/pull/179565,draft/noise,23,"PR is draft; review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for contributor update,2026-05-15T05:47:27Z +pr,180225,Enable ACL operator fusions for aarch64,https://github.com/pytorch/pytorch/pull/180225,draft/noise,23,"PR is draft; review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/linux-aarch64",wait for contributor update,2026-05-11T09:07:09Z +pr,97825,Support sparse COO/CSR/CSC/BSR/BSC return values in gradcheck input function,https://github.com/pytorch/pytorch/pull/97825,draft/noise,23,PR is draft; review decision requests changes; PR has not updated for at least 45 days,wait for contributor update,2025-06-19T08:46:33Z +pr,140372,[Intel GPU] Enable mkldnn::_convolution.pointwise at XPU backend,https://github.com/pytorch/pytorch/pull/140372,draft/noise,23,PR is draft; review decision requests changes; PR has not updated for at least 45 days; CI lane labels: ciflow/xpu,wait for contributor update,2025-05-24T06:21:32Z +issue,93419,"torchdynamo is not properly setting up input tracking (e.g., for symbolic shape guards) for view bases",https://github.com/pytorch/pytorch/issues/93419,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",route to design/research decision before patch planning,2026-03-31T23:18:40Z +issue,50112,need a clear guide for when and how to use torch.cuda.set_device(),https://github.com/pytorch/pytorch/issues/50112,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",route to design/research decision before patch planning,2025-12-26T12:57:46Z +issue,170944,Indexing a tensor with numpy boolean array is very slow,https://github.com/pytorch/pytorch/issues/170944,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2025-12-22T18:24:45Z +issue,83529,Adding Levenberg-marquardt optimizer in PyTorch,https://github.com/pytorch/pytorch/issues/83529,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2025-12-08T14:18:29Z +issue,160553,[MPS] Out of bounds indexing inconsistency between MPS and CPU/CUDA,https://github.com/pytorch/pytorch/issues/160553,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",route to design/research decision before patch planning,2025-12-02T09:28:54Z +issue,156424,research adding cuda-bindings to core,https://github.com/pytorch/pytorch/issues/156424,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,route to design/research decision before patch planning,2025-11-18T18:31:46Z +issue,130451,vmap over torch.allclose doesn't work,https://github.com/pytorch/pytorch/issues/130451,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,route to design/research decision before patch planning,2025-11-17T13:31:48Z +issue,158043,Add GoLU Activation Function 🚀,https://github.com/pytorch/pytorch/issues/158043,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-11-06T14:41:14Z +issue,99873,Dynamo config patching in our code is brittle,https://github.com/pytorch/pytorch/issues/99873,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",route to design/research decision before patch planning,2025-11-05T19:40:03Z +issue,166826,PyTorch Layer for Pre-Summation Nonlinear Feature Transformation,https://github.com/pytorch/pytorch/issues/166826,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend",route to design/research decision before patch planning,2025-11-04T19:29:44Z +issue,6662,Autogenerate code example / tutorial outputs in documentation,https://github.com/pytorch/pytorch/issues/6662,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2025-10-27T15:13:59Z +issue,166208,OptimizerDict - design and write an abstraction for a single Optim that'd control multiple optimizers,https://github.com/pytorch/pytorch/issues/166208,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2025-10-27T14:09:30Z +issue,165577,CI: What is the purpose of `slow.yml`,https://github.com/pytorch/pytorch/issues/165577,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: ci,route to design/research decision before patch planning,2025-10-16T09:24:45Z +issue,83264,Emulating FP64 and increased precisions on Apple silicon,https://github.com/pytorch/pytorch/issues/83264,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: mps,route to design/research decision before patch planning,2025-10-04T13:07:30Z +issue,24826,Transformer Lack of Embedding Layer and Positional Encodings,https://github.com/pytorch/pytorch/issues/24826,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",route to design/research decision before patch planning,2025-09-29T23:14:42Z +issue,65392,torch.sparse.sum result has wrong dtype when reducing over all dimensions,https://github.com/pytorch/pytorch/issues/65392,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: sparse",route to design/research decision before patch planning,2025-09-27T03:36:32Z +issue,55152,Better handling of OpInfo.sample_inputs,https://github.com/pytorch/pytorch/issues/55152,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: tests,route to design/research decision before patch planning,2025-09-26T23:51:21Z +issue,84347,Lack of newly raised optimizers,https://github.com/pytorch/pytorch/issues/84347,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2025-09-26T23:37:46Z +issue,68332,[feature request] Provide functional form of scheduler formulas (and reconsider older decisions of not doing it),https://github.com/pytorch/pytorch/issues/68332,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",route to design/research decision before patch planning,2025-09-19T01:43:33Z +issue,160637,Support optimisation of sparse parameters via LBFGS,https://github.com/pytorch/pytorch/issues/160637,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: sparse",route to design/research decision before patch planning,2025-09-17T15:58:39Z +issue,141994,Add torch.set_default_int_dtype() / extend set_default_dtype() to allow setting default signed integer dtype,https://github.com/pytorch/pytorch/issues/141994,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2025-09-16T18:25:56Z +issue,153669,Make tlparse able to show a summary of distinct graph breaks,https://github.com/pytorch/pytorch/issues/153669,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-09-12T20:16:04Z +issue,160322,`torch.Tensor` methods return type annotation too broad for `torch.Tensor` subclasses,https://github.com/pytorch/pytorch/issues/160322,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: typing,route to design/research decision before patch planning,2025-09-02T03:09:22Z +issue,161129,autograd assertion error in cuda backend when privateuse1 is renamed,https://github.com/pytorch/pytorch/issues/161129,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: privateuse1",route to design/research decision before patch planning,2025-08-29T20:17:25Z +issue,160489,[CI] Mergebot tests are no longer updatable,https://github.com/pytorch/pytorch/issues/160489,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",route to design/research decision before patch planning,2025-08-26T20:06:48Z +issue,89116,Implement generic batch normalization layer.,https://github.com/pytorch/pytorch/issues/89116,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, oncall: distributed",route to design/research decision before patch planning,2025-08-25T10:22:07Z +issue,109691,Extends the functionality of `nn.BatchNorm1d`.,https://github.com/pytorch/pytorch/issues/109691,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, oncall: distributed",route to design/research decision before patch planning,2025-08-25T10:21:04Z +issue,158709,autograd precision: float32 gradients of bfloat16 weights,https://github.com/pytorch/pytorch/issues/158709,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2025-08-20T16:29:03Z +issue,125422,Tensor roll with different shifts,https://github.com/pytorch/pytorch/issues/125422,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy,route to design/research decision before patch planning,2025-08-11T19:08:53Z +issue,159765,Add nn.GradBank for gradient scaling to prevent vanishing/exploding gradients,https://github.com/pytorch/pytorch/issues/159765,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-08-08T18:54:19Z +issue,112303,"[dynamo]RFC/Feature request - cache, guard, and reuse InliningInstructionTranslator inline_call",https://github.com/pytorch/pytorch/issues/112303,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",route to design/research decision before patch planning,2025-08-01T22:16:05Z +issue,80168,GEGLU activation,https://github.com/pytorch/pytorch/issues/80168,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-07-30T18:41:46Z +issue,130513,RuntimeError Occurred In getBar1SizeOfGpu when initializing torch RPC,https://github.com/pytorch/pytorch/issues/130513,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: tensorpipe, oncall: distributed",route to design/research decision before patch planning,2025-07-30T14:08:49Z +issue,155322,Segmentation fault after manual tensor assignment with autograd enabled on CUDA,https://github.com/pytorch/pytorch/issues/155322,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",route to design/research decision before patch planning,2025-07-29T14:18:11Z +issue,104849,"[feature request] torch.mix function to generalize/symmetrize addcmul + can serve as a more versatlie, generalized ""optimization update rule"" function",https://github.com/pytorch/pytorch/issues/104849,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2025-07-29T12:41:47Z +issue,104247,"[proposal] ""Name"" string attribute for modules, parameters, buffers, tensors for more pleasant debugging (including distributed FSDP2 logging)",https://github.com/pytorch/pytorch/issues/104247,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2025-07-29T12:40:35Z +issue,158704,Enable DataLoaders to prefetch batches across epochs,https://github.com/pytorch/pytorch/issues/158704,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,route to design/research decision before patch planning,2025-07-23T18:44:54Z +issue,69431,"[feature request] `torch.to(obj, device, dtype)` supporting recursive lists/dicts/tuples of tensors probably by uplifting/promoting `torch.distributed.utils._recursive_to`",https://github.com/pytorch/pytorch/issues/69431,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2025-07-19T20:30:25Z +issue,157547,Several `torch.*` functions raise uninformative `NotImplementedError`s when called with integer `dtype`,https://github.com/pytorch/pytorch/issues/157547,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: error checking,route to design/research decision before patch planning,2025-07-14T17:18:48Z +issue,150742,Install pytorch from pypi using local CUDA build,https://github.com/pytorch/pytorch/issues/150742,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: ci, oncall: releng",route to design/research decision before patch planning,2025-07-09T17:01:38Z +issue,61654,Allow creation of pseudo devices for testing purposes,https://github.com/pytorch/pytorch/issues/61654,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2025-07-09T13:55:56Z +issue,141627,Support reductions in FlexAttention's score_mod/mask_mod,https://github.com/pytorch/pytorch/issues/141627,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2025-07-08T00:06:21Z +issue,154260,Pypi Support for Windows arm64,https://github.com/pytorch/pytorch/issues/154260,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: windows, oncall: releng",route to design/research decision before patch planning,2025-06-30T12:12:45Z +issue,155983,as_tensor of list of tensors should keep grad history,https://github.com/pytorch/pytorch/issues/155983,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2025-06-28T17:22:17Z +issue,154767,Turn gradient_accumulate into a separate aten op,https://github.com/pytorch/pytorch/issues/154767,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2025-06-28T15:07:49Z +issue,152877,Motivate Pytorch's forward mode AD APIs with training examples,https://github.com/pytorch/pytorch/issues/152877,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs, module: forward ad",route to design/research decision before patch planning,2025-06-28T15:07:02Z +issue,155986,Vulkan interoperability,https://github.com/pytorch/pytorch/issues/155986,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: vulkan,route to design/research decision before patch planning,2025-06-17T17:10:23Z +issue,154906,load_state_dict with strict should be able to remove_duplicate?,https://github.com/pytorch/pytorch/issues/154906,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-06-13T21:45:37Z +issue,155368,Change error to warnning when tensor size mismatch while loading params,https://github.com/pytorch/pytorch/issues/155368,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-06-13T19:21:33Z +issue,155878,RFC: Add Adaptive Entropy-Gated Contrastive Fusion (AECF) for Robust Multimodal Attention Pooling,https://github.com/pytorch/pytorch/issues/155878,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sdpa",route to design/research decision before patch planning,2025-06-13T15:47:43Z +issue,122680,Please consider adding MIG (MI-rror with G-radient modification) to torch.nn,https://github.com/pytorch/pytorch/issues/122680,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-06-08T10:32:43Z +issue,111441,torch.compile of simple loop takes 34 seconds,https://github.com/pytorch/pytorch/issues/111441,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: higher order operators, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2025-06-04T02:23:26Z +issue,150698,[Feature Request] Memory optimization for backward propagation in GPU,https://github.com/pytorch/pytorch/issues/150698,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",route to design/research decision before patch planning,2025-06-03T19:08:45Z +issue,149133,Implement einsum backprop rather than decomposing,https://github.com/pytorch/pytorch/issues/149133,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: python frontend",route to design/research decision before patch planning,2025-06-03T15:52:54Z +issue,115937,Building extensions with CMake,https://github.com/pytorch/pytorch/issues/115937,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions",route to design/research decision before patch planning,2025-05-27T12:52:27Z +issue,153854,Better padding API,https://github.com/pytorch/pytorch/issues/153854,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",route to design/research decision before patch planning,2025-05-20T01:06:22Z +issue,153566,Improve PyTorch Logging to Distinguish ROCm from CUDA,https://github.com/pytorch/pytorch/issues/153566,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: rocm,route to design/research decision before patch planning,2025-05-15T20:23:45Z +issue,152875,Incorporate CUDA Memory Trimming Into DeviceCachingAllocator,https://github.com/pytorch/pytorch/issues/152875,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: windows",route to design/research decision before patch planning,2025-05-13T23:10:32Z +issue,153378,MPS backend silently ignores dimension mismatch,https://github.com/pytorch/pytorch/issues/153378,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",route to design/research decision before patch planning,2025-05-13T02:03:01Z +issue,118225,Support for banded matrix operations,https://github.com/pytorch/pytorch/issues/118225,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,route to design/research decision before patch planning,2025-05-12T08:14:06Z +issue,153035,Add Split Softmax,https://github.com/pytorch/pytorch/issues/153035,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-05-07T19:12:02Z +issue,150432,pytorch pip install instructions: always include the cuda index,https://github.com/pytorch/pytorch/issues/150432,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: releng",route to design/research decision before patch planning,2025-05-04T14:50:30Z +issue,151722,Add maxcount Parameter to torch.unique and torch.unique_consecutive,https://github.com/pytorch/pytorch/issues/151722,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: performance, module: python frontend",route to design/research decision before patch planning,2025-04-28T18:40:11Z +issue,101699,[discussion] [feature request] Native tensor-backed array of strings (at least a read-only one) and basic string processing functions for addition into core + discussion of extremely basic data frames (also for reducing python object heap pressure),https://github.com/pytorch/pytorch/issues/101699,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2025-04-16T08:16:38Z +issue,117806,Supporting custom attributes with `__torch_function__` tensor subclasses,https://github.com/pytorch/pytorch/issues/117806,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: dynamo",scan when time allows,2025-04-11T21:54:41Z +issue,25883,Python/C++ API Parity: torch.nn modules and functional,https://github.com/pytorch/pytorch/issues/25883,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2025-04-11T20:00:45Z +issue,149118,Add dim parameter to torch.bucketize,https://github.com/pytorch/pytorch/issues/149118,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2025-04-09T20:33:12Z +issue,4132,x.grad should be 0 but get NaN after x/0,https://github.com/pytorch/pytorch/issues/4132,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs, module: nans and infs",route to design/research decision before patch planning,2025-04-01T14:38:32Z +issue,132446,[Feature Request] Support Custom CPU Allocator to Enable torch.cpu.change_current_allocator,https://github.com/pytorch/pytorch/issues/132446,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu,route to design/research decision before patch planning,2025-04-01T07:36:13Z +issue,148843,make `to_empty` a no-op if parameter/buffer already on `device`,https://github.com/pytorch/pytorch/issues/148843,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2025-03-15T20:41:29Z +issue,144228,grad_fn function disobeys broadcast rules,https://github.com/pytorch/pytorch/issues/144228,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2025-03-04T18:27:51Z +issue,148345,Installation of `pytorch==2.6.0+cu124` doesn't install `triton` and `nvidia` libraries,https://github.com/pytorch/pytorch/issues/148345,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,route to design/research decision before patch planning,2025-03-03T18:10:28Z +issue,147437,"To avoid default eps ineffective in some precision(float16), dynamically set the default value of eps in optimizers such as Adam based on precision.",https://github.com/pytorch/pytorch/issues/147437,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2025-02-26T16:24:01Z +issue,146980,Feature Request: Add dlfloat16 support as a new dtype,https://github.com/pytorch/pytorch/issues/146980,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2025-02-16T15:45:20Z +issue,147048,Feature Request: rsample for Von Mises distribution,https://github.com/pytorch/pytorch/issues/147048,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: distributions,route to design/research decision before patch planning,2025-02-13T18:23:37Z +issue,144159,Allow generic python data structure input for torch.autograd.Function,https://github.com/pytorch/pytorch/issues/144159,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2025-02-04T01:51:28Z +issue,145739,No broadcasting by default.,https://github.com/pytorch/pytorch/issues/145739,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2025-02-03T19:09:13Z +issue,144536,[bug report template format] Simplify version information with HTML tags,https://github.com/pytorch/pytorch/issues/144536,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: collect_env.py,route to design/research decision before patch planning,2025-01-14T02:16:08Z +issue,120589,Softmax plus one for Flash attention,https://github.com/pytorch/pytorch/issues/120589,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: sdpa,route to design/research decision before patch planning,2025-01-12T08:32:20Z +issue,84782,macOS Pyinstaller: libc++abi: terminating with uncaught exception of type c10::Error: Type c10::intrusive_ptr> could not be converted to any of the known types,https://github.com/pytorch/pytorch/issues/84782,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: m1, module: third_party",route to design/research decision before patch planning,2025-01-07T19:23:23Z +issue,136433,SAC doesn't support nesting with different recompute plans,https://github.com/pytorch/pytorch/issues/136433,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd",route to design/research decision before patch planning,2025-01-06T02:16:57Z +issue,138118,[autograd] do not clear all variables when retain_graph=False,https://github.com/pytorch/pytorch/issues/138118,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2025-01-06T02:16:07Z +issue,140378,Differentiable quadratic programing solver,https://github.com/pytorch/pytorch/issues/140378,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2025-01-06T02:06:23Z +issue,142320,Proposal for Official Cross-Platform Support for 3D Gaussian Splatting,https://github.com/pytorch/pytorch/issues/142320,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-12-23T18:08:35Z +issue,120189,Making Mamba first-class citizen in PyTorch,https://github.com/pytorch/pytorch/issues/120189,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",route to design/research decision before patch planning,2024-12-14T22:33:49Z +issue,142381,Add a linter for Docker images which are not included in docker-builds.yml,https://github.com/pytorch/pytorch/issues/142381,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: ci,route to design/research decision before patch planning,2024-12-09T17:11:20Z +issue,141678,torch.cuda adds support for the changing `CUDA_VISIBLE_DEVICES`,https://github.com/pytorch/pytorch/issues/141678,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: third_party",route to design/research decision before patch planning,2024-11-30T20:36:52Z +issue,44945,Fuse softmax and masking in MultiheadAttention,https://github.com/pytorch/pytorch/issues/44945,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-11-27T19:47:15Z +issue,35458,Shared-QK transformer for the transformer (nn.activation.MultiheadAttention) module in PyTorch?,https://github.com/pytorch/pytorch/issues/35458,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-11-27T19:47:12Z +issue,74363,"Efficient very large (e.g., 31x31) depth-wise convolution",https://github.com/pytorch/pytorch/issues/74363,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution,route to design/research decision before patch planning,2024-11-20T05:33:00Z +issue,50566,Enhance supported fill value type for constant pad in functional.pad,https://github.com/pytorch/pytorch/issues/50566,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",route to design/research decision before patch planning,2024-11-10T01:44:00Z +issue,138952,[Feature request] Flag to prevent certain parameters from being typecast by nn.module.to(),https://github.com/pytorch/pytorch/issues/138952,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-11-08T00:39:28Z +issue,138502,Contribution Proposal: ScatterND Implementation for PyTorch,https://github.com/pytorch/pytorch/issues/138502,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,route to design/research decision before patch planning,2024-11-04T18:14:38Z +issue,111538,Propose to add constant padding mode to the `torch.nn.functional.grid_sample` function,https://github.com/pytorch/pytorch/issues/111538,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-11-03T03:22:28Z +issue,106360,Improving save_on_cpu's performance by overlapping memory transfers with compute,https://github.com/pytorch/pytorch/issues/106360,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: distributed",route to design/research decision before patch planning,2024-10-31T15:58:23Z +issue,71386,Direct Implementation of K-Nearest neighbor (KNN) in pytorch,https://github.com/pytorch/pytorch/issues/71386,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-10-29T05:23:51Z +issue,138906,"RuntimeError: ""addmm_cuda"" not implemented for 'ComplexHalf'",https://github.com/pytorch/pytorch/issues/138906,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cuda",route to design/research decision before patch planning,2024-10-28T15:46:40Z +issue,137680,[FEA] Add nvsci support in pytorch multiprocessing module for nvidia jetson devices.,https://github.com/pytorch/pytorch/issues/137680,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,route to design/research decision before patch planning,2024-10-17T06:27:09Z +issue,135816,Can we remove @torch._dynamo_disable from torch.utils.checkpoint.checkpoint?,https://github.com/pytorch/pytorch/issues/135816,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",route to design/research decision before patch planning,2024-09-24T06:12:00Z +issue,113530,"Make ""torch.load"" multi threaded process",https://github.com/pytorch/pytorch/issues/113530,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,route to design/research decision before patch planning,2024-09-20T16:59:41Z +issue,43101,New function for CTC decoding,https://github.com/pytorch/pytorch/issues/43101,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-09-14T12:00:32Z +issue,107224,"[CPP API] Add Adadelta, Adamax, ASGD, NAdam, RAdam and Rprop",https://github.com/pytorch/pytorch/issues/107224,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: optimizer",route to design/research decision before patch planning,2024-09-01T18:44:21Z +issue,133288,Give PrivateUse2 and PrivateUse3 the same treatment as PrivateUse1,https://github.com/pytorch/pytorch/issues/133288,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: backend, module: privateuse1",route to design/research decision before patch planning,2024-08-29T14:17:01Z +issue,133566,Performance: Autodiff stores inputs of convolution layers with non-trainable weight,https://github.com/pytorch/pytorch/issues/133566,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",route to design/research decision before patch planning,2024-08-29T13:44:17Z +issue,134325,Proposal: When serializing to state_dict/parameters use different separators for ParameterList and ParameterDict,https://github.com/pytorch/pytorch/issues/134325,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization",route to design/research decision before patch planning,2024-08-26T22:43:19Z +issue,133466,"In-place operations should be deprecated and in the future, they should be removed from PyTorch.",https://github.com/pytorch/pytorch/issues/133466,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-08-16T20:35:14Z +issue,132378,per sample gradient with torch.autograd.grad,https://github.com/pytorch/pytorch/issues/132378,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: vmap",route to design/research decision before patch planning,2024-08-13T22:19:06Z +issue,129867,`saved_tensors_hooks`/`_pop_saved_tensors_default_hooks` API extension,https://github.com/pytorch/pytorch/issues/129867,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd, module: multithreading",route to design/research decision before patch planning,2024-08-13T15:38:13Z +issue,105465,[proposal] Bit ops: e.g. setbit/getbit/togglebit/byteswap,https://github.com/pytorch/pytorch/issues/105465,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2024-08-08T20:48:17Z +issue,131881,Decouple some generic APIs with CUDA building,https://github.com/pytorch/pytorch/issues/131881,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,route to design/research decision before patch planning,2024-08-08T07:00:36Z +issue,128380,Conformal Prediction framework to enhance reliability in risk sensitive industrial applications,https://github.com/pytorch/pytorch/issues/128380,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2024-08-07T20:46:50Z +issue,131927,exclude_dim=[...] as a keyword argument to reduce type operations,https://github.com/pytorch/pytorch/issues/131927,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: reductions,route to design/research decision before patch planning,2024-07-29T04:35:01Z +issue,128283,control output type in matmul,https://github.com/pytorch/pytorch/issues/128283,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2024-07-25T13:20:01Z +issue,130428,Extend the device context manager to support existing tensors as well,https://github.com/pytorch/pytorch/issues/130428,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2024-07-16T16:47:03Z +issue,130480,[Feature Request] Lazy upcasting for mmap'd state dicts,https://github.com/pytorch/pytorch/issues/130480,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization",route to design/research decision before patch planning,2024-07-11T20:07:11Z +issue,127996,Tensor subclass constructor via autograd function is error prone re mutation,https://github.com/pytorch/pytorch/issues/127996,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2024-06-28T16:12:42Z +issue,120107,[feature request] keep gradcheck numerical jacobian from being depeciated,https://github.com/pytorch/pytorch/issues/120107,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2024-06-28T13:39:47Z +issue,128793,"GradType: a subset of dtype that is differentiable, containing all float and complex dtypes",https://github.com/pytorch/pytorch/issues/128793,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",route to design/research decision before patch planning,2024-06-27T20:11:38Z +issue,128934,version inquiry,https://github.com/pytorch/pytorch/issues/128934,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,route to design/research decision before patch planning,2024-06-27T09:30:36Z +issue,50122,"[proposal] Pseudo-functions to support common gradient patch use-cases like replacing inf / nan, gradient clipping / reversal",https://github.com/pytorch/pytorch/issues/50122,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2024-06-25T19:19:52Z +issue,124902,Add support for IPC features for PrivateUse devices,https://github.com/pytorch/pytorch/issues/124902,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: privateuse1",route to design/research decision before patch planning,2024-06-06T11:32:19Z +issue,127932,Add SinusoidalPositionalEmbedding module for use in Transformers and Diffusion models,https://github.com/pytorch/pytorch/issues/127932,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-06-05T02:20:50Z +issue,127699,torch.Tensor for optimizer parameters,https://github.com/pytorch/pytorch/issues/127699,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2024-06-03T19:52:37Z +issue,127284,Fused AdamW maybe should accept lr_dict directly?,https://github.com/pytorch/pytorch/issues/127284,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2024-05-28T14:58:02Z +issue,126762,Feature request to add the SReLU activation function,https://github.com/pytorch/pytorch/issues/126762,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-05-22T14:23:28Z +issue,125767,RFC: Integrate Arm Compute Library (ACL) into PyTorch as a submodule and add build support in setup.py,https://github.com/pytorch/pytorch/issues/125767,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",route to design/research decision before patch planning,2024-05-10T23:44:51Z +issue,121791,[RFC] Set Float32 Precision for CONV/RNN,https://github.com/pytorch/pytorch/issues/121791,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2024-05-08T06:45:02Z +issue,125223,torch.uniform_() is single-threaded on CPU,https://github.com/pytorch/pytorch/issues/125223,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: distributions, module: nn, module: performance",route to design/research decision before patch planning,2024-05-06T01:52:56Z +issue,58122,Generate special clear error messages for known common misuses in TorchScript,https://github.com/pytorch/pytorch/issues/58122,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2024-04-21T17:55:38Z +issue,49227,Looking for a more convenient way to getter value of the upper/lower triangular matrix into 1D,https://github.com/pytorch/pytorch/issues/49227,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: numpy",route to design/research decision before patch planning,2024-04-15T14:06:34Z +issue,63265,torch.scatter_: support index.size(d) > src.size(d),https://github.com/pytorch/pytorch/issues/63265,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2024-04-03T02:24:43Z +issue,47055,"[types] torch.utils.data.{Dataset, Sampler} are not Sized",https://github.com/pytorch/pytorch/issues/47055,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: data,route to design/research decision before patch planning,2024-04-02T15:54:07Z +issue,122551,Global Response Normalization (GRN),https://github.com/pytorch/pytorch/issues/122551,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-03-25T15:23:00Z +issue,121933,Please provide `total_repeat_length` field to PyTorch `repeat` / `repeat_interleave` API,https://github.com/pytorch/pytorch/issues/121933,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,route to design/research decision before patch planning,2024-03-20T16:46:51Z +issue,122099,No reduction of disk usage on slices of torch tensor,https://github.com/pytorch/pytorch/issues/122099,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,route to design/research decision before patch planning,2024-03-19T17:17:42Z +issue,119475,`autograd.Function` leads to `pylint` warnings in valid user code,https://github.com/pytorch/pytorch/issues/119475,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: typing",route to design/research decision before patch planning,2024-03-19T15:05:31Z +issue,109052,Allow reductions to write into pinned memory,https://github.com/pytorch/pytorch/issues/109052,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: reductions",route to design/research decision before patch planning,2024-03-14T02:24:02Z +issue,58228,"""_amp_foreach_non_finite_check_and_unscale_cuda"" not implemented for 'ComplexFloat'",https://github.com/pytorch/pytorch/issues/58228,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: complex, module: cuda",route to design/research decision before patch planning,2024-03-10T09:07:08Z +issue,120700,Invert the model architecture from the unknown neural network model (the python object of the model),https://github.com/pytorch/pytorch/issues/120700,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-03-01T08:14:44Z +issue,120072,Please consider adding more optimization support for multi-layer recurrent neural networks to improve fitting speed.,https://github.com/pytorch/pytorch/issues/120072,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",route to design/research decision before patch planning,2024-02-23T21:57:56Z +issue,100628,Feature request: fast way to approximate the diagonal of the hessian,https://github.com/pytorch/pytorch/issues/100628,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",route to design/research decision before patch planning,2024-02-21T11:13:19Z +issue,114462,Transformer with convolutional network,https://github.com/pytorch/pytorch/issues/114462,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-02-13T23:54:42Z +issue,102447,"Add additional ""sigmoid"" approximation to GeLu activation?",https://github.com/pytorch/pytorch/issues/102447,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-02-08T13:46:03Z +issue,110761,Using ChainedScheduler with ReduceLROnPlateau leads to unexpected keyword argument error,https://github.com/pytorch/pytorch/issues/110761,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",route to design/research decision before patch planning,2024-02-07T11:10:11Z +issue,20822,"[Proposal] Data reading framework for PyTorch (Hive, MySQL, S3 etc.)",https://github.com/pytorch/pytorch/issues/20822,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,route to design/research decision before patch planning,2024-02-05T23:35:05Z +issue,68879,Stop gradient option for padding,https://github.com/pytorch/pytorch/issues/68879,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",route to design/research decision before patch planning,2024-02-02T15:50:13Z +issue,118133,Better test disabling system,https://github.com/pytorch/pytorch/issues/118133,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: devx,route to design/research decision before patch planning,2024-02-01T14:05:00Z +issue,99278,Build error on libstc++ header stl_alogbase.h on riscv,https://github.com/pytorch/pytorch/issues/99278,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-01-21T11:52:05Z +issue,117217,"BatchNorm1d, running_mean and running_var",https://github.com/pytorch/pytorch/issues/117217,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-01-11T22:18:05Z +issue,72831,Toggling deterministic mode for individual autograd backward functions,https://github.com/pytorch/pytorch/issues/72831,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: determinism",route to design/research decision before patch planning,2024-01-10T18:32:09Z +issue,110295,Tests modify global state cause later tests to fail,https://github.com/pytorch/pytorch/issues/110295,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: devx",route to design/research decision before patch planning,2024-01-09T21:53:06Z +issue,115993,[FR] Add a way to reserve memory that survives `torch.cuda.empty_cache()`,https://github.com/pytorch/pytorch/issues/115993,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",route to design/research decision before patch planning,2024-01-08T23:34:36Z +issue,116555,The performance of `torch.index_select` and regular indexing differs dramatically based on the size of the tensor it is indexing,https://github.com/pytorch/pytorch/issues/116555,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: autograd, module: cpu, module: performance",route to design/research decision before patch planning,2024-01-08T18:58:39Z +issue,116354,Abstract torch.device for GPU/NE/TPU computations in the cloud-based agent,https://github.com/pytorch/pytorch/issues/116354,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2024-01-08T18:37:47Z +issue,115852,`groups` parameter for `nn.Linear`,https://github.com/pytorch/pytorch/issues/115852,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2024-01-08T13:32:33Z +issue,116490,Legitimize/harden differentiable optimizers in torch.optim,https://github.com/pytorch/pytorch/issues/116490,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: optimizer",route to design/research decision before patch planning,2023-12-28T07:38:44Z +issue,105255,[discussion] Integrate widely used utilities from fvcore into the core repo,https://github.com/pytorch/pytorch/issues/105255,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: nn, oncall: distributed",route to design/research decision before patch planning,2023-12-22T00:11:54Z +issue,115273,Consider improving project management (12 pages of HighPriority tickets > 1 year old),https://github.com/pytorch/pytorch/issues/115273,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: devx,route to design/research decision before patch planning,2023-12-18T18:08:24Z +issue,115588,`torch.nn.KLDivLoss` fails `test_grad` and `test_gradgrad`,https://github.com/pytorch/pytorch/issues/115588,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-12-12T21:59:19Z +issue,115585,`torch.nn.CTCLoss` doesn't seem to properly support noncontiguous inputs for 1d `target` with `target_dtype=torch.long`,https://github.com/pytorch/pytorch/issues/115585,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-12-11T20:52:25Z +issue,108491,Suppport Fused AdamW on CPU,https://github.com/pytorch/pytorch/issues/108491,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: performance",route to design/research decision before patch planning,2023-12-01T21:02:56Z +issue,114650,Error when calculating the Jacobian of torch.conj using forward-mode differentiation,https://github.com/pytorch/pytorch/issues/114650,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex",route to design/research decision before patch planning,2023-11-29T01:15:40Z +issue,40316,Refactor the adaptive avg pool code,https://github.com/pytorch/pytorch/issues/40316,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: pooling, oncall: quantization",scan when time allows,2023-11-17T22:22:33Z +issue,113730,Can the official provide cpu offload method for large model inference?,https://github.com/pytorch/pytorch/issues/113730,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2023-11-17T17:33:19Z +issue,110885,Depthwise conv3d slower than normal conv3d,https://github.com/pytorch/pytorch/issues/110885,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",route to design/research decision before patch planning,2023-11-03T18:53:06Z +issue,102201,scipy.ndimage.find_objects,https://github.com/pytorch/pytorch/issues/102201,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-11-01T20:01:51Z +issue,16663,new_zeros is not traced correctly,https://github.com/pytorch/pytorch/issues/16663,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-11-01T17:43:43Z +issue,111563,"Higher-order derivatives extremely slow, increasing exponentially",https://github.com/pytorch/pytorch/issues/111563,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2023-11-01T05:36:13Z +issue,112352,RFC: [pytree] node registration namespaces,https://github.com/pytorch/pytorch/issues/112352,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: pytree,route to design/research decision before patch planning,2023-10-30T18:31:46Z +issue,110780,backward and grad behave inconsistently w.r.t. set_ on leaf variable,https://github.com/pytorch/pytorch/issues/110780,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2023-10-09T17:01:06Z +issue,92330,Simplify module backward hooks to use multi-grad hooks instead,https://github.com/pytorch/pytorch/issues/92330,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: bc-breaking, module: nn",route to design/research decision before patch planning,2023-10-05T22:31:45Z +issue,108090,[Optimizer Perf] Improve speed of _init_group to c++,https://github.com/pytorch/pytorch/issues/108090,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: performance",route to design/research decision before patch planning,2023-10-05T15:10:28Z +issue,87734,Checkpointing Support for Modularized Optimizers,https://github.com/pytorch/pytorch/issues/87734,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2023-10-04T21:58:58Z +issue,97469,Improve collectives fingerprinting,https://github.com/pytorch/pytorch/issues/97469,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: c10d,scan when time allows,2023-09-26T10:12:30Z +issue,109948,Simple script segfaulting when grad is enabled,https://github.com/pytorch/pytorch/issues/109948,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",route to design/research decision before patch planning,2023-09-25T16:16:11Z +issue,71479,[numpy compat] torch.stack and torch.tensor doesn't support nested list+tensors (NumPy does support) - at least document the difference in the error message,https://github.com/pytorch/pytorch/issues/71479,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: viewing and reshaping",route to design/research decision before patch planning,2023-09-23T18:53:56Z +issue,103449,Asynchronous CUDA AveragedModel,https://github.com/pytorch/pytorch/issues/103449,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2023-09-22T18:18:52Z +issue,109706,Make standard container classes satisfy container Protocols.,https://github.com/pytorch/pytorch/issues/109706,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-09-22T18:04:05Z +issue,109582,Make Dropout take a dim=... argument,https://github.com/pytorch/pytorch/issues/109582,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-09-22T16:27:39Z +issue,108692,Adding Maximal Update Parametrization (µP) to torch.nn.init,https://github.com/pytorch/pytorch/issues/108692,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-09-14T14:56:07Z +issue,108784,Collect more env variables in `collect_env.py`,https://github.com/pytorch/pytorch/issues/108784,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2023-09-12T11:14:52Z +issue,66813,Use `__slots__` for the `nn.Module` class,https://github.com/pytorch/pytorch/issues/66813,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2023-08-28T14:13:25Z +issue,67087,Feature Request: 1d grid sample,https://github.com/pytorch/pytorch/issues/67087,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-08-26T04:19:28Z +issue,62902,distributed/elastic/multiprocessing/api_test is flaky on ROCm,https://github.com/pytorch/pytorch/issues/62902,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: rocm, oncall: r2p",route to design/research decision before patch planning,2023-08-22T21:32:48Z +issue,106784,"[ux] Suppot torch.tensor(set([1,2,3]))",https://github.com/pytorch/pytorch/issues/106784,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2023-08-21T18:27:56Z +issue,105365,Quadric Layer,https://github.com/pytorch/pytorch/issues/105365,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-08-09T20:19:27Z +issue,96319,views created in __torch_dispatch__ share storage but not version_counter,https://github.com/pytorch/pytorch/issues/96319,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: autograd, module: molly-guard",route to design/research decision before patch planning,2023-07-24T15:20:16Z +issue,39765,Support mainstream pruning techniques,https://github.com/pytorch/pytorch/issues/39765,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: pruning,route to design/research decision before patch planning,2023-07-14T12:23:52Z +issue,102337,Tensors that share same underlying storage to also share gradient storage,https://github.com/pytorch/pytorch/issues/102337,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: python frontend",route to design/research decision before patch planning,2023-07-13T14:37:49Z +issue,103588,LSTM/RNN operation agnostic,https://github.com/pytorch/pytorch/issues/103588,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-07-07T16:15:11Z +issue,102417,_view_func but without keeping original view tensor alive,https://github.com/pytorch/pytorch/issues/102417,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: viewing and reshaping",route to design/research decision before patch planning,2023-06-23T15:24:01Z +issue,96764,Improve checkpoint thread-safety,https://github.com/pytorch/pytorch/issues/96764,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: multithreading",route to design/research decision before patch planning,2023-06-23T15:22:12Z +issue,96579,Wrong return type from operation on custom tensor inside registered hook,https://github.com/pytorch/pytorch/issues/96579,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: autograd",route to design/research decision before patch planning,2023-06-23T15:16:09Z +issue,103109,Increased / more verbose type aliases for improved readability of user defined content,https://github.com/pytorch/pytorch/issues/103109,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: typing,route to design/research decision before patch planning,2023-06-19T20:22:13Z +issue,102905,nn.ChannelShuffle1d,https://github.com/pytorch/pytorch/issues/102905,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-06-09T19:20:49Z +issue,102730,Best practices clarification for initialization strategies,https://github.com/pytorch/pytorch/issues/102730,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: initialization, module: nn, module: python frontend",route to design/research decision before patch planning,2023-06-09T19:12:08Z +issue,10756,Unexpected Behavior when Pointwise Operations Write to Expanded Tensors,https://github.com/pytorch/pytorch/issues/10756,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: internals, module: partial aliasing",scan when time allows,2023-06-07T22:36:05Z +issue,102938,Support for efficiently processing categorical distributions with varying dimensions,https://github.com/pytorch/pytorch/issues/102938,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: distributions,route to design/research decision before patch planning,2023-06-05T15:30:35Z +issue,95374,Add oscillating activation functions to PyTorch.,https://github.com/pytorch/pytorch/issues/95374,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: loss,route to design/research decision before patch planning,2023-05-31T00:04:45Z +issue,44954,Optimization with constraints for torch.optim,https://github.com/pytorch/pytorch/issues/44954,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2023-05-13T12:07:01Z +issue,100990,Extending compatibility of LibTorch,https://github.com/pytorch/pytorch/issues/100990,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,route to design/research decision before patch planning,2023-05-10T01:51:22Z +issue,48830,"Please add ""dim"" feature for function ""torch.masked_select""",https://github.com/pytorch/pytorch/issues/48830,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: sorting and selection",route to design/research decision before patch planning,2023-04-21T16:58:13Z +issue,83800,torch.var_mean is slower than layer norm,https://github.com/pytorch/pytorch/issues/83800,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",route to design/research decision before patch planning,2023-04-14T18:17:41Z +issue,97788,Add SSIM as Loss Function,https://github.com/pytorch/pytorch/issues/97788,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: loss,route to design/research decision before patch planning,2023-03-31T21:10:15Z +issue,98089,GroupNorm cpu/gpu parity tests fail with pretty large differences,https://github.com/pytorch/pytorch/issues/98089,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-03-31T21:04:04Z +issue,97248,"A more systematic API for resolving the ""vmap-incompatible in-place operation"" error",https://github.com/pytorch/pytorch/issues/97248,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,route to design/research decision before patch planning,2023-03-21T22:35:35Z +issue,96555,early stopping,https://github.com/pytorch/pytorch/issues/96555,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2023-03-14T17:53:20Z +issue,95916,autograd.functional.jacobian : tensor instead of function as input for reverse mode?,https://github.com/pytorch/pytorch/issues/95916,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2023-03-10T22:50:34Z +issue,95337,Changing behavior of module.to() to better support mixed real- and complex-valued parameters,https://github.com/pytorch/pytorch/issues/95337,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nn",route to design/research decision before patch planning,2023-03-03T20:11:58Z +issue,88325,Group losses in a common namespace,https://github.com/pytorch/pytorch/issues/88325,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",route to design/research decision before patch planning,2023-03-03T20:09:25Z +issue,93684,Can we rewrite numpy operators to pytorch operators?,https://github.com/pytorch/pytorch/issues/93684,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-02-21T15:31:52Z +issue,94904,"Optimizer ""Lion"" in Symbolic Discovery of Optimization Algorithms",https://github.com/pytorch/pytorch/issues/94904,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2023-02-17T15:33:50Z +issue,84751,Allow passing dict (as opposed to OrderedDict) to nn.Sequential,https://github.com/pytorch/pytorch/issues/84751,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-02-10T20:31:34Z +issue,62148,[feature request] Function to broadcast right,https://github.com/pytorch/pytorch/issues/62148,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,route to design/research decision before patch planning,2023-02-06T23:24:59Z +issue,93154,Support for VeLO optimizer.,https://github.com/pytorch/pytorch/issues/93154,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2023-02-03T20:06:26Z +issue,84524,nn.Softmax should not allow default/implicit/unset dim constructor argument,https://github.com/pytorch/pytorch/issues/84524,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: nn",route to design/research decision before patch planning,2023-01-27T21:03:08Z +issue,89724,Sample Weighted BatchNorm1d,https://github.com/pytorch/pytorch/issues/89724,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",route to design/research decision before patch planning,2023-01-27T20:36:35Z +issue,87366,"Implementation of CG, and BICGSTAB methods",https://github.com/pytorch/pytorch/issues/87366,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2023-01-27T20:36:23Z +issue,86590,"Easy way to ""freeze"" BatchNorm running_mean/running_var",https://github.com/pytorch/pytorch/issues/86590,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2023-01-27T20:36:17Z +issue,59787,Remove support for multiple ellipses in slicing,https://github.com/pytorch/pytorch/issues/59787,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: deprecation, module: python array api",route to design/research decision before patch planning,2023-01-11T08:04:38Z +issue,91073,Implement L1 and L2 gradient as hooks with the option of changing the weight decay value.,https://github.com/pytorch/pytorch/issues/91073,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer",route to design/research decision before patch planning,2022-12-20T08:40:00Z +issue,84593,Uneven and/or Dynamically sized collectives,https://github.com/pytorch/pytorch/issues/84593,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: c10d,scan when time allows,2022-12-14T23:50:15Z +issue,80553,Nonliner conjugate gradient optimizer + Hager-Zhang line search,https://github.com/pytorch/pytorch/issues/80553,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2022-11-25T02:54:51Z +issue,86641,RFC(from users): nn.Module behavior with in-place changes,https://github.com/pytorch/pytorch/issues/86641,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,route to design/research decision before patch planning,2022-10-11T13:51:18Z +issue,64067,Execute C++ based GTest unit tests from a python wrapper,https://github.com/pytorch/pytorch/issues/64067,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-09-28T18:58:34Z +issue,67321,alias to generate tensor with random uniform distribution.,https://github.com/pytorch/pytorch/issues/67321,good volunteer slice,20,label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2022-09-09T15:29:27Z +issue,83492,`Frozen` module for transfer learning.,https://github.com/pytorch/pytorch/issues/83492,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2022-08-26T19:13:33Z +issue,83854,Pytorch/Nova CI should monitor service outages for major dependencies,https://github.com/pytorch/pytorch/issues/83854,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: ci,route to design/research decision before patch planning,2022-08-23T13:32:52Z +issue,76532,Add ability to add custom suffixes to tensor repr,https://github.com/pytorch/pytorch/issues/76532,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: printing, module: python frontend",route to design/research decision before patch planning,2022-08-23T06:52:09Z +issue,61343,Feature Request: Remove Optimizer Lazy State Initialization,https://github.com/pytorch/pytorch/issues/61343,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2022-08-18T14:41:58Z +issue,67587,Re-implement `Optimizer.__repr__`,https://github.com/pytorch/pytorch/issues/67587,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: python frontend",route to design/research decision before patch planning,2022-08-18T14:34:50Z +issue,79205,out-of-place functional optimizers: functional optimizers may not be composite compliant,https://github.com/pytorch/pytorch/issues/79205,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: optimizer",route to design/research decision before patch planning,2022-08-18T14:16:19Z +issue,80308,CosineAnnealingWarmRestarts with initial warm up and weight decay applied on consecutive cycles without warm up,https://github.com/pytorch/pytorch/issues/80308,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2022-08-18T14:10:40Z +issue,83159,Expand Learning rate scheduling to any optimization hyperparameter,https://github.com/pytorch/pytorch/issues/83159,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",route to design/research decision before patch planning,2022-08-18T14:05:34Z +issue,79452,Use c10d broadcast_object in Zero,https://github.com/pytorch/pytorch/issues/79452,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",scan when time allows,2022-07-28T22:35:47Z +issue,33181,[feature request] [dataloader] Introduce Dataset.__collate__,https://github.com/pytorch/pytorch/issues/33181,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,route to design/research decision before patch planning,2022-07-26T18:24:03Z +issue,68699,Transpose of a sparse tensor is not a view operation,https://github.com/pytorch/pytorch/issues/68699,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse, module: viewing and reshaping",route to design/research decision before patch planning,2022-07-26T15:30:27Z +issue,81333,Reductions on tensors larger than GPU memory,https://github.com/pytorch/pytorch/issues/81333,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,route to design/research decision before patch planning,2022-07-12T16:57:38Z +issue,80380,Support for learnable p Values in LPPOOL like Pool,https://github.com/pytorch/pytorch/issues/80380,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2022-06-28T15:22:40Z +issue,79877,test_meta_vstack_cuda_int16 (__main__.TestMetaCUDA) Fails with DEBUG=1,https://github.com/pytorch/pytorch/issues/79877,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,route to design/research decision before patch planning,2022-06-25T03:25:08Z +issue,79842,Automatically use CUDA,https://github.com/pytorch/pytorch/issues/79842,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,route to design/research decision before patch planning,2022-06-21T20:32:54Z +issue,68516,"[ux] [feature request] Arguments to modules(), named_modules(), children(), named_children() to filter modules of specific types",https://github.com/pytorch/pytorch/issues/68516,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2022-06-02T18:12:51Z +issue,14844,Negative indexing for nn.Embedding inputs,https://github.com/pytorch/pytorch/issues/14844,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2022-05-24T21:51:52Z +issue,72712,Better support for pypip packages implementing torch cuda extentions,https://github.com/pytorch/pytorch/issues/72712,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,route to design/research decision before patch planning,2022-03-31T11:28:20Z +issue,74904,Adding novel 'AdaFamily' optimizer,https://github.com/pytorch/pytorch/issues/74904,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2022-03-30T09:14:02Z +issue,73143,Add nan-safe einsum and bilinear,https://github.com/pytorch/pytorch/issues/73143,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,route to design/research decision before patch planning,2022-03-14T07:20:41Z +issue,73172,LazyLinear with equal in_features and out_features,https://github.com/pytorch/pytorch/issues/73172,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2022-03-01T18:18:43Z +issue,73513,Wrapper tensor level confusion,https://github.com/pytorch/pytorch/issues/73513,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,route to design/research decision before patch planning,2022-03-01T14:44:13Z +issue,67634,Feature request: Grid sample on complex tensors with float grid,https://github.com/pytorch/pytorch/issues/67634,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: complex, module: interpolation, module: nn",route to design/research decision before patch planning,2022-02-06T21:40:47Z +issue,68798,Clarify variables of BatchNorm*d functions,https://github.com/pytorch/pytorch/issues/68798,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: norms and normalization",route to design/research decision before patch planning,2022-01-23T18:03:00Z +issue,69757,[Feature request] nn.Model API: Standard model interface,https://github.com/pytorch/pytorch/issues/69757,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2021-12-17T18:15:39Z +issue,44827,[RFC] Pipeline Parallelism in PyTorch,https://github.com/pytorch/pytorch/issues/44827,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",route to design/research decision before patch planning,2021-12-09T17:30:40Z +issue,68648,[feature request] Sticky/force regime of train/eval modes,https://github.com/pytorch/pytorch/issues/68648,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: ux",route to design/research decision before patch planning,2021-11-27T11:41:20Z +issue,68514,Add common regularizations to PyTorch.,https://github.com/pytorch/pytorch/issues/68514,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",route to design/research decision before patch planning,2021-11-20T16:16:30Z +issue,65913,Support differentiability through clone and update,https://github.com/pytorch/pytorch/issues/65913,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",route to design/research decision before patch planning,2021-11-05T17:35:03Z +issue,64730,Enable BatchNorm to use running mean/variance during train,https://github.com/pytorch/pytorch/issues/64730,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",route to design/research decision before patch planning,2021-10-07T13:13:35Z +issue,57724,Incorporate Posit Support,https://github.com/pytorch/pytorch/issues/57724,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2021-09-17T13:28:03Z +issue,46168,[discussion] In-place gradient (grad_input) computation for better memory utilisation,https://github.com/pytorch/pytorch/issues/46168,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",route to design/research decision before patch planning,2021-09-07T03:34:19Z +issue,49890,GEMM with int8 datatype throws RuntimeError on GPU,https://github.com/pytorch/pytorch/issues/49890,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,route to design/research decision before patch planning,2021-09-01T05:27:12Z +issue,46544,PyTorch NaN behavior and API design,https://github.com/pytorch/pytorch/issues/46544,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: numpy, module: ux",route to design/research decision before patch planning,2021-08-30T21:08:20Z +issue,60223,[testing] SkipInfo should error if `cls_name` and `test_name` combination is not valid,https://github.com/pytorch/pytorch/issues/60223,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: tests,route to design/research decision before patch planning,2021-08-30T02:53:38Z +issue,20704,Remove unpack() in torch/csrc/nn/type_checks.h and its caller functions in the codebase,https://github.com/pytorch/pytorch/issues/20704,good volunteer slice,20,"label says issue is suitable for a scoped external contribution; old update age or low-urgency/stale label; component/oncall labels: module: internals, module: nn",scan when time allows,2021-08-25T16:58:36Z +issue,63038,Use pytorch-probot for PyTorch-specific stuff + MOTD in Dr. CI comment,https://github.com/pytorch/pytorch/issues/63038,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: infra,route to design/research decision before patch planning,2021-08-16T21:38:33Z +issue,61490,Deprecate torch.(min|max|median|mode) to only return values and not indices,https://github.com/pytorch/pytorch/issues/61490,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy, module: python array api, module: reductions",route to design/research decision before patch planning,2021-08-12T16:14:00Z +issue,62155,[docs] Change code snippet formatting to be directly copy-pasteable into REPL or into source files,https://github.com/pytorch/pytorch/issues/62155,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: docs,route to design/research decision before patch planning,2021-07-27T09:08:08Z +issue,61901,Reducing over empty dimensions for reductions without identity,https://github.com/pytorch/pytorch/issues/61901,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: reductions",route to design/research decision before patch planning,2021-07-22T13:16:16Z +issue,61664,Support adding new keyword-only parameters without breaking FC,https://github.com/pytorch/pytorch/issues/61664,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: internals",route to design/research decision before patch planning,2021-07-19T17:19:53Z +issue,61585,"Proposal: allow using axis, axes, dim and dims interchangeably",https://github.com/pytorch/pytorch/issues/61585,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: ux",route to design/research decision before patch planning,2021-07-14T23:50:47Z +issue,59868,Python scalars should be promoted to the same `dtype` as the respective tensor,https://github.com/pytorch/pytorch/issues/59868,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: type promotion",route to design/research decision before patch planning,2021-06-16T21:05:19Z +issue,51962,PyTorch for scientific computing,https://github.com/pytorch/pytorch/issues/51962,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2021-05-05T18:59:06Z +issue,41803,out-variant for tensor.bitwise_and (exists for torch.bitwise_and) + bitwise_friends,https://github.com/pytorch/pytorch/issues/41803,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: ux",route to design/research decision before patch planning,2021-03-24T15:26:45Z +issue,35801,Add KFAC optimizer,https://github.com/pytorch/pytorch/issues/35801,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2021-03-09T10:55:54Z +issue,51156,[Feature Request] Support tensor creation from objects that implement the __array__ interface,https://github.com/pytorch/pytorch/issues/51156,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy,route to design/research decision before patch planning,2021-03-06T17:44:18Z +issue,52207,Wasserstein metric,https://github.com/pytorch/pytorch/issues/52207,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2021-02-26T17:46:36Z +issue,52036,Automatic splitting of batch during model feed-forwarding to avoid memory errors caused by the too big batch_size,https://github.com/pytorch/pytorch/issues/52036,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage,route to design/research decision before patch planning,2021-02-11T09:27:35Z +issue,43450,Training Large Neural Networks with Constant Memory using a New Execution Algorithm,https://github.com/pytorch/pytorch/issues/43450,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: nn",route to design/research decision before patch planning,2021-02-08T05:17:12Z +issue,50204,Find the joint eigenvalue of two matrices.,https://github.com/pytorch/pytorch/issues/50204,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,route to design/research decision before patch planning,2021-02-03T11:48:09Z +issue,46829,Mixing Numpy's arrays and PyTorch tensors,https://github.com/pytorch/pytorch/issues/46829,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: ux",route to design/research decision before patch planning,2021-01-27T14:23:19Z +issue,50712,ignore_index for nn_mse_loss,https://github.com/pytorch/pytorch/issues/50712,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",route to design/research decision before patch planning,2021-01-20T14:07:49Z +issue,45051,Custom Datatypes in Tensors,https://github.com/pytorch/pytorch/issues/45051,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy,route to design/research decision before patch planning,2021-01-12T16:43:34Z +issue,8837,Inconsistency in implementation of _LRScheduler,https://github.com/pytorch/pytorch/issues/8837,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer",route to design/research decision before patch planning,2021-01-05T17:15:17Z +issue,49338,a time series library,https://github.com/pytorch/pytorch/issues/49338,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2020-12-21T21:45:00Z +issue,46516,Batchnorm support for tracking buffer statistics when using gradient accumulation,https://github.com/pytorch/pytorch/issues/46516,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,route to design/research decision before patch planning,2020-12-19T02:10:02Z +issue,48747,[bug][libtorch] torch::load seeks to the start of the input stream before reading,https://github.com/pytorch/pytorch/issues/48747,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,route to design/research decision before patch planning,2020-12-03T01:31:46Z +issue,48494,Inconsistent type of property stride of Conv1d and MaxPool1d,https://github.com/pytorch/pytorch/issues/48494,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: ux",route to design/research decision before patch planning,2020-11-27T11:43:47Z +issue,47915,sparse filter layers (more specifically convolutions),https://github.com/pytorch/pytorch/issues/47915,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: sparse,route to design/research decision before patch planning,2020-11-16T13:28:19Z +issue,47037,"Change ""_next_index()"" in DataLoader to a public and stable API",https://github.com/pytorch/pytorch/issues/47037,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,route to design/research decision before patch planning,2020-10-29T14:39:41Z +issue,6657,[feature request] Stochastic Variance Reduced Gradient (SVRG) optimizer,https://github.com/pytorch/pytorch/issues/6657,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2020-10-20T14:40:01Z +issue,44591,[Feature] Fused Matmul & Min/Max/Sum/Prod,https://github.com/pytorch/pytorch/issues/44591,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",route to design/research decision before patch planning,2020-10-06T09:49:47Z +issue,45808,Support for Locality Sensitive Hashing Optimizations,https://github.com/pytorch/pytorch/issues/45808,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2020-10-05T20:52:47Z +issue,45133,[Feature Request] Add support for Hidden Markov Models in torch.distributions,https://github.com/pytorch/pytorch/issues/45133,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: distributions,route to design/research decision before patch planning,2020-10-01T00:01:23Z +issue,45295,Adaptive usage of memory during training inference,https://github.com/pytorch/pytorch/issues/45295,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",route to design/research decision before patch planning,2020-09-28T15:37:30Z +issue,44777,Hybrid Memory,https://github.com/pytorch/pytorch/issues/44777,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",route to design/research decision before patch planning,2020-09-18T09:00:43Z +issue,42652,Add non_blocking=True copy of torch.Storage,https://github.com/pytorch/pytorch/issues/42652,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2020-08-10T22:12:26Z +issue,4660,[Feature proposal] Add MC-derived optimizers,https://github.com/pytorch/pytorch/issues/4660,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2020-08-04T18:07:49Z +issue,41559,momentum in BatchNorm,https://github.com/pytorch/pytorch/issues/41559,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",route to design/research decision before patch planning,2020-07-21T19:24:25Z +issue,41556,Implement LSH Optimizations for Enhanced CPU-Only Performance,https://github.com/pytorch/pytorch/issues/41556,needs design decision,20,"needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",route to design/research decision before patch planning,2020-07-17T17:20:11Z +issue,39780,LRP-based explainability,https://github.com/pytorch/pytorch/issues/39780,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label,route to design/research decision before patch planning,2020-06-10T17:47:44Z +issue,3904,Implement DE in pytorch.optim,https://github.com/pytorch/pytorch/issues/3904,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,route to design/research decision before patch planning,2020-05-29T14:18:10Z +issue,37155,Per-cluster biases in AdaptiveLogSoftmaxWithLoss,https://github.com/pytorch/pytorch/issues/37155,needs design decision,20,needs design/research label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,route to design/research decision before patch planning,2020-04-28T19:05:20Z +pr,184042,Tma scatter add remove software pipelining,https://github.com/pytorch/pytorch/pull/184042,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-22T22:08:36Z +pr,184753,Fix Dynamo frozenset copy identity semantics,https://github.com/pytorch/pytorch/pull/184753,stale/low urgency,17,"PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-05-21T21:46:27Z +pr,184605,Add CPython Dynamo agentic loop plan,https://github.com/pytorch/pytorch/pull/184605,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-21T19:51:43Z +pr,182269,[MPS] Fix channels_last batch_norm backward,https://github.com/pytorch/pytorch/pull/182269,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/mps,review when ready,2026-05-21T06:25:07Z +pr,183386,Add tests for OneCycleLR zero divisor initialization behavior,https://github.com/pytorch/pytorch/pull/183386,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-19T22:50:52Z +pr,184116,[distributed] Add NVFP4 scaled matmul + reduce-scatter fast path #182915,https://github.com/pytorch/pytorch/pull/184116,stale/low urgency,17,"PR has not updated for at least 45 days; CI lane labels: ciflow/h100-symm-mem, ciflow/torchtitan",review when ready,2026-05-19T17:20:45Z +pr,184067,[submodule] Drop unused third_party/gemmlowp submodule,https://github.com/pytorch/pytorch/pull/184067,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,review when ready,2026-05-19T15:21:11Z +pr,183976,[CUDA] Correct inner scope Math.cuh int i -> int j,https://github.com/pytorch/pytorch/pull/183976,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-19T14:45:02Z +pr,183568,Fix timeout in `test_share_memory` on error,https://github.com/pytorch/pytorch/pull/183568,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/h100-symm-mem,review when ready,2026-05-18T21:32:41Z +pr,182424,Warning oversized storage copy,https://github.com/pytorch/pytorch/pull/182424,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-18T09:53:37Z +pr,183203,perf(dataloader): zero-copy contiguous-slice fast path for TensorDataset,https://github.com/pytorch/pytorch/pull/183203,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-16T01:06:58Z +pr,183726,[poc] yet another nonstrict api,https://github.com/pytorch/pytorch/pull/183726,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-15T21:54:44Z +pr,183787,[PyTorch] Support selective TorchScript tensor data elision (#183787),https://github.com/pytorch/pytorch/pull/183787,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T22:48:56Z +pr,177450,DataPipes: prevent failing properties from masking invalid-iterator errors,https://github.com/pytorch/pytorch/pull/177450,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T21:34:59Z +pr,175320,[MPS] Ensure pool2d inputs are contiguous in target memory format,https://github.com/pytorch/pytorch/pull/175320,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/mps,review when ready,2026-05-14T19:40:35Z +pr,182114,Add torch.utils.nan_guard for forward-pass NaN/Inf debugging,https://github.com/pytorch/pytorch/pull/182114,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T17:51:21Z +pr,182583,Skip RST<->MD source conversions in doc-redirects-check,https://github.com/pytorch/pytorch/pull/182583,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T16:37:32Z +pr,181262,"Show node.meta[""custom""][""stream""] in Graph.__str__",https://github.com/pytorch/pytorch/pull/181262,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,review when ready,2026-05-14T16:23:10Z +pr,183405,Fix unused groups in test_convs,https://github.com/pytorch/pytorch/pull/183405,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T16:06:31Z +pr,183502,Remove outdated code for macOS < 12,https://github.com/pytorch/pytorch/pull/183502,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T16:06:15Z +pr,182788,Fix decorators of `ProcessGroupGlooTest` & `ProcessGroupNCCLTest`,https://github.com/pytorch/pytorch/pull/182788,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T16:04:27Z +pr,182836,[ROCm][CI] Update mi200 workflows to use arc label,https://github.com/pytorch/pytorch/pull/182836,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T16:03:52Z +pr,182746,[MPS] Memory-efficient SDPA backward (opt-in),https://github.com/pytorch/pytorch/pull/182746,stale/low urgency,17,"PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/mps",review when ready,2026-05-14T16:03:47Z +pr,182873,test: skip some test_non_standard_bool_values test cases,https://github.com/pytorch/pytorch/pull/182873,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-14T16:03:28Z +pr,183037,[Distributed] Add `ranks_only` to `StateDictOptions` for controlling which ranks receive the full state dict,https://github.com/pytorch/pytorch/pull/183037,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,review when ready,2026-05-14T16:02:20Z +pr,171686,Add failing test documenting implicit step during `LRScheduler` initialization,https://github.com/pytorch/pytorch/pull/171686,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-12T23:39:52Z +pr,182857,[inductor][mtia] Skip matvec bmm/addmm decomposition on MTIA; add MTIA matvec bmm lowering (#182857),https://github.com/pytorch/pytorch/pull/182857,stale/low urgency,17,"PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",review when ready,2026-05-12T15:37:03Z +pr,166625,Fix skip condition in test_overrides,https://github.com/pytorch/pytorch/pull/166625,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-12T10:35:17Z +pr,182934,[RFC][distributed] Add parallel_map with collective operation interception…,https://github.com/pytorch/pytorch/pull/182934,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,review when ready,2026-05-11T22:15:00Z +pr,182975,Document 5 public APIs in docs/source/utils.md,https://github.com/pytorch/pytorch/pull/182975,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T21:26:01Z +pr,176430,Don't use sharding for tests with a single shard,https://github.com/pytorch/pytorch/pull/176430,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T11:21:08Z +pr,174768,Normalize object names in TreeProfiler test,https://github.com/pytorch/pytorch/pull/174768,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T11:20:52Z +pr,165986,Skip tests requiring NCCL if no GPU is found,https://github.com/pytorch/pytorch/pull/165986,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T11:20:51Z +pr,174959,Make server creation in distributed/elastic/test_control_plane more reliable,https://github.com/pytorch/pytorch/pull/174959,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T10:01:53Z +pr,177126,Fix search for OpenMP using e.g. GCC,https://github.com/pytorch/pytorch/pull/177126,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T08:45:15Z +pr,169936,Fix `TestSaveLoad.test_version_error`,https://github.com/pytorch/pytorch/pull/169936,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T08:45:03Z +pr,176709,Remove Eigen d_cnjg and r_cnjg source reference,https://github.com/pytorch/pytorch/pull/176709,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-05-11T07:56:57Z +pr,141842,Refactor NJT to hold metadata on nested int,https://github.com/pytorch/pytorch/pull/141842,stale/low urgency,17,"PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/trunk",review when ready,2026-05-06T20:42:48Z +pr,119977,[NJT] Factory function support,https://github.com/pytorch/pytorch/pull/119977,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2026-05-04T22:34:38Z +pr,132414,Ensure FakeTensor post-creation has associated nested int cached,https://github.com/pytorch/pytorch/pull/132414,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2026-05-04T22:34:33Z +pr,168223,Add failing test for negative epoch crash in LR schedulers (ZeroDivisionError),https://github.com/pytorch/pytorch/pull/168223,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-04-04T12:42:00Z +pr,130752,support zb1p and zb2p algorithms,https://github.com/pytorch/pytorch/pull/130752,stale/low urgency,17,"PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan, ciflow/trunk",review when ready,2026-03-26T21:24:16Z +pr,168895,Add `pad_to_multiple_of` to `pad_sequence` (C++ only),https://github.com/pytorch/pytorch/pull/168895,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-02-03T06:32:28Z +pr,167541,[MPS] Fix crashes due to a race condition in the MPS backend,https://github.com/pytorch/pytorch/pull/167541,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2026-01-26T10:21:33Z +pr,162079,Tune linalg_eigh_cusolver: better heuristic for syevj_batched selection on cuda,https://github.com/pytorch/pytorch/pull/162079,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2025-11-15T20:20:24Z +pr,152158,Add AC_TRACER Infra TorchDispatchMode key,https://github.com/pytorch/pytorch/pull/152158,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2025-07-24T18:26:34Z +pr,102613,Support building pytorch using MKL ILP64 model.,https://github.com/pytorch/pytorch/pull/102613,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2025-06-19T08:43:08Z +pr,147927,Custom ops support arbitrary input types by migrating to python dispatcher,https://github.com/pytorch/pytorch/pull/147927,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2025-05-30T15:02:48Z +pr,136835,Add back DistributedDataParallel types that were lost when pyi was removed,https://github.com/pytorch/pytorch/pull/136835,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2025-05-06T09:21:08Z +pr,142114,[DO NOT MERGE][WIP] CI: Dispatch PR events to the out-of-tree test infra,https://github.com/pytorch/pytorch/pull/142114,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2025-04-16T07:00:50Z +pr,146101,(WIP) Update NJT ops to check data for raggedness check,https://github.com/pytorch/pytorch/pull/146101,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2025-04-15T16:01:07Z +pr,145922,Update NestedInt equality to take into account all metadata,https://github.com/pytorch/pytorch/pull/145922,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2025-04-08T21:20:16Z +pr,146318,Hack AC to not clear recomputed activations,https://github.com/pytorch/pytorch/pull/146318,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2025-04-07T21:46:21Z +pr,138519,[Inductor] introduce comm buffer planning,https://github.com/pytorch/pytorch/pull/138519,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2025-03-06T02:03:37Z +pr,140321,[scan] Refactored testcases,https://github.com/pytorch/pytorch/pull/140321,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2025-01-18T18:23:50Z +pr,119496,Switch batch norm stack to consolidated ops,https://github.com/pytorch/pytorch/pull/119496,stale/low urgency,17,"PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/mps",review when ready,2024-09-24T23:02:33Z +pr,130462,[testing] Allow creating offsets within AC region for NJT without SAC,https://github.com/pytorch/pytorch/pull/130462,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2024-09-23T19:55:20Z +pr,117905,Factory function and basic .sizes() support for C++ NestedTensor,https://github.com/pytorch/pytorch/pull/117905,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2024-08-26T17:00:50Z +pr,110155,[TESTING] Capture scalar/dynamic by default,https://github.com/pytorch/pytorch/pull/110155,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,review when ready,2024-04-07T06:26:26Z +pr,108303,[optim] Make casting to match params a hook (2nd try),https://github.com/pytorch/pytorch/pull/108303,stale/low urgency,17,PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,review when ready,2024-01-12T19:19:46Z +pr,82997,Implement refs.var as a real reference,https://github.com/pytorch/pytorch/pull/82997,stale/low urgency,17,PR has not updated for at least 45 days,review when ready,2023-04-17T00:35:59Z +pr,184358,Add XPU to externalId,https://github.com/pytorch/pytorch/pull/184358,draft/noise,15,PR is draft; PR closes or references open issue numbers; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-25T00:03:21Z +pr,183507,einsum: fast path for small-K batched-dot to avoid cuBLAS launch overhead,https://github.com/pytorch/pytorch/pull/183507,draft/noise,15,PR is draft; PR closes or references open issue numbers; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-18T02:07:30Z +pr,183009,[Inductor] Fall back native_layer_norm; route var_mean through inductor lowering,https://github.com/pytorch/pytorch/pull/183009,draft/noise,15,"PR is draft; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-11T03:11:51Z +pr,153893,[PP] Fix double backward error in stage_backward,https://github.com/pytorch/pytorch/pull/153893,draft/noise,15,PR is draft; PR closes or references open issue numbers; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,wait for author to mark ready,2026-03-20T22:40:30Z +pr,124624,[NJT] Allow construction of NJT within graph using offsets from inputs,https://github.com/pytorch/pytorch/pull/124624,draft/noise,15,PR is draft; PR closes or references open issue numbers; PR has not updated for at least 45 days,wait for author to mark ready,2024-08-12T20:07:16Z +issue,158986,`nvlink error: Entry function '' uses too much data for compiler-generated constants`,https://github.com/pytorch/pytorch/issues/158986,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-07T16:46:46Z +issue,156010,`torch.prod` or `torch.special.entr` triggers `CUDA driver error: invalid argument` on GPU unless kernel cache is cleared,https://github.com/pytorch/pytorch/issues/156010,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2026-01-04T08:16:18Z +issue,165798,matmul/mm/bmm crashes on CUDA 13.0 + Libtorch 2.9 on Ubuntu 24.04,https://github.com/pytorch/pytorch/issues/165798,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-27T06:21:53Z +issue,171092,Load nVidia libraries in reverse order so that the newest version can take control,https://github.com/pytorch/pytorch/issues/171092,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-22T18:56:07Z +issue,163510,building pytorch c++ extension with cuda 13 causes unexpected behavior when pytorch is built with cuda 12.8,https://github.com/pytorch/pytorch/issues/163510,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cpp-extensions",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-20T20:24:14Z +issue,170278,macOS Parallel Builds Fail with jobs > 4,https://github.com/pytorch/pytorch/issues/170278,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: ci, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-18T21:19:28Z +issue,170524,RuntimeError occurred when I train “Resnet”,https://github.com/pytorch/pytorch/issues/170524,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-17T01:15:40Z +issue,170038,Conv2D numerical mismatch vs TensorFlow,https://github.com/pytorch/pytorch/issues/170038,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-15T09:52:08Z +issue,170016,Conv2d produces numerically inconsistent results compared to other DL frameworks (TensorFlow / Paddle / MindSpore),https://github.com/pytorch/pytorch/issues/170016,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-11T16:49:44Z +issue,169342,[MPS] Batch inference produces incorrect results when chunk() views are passed through conv operations,https://github.com/pytorch/pytorch/issues/169342,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: correctness (silent), module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-09T11:42:07Z +issue,169082,Reduce Overhead and Max Autotune NaN,https://github.com/pytorch/pytorch/issues/169082,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: inductor, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-12-04T01:48:55Z +issue,168362,Triton 3.4.0 wheel hash mismatch between PyPI metadata and download server,https://github.com/pytorch/pytorch/issues/168362,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-21T17:41:59Z +issue,167820,Why torch==2.9 compile qwen3 model with block ptr will crash?,https://github.com/pytorch/pytorch/issues/167820,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: vllm, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-18T06:07:13Z +issue,137260,`import torch` takes forever,https://github.com/pytorch/pytorch/issues/137260,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-16T01:37:22Z +issue,167833,Incorrect SHA256 on index for torchvision==0.24.1+cu130,https://github.com/pytorch/pytorch/issues/167833,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-15T20:12:01Z +issue,114206,[PT2] Compile Cold Start (Persistent Cacheing) - AOTAutograd may be bottleneck when `TORCHINDUCTOR_FX_GRAPH_CACHE=1`,https://github.com/pytorch/pytorch/issues/114206,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: compile-time, module: dynamo, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-06T01:38:21Z +issue,105700,Don't use weak ref finalization for freeing resources when code objects die,https://github.com/pytorch/pytorch/issues/105700,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-05T19:46:28Z +issue,82218,OOM during backward() leads to memory leaks,https://github.com/pytorch/pytorch/issues/82218,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-04T16:24:09Z +issue,60477,ROCm miopenStatusInternalError /MIOpen/src/sqlite_db.cpp:109: open memvfs: unable to open database file,https://github.com/pytorch/pytorch/issues/60477,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: rocm,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-04T05:15:26Z +issue,104259,ImportError: libcudnn.so.8: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/104259,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-11-01T14:44:05Z +issue,163311,[build] `CMake (or ninja)` configuration can't index `uv` environment for incremental compilation,https://github.com/pytorch/pytorch/issues/163311,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-29T02:08:13Z +issue,166303,Pytorch Operators on older pytorch version,https://github.com/pytorch/pytorch/issues/166303,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: docs,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-27T16:55:38Z +issue,165590,RuntimeError: non-positive groups is not supported,https://github.com/pytorch/pytorch/issues/165590,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-17T18:59:35Z +issue,163923,Enabling activation checkpoint debug makes throughput gradually slow down to crawl.,https://github.com/pytorch/pytorch/issues/163923,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-15T11:44:30Z +issue,159224,"Import Error: Undefined Symbol, _PyCode_SetExtra: Torch Extensions doesn't seem to support Python 3.12",https://github.com/pytorch/pytorch/issues/159224,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-12T17:07:15Z +issue,155344,PyTorch CPP Extensions fail when same kernel is compiled more than once on ROCm servers,https://github.com/pytorch/pytorch/issues/155344,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: rocm",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-10-09T17:09:00Z +issue,99573,"Jacfwd become slower after update pytorch (""We’ve integrated functorch into PyTorch---Documentation"")",https://github.com/pytorch/pytorch/issues/99573,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-29T23:25:58Z +issue,163347,Incorrect bf16 results in Conv,https://github.com/pytorch/pytorch/issues/163347,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: correctness (silent), module: mkldnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-24T19:17:48Z +issue,137524,FSDP-2 doesnt do overlapping when composed with TP,https://github.com/pytorch/pytorch/issues/137524,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-19T18:06:09Z +issue,161104,Eager CUDAGraph + stream performance,https://github.com/pytorch/pytorch/issues/161104,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-15T17:24:12Z +issue,129992,Got this error when just running my flask script,https://github.com/pytorch/pytorch/issues/129992,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: internals,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-06T11:00:36Z +issue,45273,"Training fast with small dataset, slow with large dataset",https://github.com/pytorch/pytorch/issues/45273,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: data parallel, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-09-03T15:48:42Z +issue,161494,OOM,https://github.com/pytorch/pytorch/issues/161494,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-26T13:50:47Z +issue,160846,Can't run a model trained in MPS-system on CPU-only system.,https://github.com/pytorch/pytorch/issues/160846,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-19T19:24:18Z +issue,146769,MPS Error on sequoia 15.3: NDArray dimension length > INT_MAX',https://github.com/pytorch/pytorch/issues/146769,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: mps,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-17T16:02:17Z +issue,159447,dataloader hits CUDA error: invalid argument in data.pin_memory(device) on RTX 3090,https://github.com/pytorch/pytorch/issues/159447,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-11T17:41:04Z +issue,160261,CPU: Pytorch >= 2.7.0 is broken,https://github.com/pytorch/pytorch/issues/160261,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-11T14:50:53Z +issue,159669,Torch Compile. CudaGraphs Memory leak,https://github.com/pytorch/pytorch/issues/159669,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-10T17:47:35Z +issue,134331,"CMake Error: When installing PyTorch from source, CUDA not being detected.",https://github.com/pytorch/pytorch/issues/134331,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-06T03:30:24Z +issue,129658,[TensorDict - compile] dynamo generator compatibility,https://github.com/pytorch/pytorch/issues/129658,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-08-01T23:36:27Z +issue,159309,torch.nn.InstanceNorm2d CPU/GPU Inconsistency,https://github.com/pytorch/pytorch/issues/159309,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half, module: numerical-reproducibility",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-30T01:37:56Z +issue,150998,Unsupported operand 118,https://github.com/pytorch/pytorch/issues/150998,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-29T14:46:09Z +issue,159196,torch.compile: device of tensor switches after the next statement is run. no repro,https://github.com/pytorch/pytorch/issues/159196,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-28T19:12:35Z +issue,158906,The make lint command got stuck.,https://github.com/pytorch/pytorch/issues/158906,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: lint,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-24T08:34:24Z +issue,111693,"[export] 14k models: AssertionError: graph-captured input # 2, of type , is not among original inputs of types",https://github.com/pytorch/pytorch/issues/111693,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-11T18:16:07Z +issue,157237,"Unexpected, batch size and device dependent NaN propagation in Conv1d",https://github.com/pytorch/pytorch/issues/157237,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: correctness (silent), module: macos, module: nans and infs",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-09T16:15:13Z +issue,151740,mps and cpu backends produce different training results with FFT and Adam,https://github.com/pytorch/pytorch/issues/151740,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: fft, module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-08T23:06:34Z +issue,157709,/lib/x86_64-linux-gnu/libstdc++.so.6: version `GLIBCXX_3.4.30' not found when IMPORT TORCH,https://github.com/pytorch/pytorch/issues/157709,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-07T22:28:00Z +issue,157398,Cannot copy data from one gpu to another using torch,https://github.com/pytorch/pytorch/issues/157398,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-07T22:27:32Z +issue,157360,torch.Tensor.addmm_ The calculation result is inconsistent with the formula calculation result,https://github.com/pytorch/pytorch/issues/157360,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-07-02T23:58:35Z +issue,151667,[MPS] MultiheadAttention with masks and dropout produces NaNs,https://github.com/pytorch/pytorch/issues/151667,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: macos, module: mps, module: nans and infs",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-28T21:26:45Z +issue,151310,cpp_extension.load 'sources' does not support Chinese paths,https://github.com/pytorch/pytorch/issues/151310,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: windows",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-27T00:43:46Z +issue,155060,MPS Memory Leak,https://github.com/pytorch/pytorch/issues/155060,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-06-13T16:57:04Z +issue,152783,"undefined symbol: __nvJitLinkCreate_12_8, version libnvJitLink.so.12",https://github.com/pytorch/pytorch/issues/152783,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-29T13:04:38Z +issue,154302,Inconsistent output of `torch.func.jvp` calculation,https://github.com/pytorch/pytorch/issues/154302,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: functorch",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-27T17:18:30Z +issue,154266,torch dynamo fails on GH200 with world size 5,https://github.com/pytorch/pytorch/issues/154266,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-27T15:02:59Z +issue,153564,Sum difference for equal channels of tensor,https://github.com/pytorch/pytorch/issues/153564,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-23T17:37:40Z +issue,143646,"Getting ""Could not initialize NNPACK! Reason: Unsupported hardware."" warning even though NNPACK is enabled",https://github.com/pytorch/pytorch/issues/143646,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nnpack,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-22T12:59:24Z +issue,152790,[CXX11ABI] torch 2.6.0-cu126 and cu124 have different exported symbols,https://github.com/pytorch/pytorch/issues/152790,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-16T13:12:13Z +issue,153574,Documentation Preview and Build Not Reflecting Latest Changes,https://github.com/pytorch/pytorch/issues/153574,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: doc infra, module: docs",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-15T16:00:10Z +issue,153033,missalignment with differenet shape in F.linear with bf16 dtype,https://github.com/pytorch/pytorch/issues/153033,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: linear algebra, module: padding",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-09T14:42:04Z +issue,90752,Adam (fused=True) issues,https://github.com/pytorch/pytorch/issues/90752,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: optimizer, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-02T21:36:34Z +issue,152592,[ratter-build] Cannot detect CUDA when build from source,https://github.com/pytorch/pytorch/issues/152592,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-05-01T14:32:14Z +issue,152135,Unaccaptable OOMs all the time.,https://github.com/pytorch/pytorch/issues/152135,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-28T17:18:23Z +issue,149508,Adam optimizer ValueError: beta1 as a Tensor,https://github.com/pytorch/pytorch/issues/149508,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-26T04:45:40Z +issue,150912,using torch.compile with torchao at the same time cause stack overflow error,https://github.com/pytorch/pytorch/issues/150912,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-21T21:33:10Z +issue,128290,[apple M2]libc++abi: terminating due to uncaught exception of type c10::Error: Could not find schema for aten::randn,https://github.com/pytorch/pytorch/issues/128290,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-16T00:45:05Z +issue,151201,I can't compile Pytorch 2.0.0 because ninja: error: build.ninja:10911: multiple rules generate caffe2/CMakeFiles/torch_cpu.dir/__/aten/src/ATen/UfuncCPUKernel_add.cpp.DEFAULT.cpp.o,https://github.com/pytorch/pytorch/issues/151201,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-14T14:59:25Z +issue,119546,Too many spawn process,https://github.com/pytorch/pytorch/issues/119546,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-14T03:42:51Z +issue,150921,`M(*[torch.from_numpy(v).to('cpu') for v in inp])` hang when start with `multiprocessing.Process`,https://github.com/pytorch/pytorch/issues/150921,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-10T02:03:07Z +issue,149279,CUDA Assertion Error in Scatter Operation During Training (RTX5090 cu128),https://github.com/pytorch/pytorch/issues/149279,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: scatter & gather ops, module: windows",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-09T22:05:39Z +issue,71545,pip installation SSLError [SSL: CERTIFICATE_VERIFY_FAILED],https://github.com/pytorch/pytorch/issues/71545,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-04-02T00:46:41Z +issue,149937,torch.inverse can't export onnx opset16,https://github.com/pytorch/pytorch/issues/149937,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: onnx,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-03-26T01:27:59Z +issue,130486,"RuntimeError: NVML_SUCCESS == DriverAPI::get()->nvmlInit_v2_() INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":813, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/130486,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-03-09T08:17:43Z +issue,148169,"PyTorch's nightly version no longer includes the CU118, CU124, and CU121 versions",https://github.com/pytorch/pytorch/issues/148169,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-03-03T18:28:27Z +issue,146579,Pytorch model (.pth format) not loading in KServe environment,https://github.com/pytorch/pytorch/issues/146579,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-02-07T12:07:52Z +issue,145827,What changed in torch 2.5.0 that cause photomaker to fail?,https://github.com/pytorch/pytorch/issues/145827,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-29T16:31:42Z +issue,47128,AMP doesn't gracefully handle optimizers for disabled regions,https://github.com/pytorch/pytorch/issues/47128,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-17T19:09:54Z +issue,130728,torch.cuda.empty_cache() doesn't work correctly for the first instance of nn.Module set to cuda (and a potential method to deal with that).,https://github.com/pytorch/pytorch/issues/130728,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-16T14:31:20Z +issue,143494,sympy.C.ConstantInteger has no method name,https://github.com/pytorch/pytorch/issues/143494,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: fx, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-13T17:17:15Z +issue,144071,torch-nightly doesn't support tesla v100,https://github.com/pytorch/pytorch/issues/144071,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-10T04:06:17Z +issue,111547,Bug with as_strided_tensorimpl for MPS devices,https://github.com/pytorch/pytorch/issues/111547,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: mps,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-09T19:15:13Z +issue,126862,"worse results by using the MPS backend, compared to the CPU",https://github.com/pytorch/pytorch/issues/126862,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-09T18:26:34Z +issue,143906,How to correctly asynchronously copy a GPU tensor to a CPU tensor in another process without introducing blocking?,https://github.com/pytorch/pytorch/issues/143906,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-03T18:13:46Z +issue,116766,[Dynamo][DeepSpeed] torch._dynamo.exc.InternalTorchDynamoError: NestedUserFunctionVariable() has no type,https://github.com/pytorch/pytorch/issues/116766,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-02T18:21:36Z +issue,143852,[CPU][Operator] one channel_shuffle test of the operator benchmark has a Performance fluctuation issue,https://github.com/pytorch/pytorch/issues/143852,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2025-01-02T06:39:46Z +issue,143905,The special size tensor containing batches has a difference of a few tens of thousands in calculation results between CPU and GPU,https://github.com/pytorch/pytorch/issues/143905,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-30T00:50:14Z +issue,143800,The tensor-based computation of exponentiation and logarithmic operations is much slower than using NumPy,https://github.com/pytorch/pytorch/issues/143800,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-26T14:53:30Z +issue,143651,`MemoryDenyWriteExecute` in systemd service causes `RuntimeError: could not create a primitive`,https://github.com/pytorch/pytorch/issues/143651,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: error checking, module: mkldnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-23T23:37:38Z +issue,110946,RuntimeError: out_ptr == out_accessor[thread_count_nonzero[tid + 1]].data() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/110946,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-22T04:23:10Z +issue,143338,c10::SmallVector unusable with gcc/g++ 12 and 13 with `-O3`,https://github.com/pytorch/pytorch/issues/143338,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-18T03:16:39Z +issue,143288,EXCEPTION : /python3.11/distutils/core.py,https://github.com/pytorch/pytorch/issues/143288,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: third_party, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-17T08:04:09Z +issue,143259,s.isIntegral(false) INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/143259,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-16T16:20:57Z +issue,137393,Cannot allocate memory for thread-local data: ABORT,https://github.com/pytorch/pytorch/issues/137393,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-14T15:06:49Z +issue,142302,torch.nn.functional.normalize get nan,https://github.com/pytorch/pytorch/issues/142302,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-13T19:47:58Z +issue,139709,less GPU memory used in Centos than ubuntu,https://github.com/pytorch/pytorch/issues/139709,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-12-06T22:52:26Z +issue,126141,Inconsistent gradients in reverse mode and forward mode AD when using `MultiheadAttention`,https://github.com/pytorch/pytorch/issues/126141,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-27T20:02:08Z +issue,126388,RAM Not Freed on CPU After Moving Model with Multiple Transformers to CUDA,https://github.com/pytorch/pytorch/issues/126388,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-27T20:02:07Z +issue,141465,Inconsistent Bias Broadcasting Results in PyTorch GPU `F.linear(...)` and `F.conv2d(...)` Operations,https://github.com/pytorch/pytorch/issues/141465,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-27T16:34:05Z +issue,132226,关于yolov5在mac设备上使用pytorch的mps加速出现的各种问题,https://github.com/pytorch/pytorch/issues/132226,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: mps,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-27T07:56:28Z +issue,141482,Can't use dataparallel,https://github.com/pytorch/pytorch/issues/141482,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: data parallel,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-26T01:45:33Z +issue,140096,upsample_bilinear backward is super slow in bf16 as compared to fp32,https://github.com/pytorch/pytorch/issues/140096,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-15T07:41:36Z +issue,139977,Assigning the weight of the torch.nn.Embedding object to other variables and then performing subsequent operations on it will cause unstable training,https://github.com/pytorch/pytorch/issues/139977,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-13T02:37:14Z +issue,140282,torch.norm produces wrong results,https://github.com/pytorch/pytorch/issues/140282,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-12T00:19:31Z +issue,139884,Persistent memory leak from failed pinned memory allocation,https://github.com/pytorch/pytorch/issues/139884,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory format, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-11T18:11:11Z +issue,111359,I have a trouble with to_symmetric,https://github.com/pytorch/pytorch/issues/111359,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: sparse,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-06T06:29:21Z +issue,137078,"Missing librt.so.1, libdl.so.2, libz.so.1 and libpthread.so.0 in PyTorch Source Build (aarch64)",https://github.com/pytorch/pytorch/issues/137078,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-05T05:24:01Z +issue,34948,TracerWarning: Output nr 1. of the traced function does not match the corresponding output of the Python function,https://github.com/pytorch/pytorch/issues/34948,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-11-01T05:52:54Z +issue,124197,Libtorch crashes docker when included in header file,https://github.com/pytorch/pytorch/issues/124197,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: abi,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-31T17:18:50Z +issue,138821,"Failed to load image Python extension: libc10_cuda.so: cannot open shared object file: No such file or directory warn(f""Failed to load image Python extension: {e}"")",https://github.com/pytorch/pytorch/issues/138821,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-25T23:00:05Z +issue,138455,Tried to instantiate dummy base class,https://github.com/pytorch/pytorch/issues/138455,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: testing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-23T15:05:36Z +issue,130530,Fail to offload FSDP model weights and optimizer states without using CPUOffload(offload_params=True),https://github.com/pytorch/pytorch/issues/130530,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-23T14:53:30Z +issue,121067,RuntimeError,https://github.com/pytorch/pytorch/issues/121067,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-22T18:39:21Z +issue,138208,NaN value produced by F.conv2d (cuda) with large values in input,https://github.com/pytorch/pytorch/issues/138208,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nans and infs",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-22T07:40:01Z +issue,137964,Network training silently fails with MPS backend,https://github.com/pytorch/pytorch/issues/137964,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-18T19:52:17Z +issue,75949,Potential memory leak in Adam optimizer in AMD chips (CPU),https://github.com/pytorch/pytorch/issues/75949,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: memory usage, module: optimizer",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-10-11T18:34:16Z +issue,87550,[onnx] export repeat_interleave TypeError: z_(): incompatible function arguments,https://github.com/pytorch/pytorch/issues/87550,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: onnx,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-09-19T03:33:05Z +issue,135693,Python's REPL trying to do invoke `.T` / `.mH` / `.mT` on 0-dim tensor when doing tab-completion for some reason (default install of Python on new Ubuntu 24.04),https://github.com/pytorch/pytorch/issues/135693,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-09-13T22:10:37Z +issue,94855,torch.compile breaks reproducibility,https://github.com/pytorch/pytorch/issues/94855,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: inductor, module: random, oncall: pt2",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-09-12T18:41:14Z +issue,134993,[bug] import torch failed after install the cpu version of pytorch,https://github.com/pytorch/pytorch/issues/134993,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-09-04T00:40:33Z +issue,133860,torch::tensor function does not work,https://github.com/pytorch/pytorch/issues/133860,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-24T05:08:19Z +issue,104598,ERROR: THESE PACKAGES DO NOT MATCH THE HASHES FROM THE REQUIREMENTS FILE.,https://github.com/pytorch/pytorch/issues/104598,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-21T12:32:49Z +issue,133808,pytorch/aten/src/ATen/cuda/CUDABlas.cpp:1422:28: error,https://github.com/pytorch/pytorch/issues/133808,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-20T01:25:47Z +issue,133093,Inconsistent results of Conv2d in pytorch between CPU and CUDA,https://github.com/pytorch/pytorch/issues/133093,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-19T17:28:18Z +issue,114455,Torch Cpu Memory Leak with FastApi uvicorn.,https://github.com/pytorch/pytorch/issues/114455,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-13T21:37:36Z +issue,131185,"RuntimeError: Expected grad_output.numel() <= std::numeric_limits::max() to be true, but got false.",https://github.com/pytorch/pytorch/issues/131185,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: autograd, module: error checking, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-13T15:49:47Z +issue,30507,"LibTorch, Error in 'xxx': free(): invalid pointer",https://github.com/pytorch/pytorch/issues/30507,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-08-01T03:26:50Z +issue,120606,“libc10.dylib” can’t be opened because Apple cannot check it for malicious software.,https://github.com/pytorch/pytorch/issues/120606,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docs, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-28T09:58:02Z +issue,58392,torch.cuda.amp.autocast nan values when model.eval(),https://github.com/pytorch/pytorch/issues/58392,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-10T15:23:19Z +issue,129643,"There is a bug, only when self",https://github.com/pytorch/pytorch/issues/129643,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-05T03:23:23Z +issue,128939,Unable to install PyTorch on M1 Macos with Python 3.10.14,https://github.com/pytorch/pytorch/issues/128939,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-07-01T08:05:33Z +issue,129303,Incorrect index from torch.mode,https://github.com/pytorch/pytorch/issues/129303,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: reductions,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-06-26T16:40:33Z +issue,128564,internal asset failure,https://github.com/pytorch/pytorch/issues/128564,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-06-13T15:47:41Z +issue,127560,[libtorch] duplicate symbol of vtable,https://github.com/pytorch/pytorch/issues/127560,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-06-10T17:28:02Z +issue,126775,The significant difference between the outputs of torch.reciprocal running on cpu and gpu,https://github.com/pytorch/pytorch/issues/126775,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-06-08T09:07:01Z +issue,127978,torch.addr outputs NaN when computing on inf vectors,https://github.com/pytorch/pytorch/issues/127978,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-06-06T16:56:07Z +issue,128131,Unable to record Memory consumption with `torch.cuda.memory._record_memory_history()`,https://github.com/pytorch/pytorch/issues/128131,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-06-06T15:38:15Z +issue,121222,Segmentation Fault in `torch.nn.Conv1d` starting from torch 2.2.0,https://github.com/pytorch/pytorch/issues/121222,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-27T17:36:30Z +issue,50779,pytorch is built without _GLIBCXX_USE_CXX11_ABI and can cause std::regex crashes (probably),https://github.com/pytorch/pytorch/issues/50779,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-26T01:10:33Z +issue,126760,"""torch.orgqr"" performs differently on cpu and gpu",https://github.com/pytorch/pytorch/issues/126760,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-21T23:07:24Z +issue,125754,Excution difference between Mac and Linux,https://github.com/pytorch/pytorch/issues/125754,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-11T00:05:08Z +issue,125911,"set_device只支持int,但如果碰到k8s分卡vcuda时报错segmentation fault",https://github.com/pytorch/pytorch/issues/125911,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-10T11:06:46Z +issue,47990,Is nn.Conv2d equivalent with Unfold + Matrix Multiplication + Fold ?,https://github.com/pytorch/pytorch/issues/47990,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: correctness (silent)",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-05-02T17:56:18Z +issue,122707,unresovled symbol when compiling a torch c++ app that using libtorch,https://github.com/pytorch/pytorch/issues/122707,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-04-12T21:40:38Z +issue,117617,"torch.topk, when the input contains +nan and -nan, the result is not sure; sometimes -nan is treated greater than normal number, sometimes -nan is the least;",https://github.com/pytorch/pytorch/issues/117617,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-04-10T07:19:06Z +issue,5953,"LBFGS always give nan results, why",https://github.com/pytorch/pytorch/issues/5953,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: optimizer",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-04-02T07:54:21Z +issue,122548,Unexpected device casting from CPU-pinned tensor to GPU tensor.,https://github.com/pytorch/pytorch/issues/122548,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-29T02:20:22Z +issue,71222,torch.linalg.lstsq is nondeterministic,https://github.com/pytorch/pytorch/issues/71222,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: linear algebra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-21T03:09:31Z +issue,121646,Installation fails on python 3.7-3.12,https://github.com/pytorch/pytorch/issues/121646,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-16T18:21:12Z +issue,120905,F.conv2d double precision error,https://github.com/pytorch/pytorch/issues/120905,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-11T17:25:18Z +issue,121287,Libtorch - clang: error: unsupported option '-fopenmp',https://github.com/pytorch/pytorch/issues/121287,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-07T15:10:28Z +issue,121190,The result of BF16 to FP32 is different in class LlamaRMSNorm training twice.,https://github.com/pytorch/pytorch/issues/121190,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: numerical-stability",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-07T05:50:46Z +issue,120791,about tensor.new(storage) issue,https://github.com/pytorch/pytorch/issues/120791,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: python frontend,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-03-06T17:43:12Z +issue,70048,Tensor transfer between gpus doesnt work,https://github.com/pytorch/pytorch/issues/70048,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-02-27T23:06:31Z +issue,120167,"When loading a text file with torch.load, it exhausts the system's physical memory.",https://github.com/pytorch/pytorch/issues/120167,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: serialization",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-02-27T17:15:05Z +issue,119147,Incorrect Results from cdist_backward Operator with Input Dimensions Exceeding int32_max,https://github.com/pytorch/pytorch/issues/119147,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: edge cases",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-02-21T07:47:34Z +issue,120081,undefined reference to `__atomic_exchange_1',https://github.com/pytorch/pytorch/issues/120081,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-02-17T02:36:30Z +issue,118263,Getting error while loading LLAMA_2_7B,https://github.com/pytorch/pytorch/issues/118263,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-25T18:18:17Z +issue,118172,"""PyTorch installation failed.""",https://github.com/pytorch/pytorch/issues/118172,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-24T15:41:02Z +issue,117962,runtime error: unknown parameter,https://github.com/pytorch/pytorch/issues/117962,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-24T03:27:34Z +issue,117210,"When running the code model.to('mps'), the program gets stuck, but it runs fine on the CPU",https://github.com/pytorch/pytorch/issues/117210,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: mps,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-19T07:15:50Z +issue,117469,"Cannot import torch on Ubuntu 20.04 from clean pip install (shared library issues, libcufft and libcudart)",https://github.com/pytorch/pytorch/issues/117469,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-16T18:43:51Z +issue,73568,Bug when using `nn.Linear`,https://github.com/pytorch/pytorch/issues/73568,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-15T12:23:40Z +issue,116931,Previous Torch versions,https://github.com/pytorch/pytorch/issues/116931,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docs",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-12T18:12:15Z +issue,116589,Cannot pickle torch model in version 2.1 when using `autograd.Function` as a model's attribute,https://github.com/pytorch/pytorch/issues/116589,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: serialization, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-10T11:59:45Z +issue,116882,RuntimeError: CUDA out of memory.,https://github.com/pytorch/pytorch/issues/116882,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-09T20:58:03Z +issue,117032,losing precision in torch.sum,https://github.com/pytorch/pytorch/issues/117032,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-09T20:32:31Z +issue,114465,Installation error: 'CMake Error at third_party/benchmark/CMakeLists.txt:304 (message): Failed to determine the source files for the regular expression backend',https://github.com/pytorch/pytorch/issues/114465,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-02T10:05:50Z +issue,50803,pytorch DDP hangs at .backward() call,https://github.com/pytorch/pytorch/issues/50803,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: deadlock",confirm whether existing evidence is enough; otherwise wait for a reproducer,2024-01-01T18:09:04Z +issue,48580,torch.save() fails when attempting to save to mounted drives,https://github.com/pytorch/pytorch/issues/48580,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-12-23T00:48:38Z +issue,114534,`PYTORCH_NO_CUDA_MEMORY_CACHING=1` with `torch.multiprocessing` shared tensors seems to perform use-after-free,https://github.com/pytorch/pytorch/issues/114534,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-12-05T16:24:37Z +issue,114572,"In version 2.1, libtorch needs to be woken up every time it is called after the model is initialized, which means that every time the model is called, it is very slow to predict the first picture.",https://github.com/pytorch/pytorch/issues/114572,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-27T04:56:17Z +issue,113840,torch.einsum is stuck in mp.Process,https://github.com/pytorch/pytorch/issues/113840,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-21T03:19:19Z +issue,113150,"Aborted (core dumped) after Run Pytorch2.0.0, which I compiled by myself",https://github.com/pytorch/pytorch/issues/113150,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-13T17:21:05Z +issue,111158,RuntimeError in run_streaming_llama.py When Using Accelerate with Streaming LLMa Model on A4500 GPU,https://github.com/pytorch/pytorch/issues/111158,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: error checking,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-07T02:02:40Z +issue,113035,Training a network SUPER slow with Pytorch 2.1,https://github.com/pytorch/pytorch/issues/113035,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-06T22:35:20Z +issue,111508,LBFGS accuracy difference between CPU and GPU,https://github.com/pytorch/pytorch/issues/111508,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-06T21:59:26Z +issue,85211,OSError libstdc++.so.6 at import,https://github.com/pytorch/pytorch/issues/85211,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-02T13:58:23Z +issue,112569,"RuntimeError: NVML_SUCCESS == r INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1695392035891/work/c10/cuda/CUDACachingAllocator.cpp"":1154, please report a bug to PyTorch",https://github.com/pytorch/pytorch/issues/112569,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-11-01T22:02:13Z +issue,111419,nonnull error,https://github.com/pytorch/pytorch/issues/111419,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-31T18:17:17Z +issue,111896,ImportError: cannot import name 'external_utils' from partially initialized module 'torch._dynamo',https://github.com/pytorch/pytorch/issues/111896,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docker",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-30T17:13:17Z +issue,110944,gdb core dump when enable DEBUG mode to compile cpu torch in centos!!!,https://github.com/pytorch/pytorch/issues/110944,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-12T13:14:17Z +issue,110701,Error with monai SwinUNETR and checkpointing,https://github.com/pytorch/pytorch/issues/110701,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-12T08:51:37Z +issue,110883,MAX_JOBS ignored when compiling pytorch from source,https://github.com/pytorch/pytorch/issues/110883,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-10T17:49:36Z +issue,110594,Multiprocessing takes forever after on .get() with mp.Queue() (Possible Deadlock),https://github.com/pytorch/pytorch/issues/110594,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-06T10:57:56Z +issue,110588,libtorch.so: error adding symbols: file in wrong format,https://github.com/pytorch/pytorch/issues/110588,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-10-05T14:40:29Z +issue,108493,RWKV + Adam exp_avg_sq will change from positive to negative after loss.backward(),https://github.com/pytorch/pytorch/issues/108493,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-09-27T18:59:36Z +issue,108676,RuntimeError when calling conv_transpose2d with groups,https://github.com/pytorch/pytorch/issues/108676,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-09-27T13:27:19Z +issue,104857,Torch's `LayerNorm` and Adam optimizer vs those in tensorflow,https://github.com/pytorch/pytorch/issues/104857,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, module: numerical-stability, module: optimizer",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-09-22T18:26:41Z +issue,109453,Inconsistent Behavior of `ConvTranspose2d` on CPU and CUDA,https://github.com/pytorch/pytorch/issues/109453,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-09-22T15:31:53Z +issue,109193,"F.conv2d(input, weight, bias, self.stride, RuntimeError: cuDNN error: CUDNN_STATUS_MAPPING_ERROR",https://github.com/pytorch/pytorch/issues/109193,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-09-14T14:45:17Z +issue,86547,Segmentation fault (core dumped) in RTX3090,https://github.com/pytorch/pytorch/issues/86547,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-09-04T06:24:01Z +issue,108013,Installation with rocm5.6 results in error: assert len(weights) == expected_node_count AssertionError,https://github.com/pytorch/pytorch/issues/108013,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: rocm",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-08-28T16:01:56Z +issue,107276,CUBLAS_STATUS_NOT_SUPPORTED,https://github.com/pytorch/pytorch/issues/107276,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-08-17T03:06:54Z +issue,107115,H100 works differently than rtx4090 on same model,https://github.com/pytorch/pytorch/issues/107115,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-08-15T12:26:15Z +issue,107133,torch.inverse throws error when DP but not in DDP or single GPU,https://github.com/pytorch/pytorch/issues/107133,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: edge cases, module: linear algebra",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-08-14T18:46:45Z +issue,21018,free(): invalid pointer Aborted (core dumped),https://github.com/pytorch/pytorch/issues/21018,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-08-04T10:01:58Z +issue,105941,Differences in the results of conv2d calculations in PyTorch 1.8,https://github.com/pytorch/pytorch/issues/105941,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cudnn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-29T15:38:11Z +issue,106126,Libtorch report C10 error when compiling on my own project,https://github.com/pytorch/pytorch/issues/106126,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-28T21:53:43Z +issue,2539,"DataLoader converts cuda FloatTensor into cpu DoubleTensor when shape is (n,)",https://github.com/pytorch/pytorch/issues/2539,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-25T20:31:08Z +issue,104284,`F.conv1d` and `F.conv2d` propagate `nan`'s incorrectly when minibatch > 15,https://github.com/pytorch/pytorch/issues/104284,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: macos, module: nans and infs, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-20T14:26:06Z +issue,55663,torch.nn.functional.log_softmax on Arm CPU gives partly NaN results,https://github.com/pytorch/pytorch/issues/55663,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-19T19:10:40Z +issue,104587,Conv1d step-by-step numerical error,https://github.com/pytorch/pytorch/issues/104587,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: determinism, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-17T17:35:33Z +issue,103752,Pytorch not calling to C code from a docker container,https://github.com/pytorch/pytorch/issues/103752,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: docker",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-07-03T02:41:55Z +issue,91942,jacrev over huber function,https://github.com/pytorch/pytorch/issues/91942,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: functorch,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-06-23T09:27:21Z +issue,103423,The document does not emphasize hidden range in nn.MaxPool2d,https://github.com/pytorch/pytorch/issues/103423,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: docs,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-06-18T09:19:33Z +issue,103424,The document does not emphasize hidden range in nn.Embedding,https://github.com/pytorch/pytorch/issues/103424,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: docs,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-06-18T08:36:41Z +issue,102078,AddressSanitizer: heap-buffer-overflow in test_comprehensive_nn_functional_embedding_bag_cpu_bfloat16,https://github.com/pytorch/pytorch/issues/102078,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: decompositions, module: embedding",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-05-31T04:10:02Z +issue,101404,problem of compilation for torch2.0,https://github.com/pytorch/pytorch/issues/101404,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-05-24T04:11:38Z +issue,63847,RuntimError:CUDA error: an illegal memory access was encountered when copying data to CUDA,https://github.com/pytorch/pytorch/issues/63847,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-05-22T03:05:08Z +issue,101880,Crash on Python // PyArrow //,https://github.com/pytorch/pytorch/issues/101880,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-05-19T20:35:07Z +issue,101331,Pytorch 2.1.0.dev20230512 cuda not available,https://github.com/pytorch/pytorch/issues/101331,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-05-16T01:13:03Z +issue,101321,Speed when installing from source is very low with CUDA 11,https://github.com/pytorch/pytorch/issues/101321,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-05-15T14:14:16Z +issue,99268,Embedding layer tensor shape,https://github.com/pytorch/pytorch/issues/99268,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: embedding,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-04-17T15:52:07Z +issue,25190,Error in python3: double free or corruption (fasttop),https://github.com/pytorch/pytorch/issues/25190,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: dataloader, module: multiprocessing",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-04-10T14:05:31Z +issue,98142,double free or corruption (fasttop),https://github.com/pytorch/pytorch/issues/98142,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-04-10T13:45:17Z +issue,98204,NaN appears when initializing tensor,https://github.com/pytorch/pytorch/issues/98204,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-04-04T13:56:02Z +issue,97872,nn.linear not support bfloat16,https://github.com/pytorch/pytorch/issues/97872,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-31T15:19:48Z +issue,97865,Unable to install torch on python 3.8.16,https://github.com/pytorch/pytorch/issues/97865,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-30T06:55:34Z +issue,97847,CUDA 10.2 cudnn 8.2.4 run Conv2d error,https://github.com/pytorch/pytorch/issues/97847,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-29T14:20:31Z +issue,97135,Incorrect gradient calculation for upsample nearest on CUDA,https://github.com/pytorch/pytorch/issues/97135,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: interpolation, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-24T19:28:09Z +issue,80588,Semi-reproducible random torch.baddbmm NaNs,https://github.com/pytorch/pytorch/issues/80588,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-06T14:17:03Z +issue,95677,Unable to import ``torch.linalg``,https://github.com/pytorch/pytorch/issues/95677,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-03-02T16:35:35Z +issue,90115,Couldn't install pytorch 2.0,https://github.com/pytorch/pytorch/issues/90115,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-02-16T17:03:28Z +issue,93900,"Why does the torch model have no memory leaks under gpu, but there is a memory leak under cpu, torch version 1.10.1",https://github.com/pytorch/pytorch/issues/93900,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-02-16T15:07:32Z +issue,94542,torch.lgamma CUDA driver error,https://github.com/pytorch/pytorch/issues/94542,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: special,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-02-13T22:33:41Z +issue,92794,Segmentation fault when running torch.ge,https://github.com/pytorch/pytorch/issues/92794,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-01-27T02:08:38Z +issue,91245,Segmentation faults in DataLoader (in latest torch version).,https://github.com/pytorch/pytorch/issues/91245,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-01-23T07:40:12Z +issue,91970,Unknown CUDA graph CaptureStatus21852,https://github.com/pytorch/pytorch/issues/91970,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2023-01-10T22:41:57Z +issue,91072,Unexpected behavior when running torch.max in cuda,https://github.com/pytorch/pytorch/issues/91072,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-12-20T18:03:42Z +issue,90521,[A ERROR in Docker] RuntimeError: CUDA error: no kernel image is available for execution on the device,https://github.com/pytorch/pytorch/issues/90521,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docker",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-12-09T17:31:10Z +issue,90138,p.block != nullptr && p.block->ptr != nullptr INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/90138,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-12-06T18:16:47Z +issue,88948,torch.linalg.matrix_rank memory leak,https://github.com/pytorch/pytorch/issues/88948,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-11-16T14:04:53Z +issue,89041,"cross compile pytoch using cmake , get an error : protobuf::protoc: command not found",https://github.com/pytorch/pytorch/issues/89041,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-11-16T00:18:47Z +issue,87597,ninja: build stopped: subcommand failed,https://github.com/pytorch/pytorch/issues/87597,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-10-24T16:04:50Z +issue,86684,[ONNX] AssertionError: A mismatch between the number of arguments (5) and their descriptors (4) was found at symbolic function 'scatter',https://github.com/pytorch/pytorch/issues/86684,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: onnx,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-10-15T22:21:35Z +issue,86877,compile torch from source,https://github.com/pytorch/pytorch/issues/86877,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-10-14T02:34:57Z +issue,86919,Importing torch 1.12.0 breaks subprocess module,https://github.com/pytorch/pytorch/issues/86919,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-10-13T18:45:45Z +issue,75458,Placing model on bfloat16 on CPU make it freeze/hang,https://github.com/pytorch/pytorch/issues/75458,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: deadlock, module: intel",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-10-07T21:50:10Z +issue,84705,Set dtype if tensor converted to numpy,https://github.com/pytorch/pytorch/issues/84705,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: numpy,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-09-13T23:43:48Z +issue,84370,Installation prefix is not passed to CMake appropriately,https://github.com/pytorch/pytorch/issues/84370,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-09-01T21:38:51Z +issue,83901,pytorch 1.12.1 Adam Optimizer Malfunction!!!,https://github.com/pytorch/pytorch/issues/83901,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-08-23T14:12:51Z +issue,82546,Libtorch C++ torch::stack error,https://github.com/pytorch/pytorch/issues/82546,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-08-03T14:16:07Z +issue,82217,backward not available for index and mask,https://github.com/pytorch/pytorch/issues/82217,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sorting and selection",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-07-27T08:07:01Z +issue,70702,conv3d padding=`same` gradgradcheck fails on CUDA,https://github.com/pytorch/pytorch/issues/70702,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: convolution, module: determinism",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-07-25T21:16:21Z +issue,81257,High GPU context memory on Torch 1.11.0 but none on Torch 1.10.0,https://github.com/pytorch/pytorch/issues/81257,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-07-19T23:03:24Z +issue,80302,AttributeError: 'LinearPackedParams' object has no attribute '_modules',https://github.com/pytorch/pytorch/issues/80302,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn, oncall: quantization",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-07-05T15:53:20Z +issue,28950,ConnectionResetError when using dataLoader with pin_memory=True,https://github.com/pytorch/pytorch/issues/28950,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-06-23T09:12:28Z +issue,78831,Mac M1 Build Failure on DEBUG=1,https://github.com/pytorch/pytorch/issues/78831,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-06-03T23:52:35Z +issue,76024,The prediction results of different equipment are inconsistent,https://github.com/pytorch/pytorch/issues/76024,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-31T16:41:29Z +issue,77893,Segmentation fault in _remove_batch_dim,https://github.com/pytorch/pytorch/issues/77893,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: vmap,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-23T17:11:00Z +issue,76960,"When using Rsqrt, the output of the 1/x process is very likely to have nan/inf",https://github.com/pytorch/pytorch/issues/76960,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-09T04:27:11Z +issue,76959,"When using Lambda, the output of the 1/x process is very likely to have nan/inf",https://github.com/pytorch/pytorch/issues/76959,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-09T04:24:45Z +issue,76956,GlobalAvgPool2d causes the inconsistency of output between frameworks,https://github.com/pytorch/pytorch/issues/76956,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-09T04:23:40Z +issue,76955,GaussianNoise causes the inconsistency of output between frameworks,https://github.com/pytorch/pytorch/issues/76955,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-09T04:22:57Z +issue,76954,ReduceSum causes the inconsistency of output between frameworks,https://github.com/pytorch/pytorch/issues/76954,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-09T04:21:51Z +issue,76643,Pytorch can't process special unicode,https://github.com/pytorch/pytorch/issues/76643,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-08T00:42:14Z +issue,76798,`gradcheck` for `torch.solve` may trigger INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/76798,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-05-05T14:28:45Z +issue,75963,Add build support for GCC 11.2,https://github.com/pytorch/pytorch/issues/75963,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-04-25T17:08:41Z +issue,56771,Problems with initial communication between GPUs,https://github.com/pytorch/pytorch/issues/56771,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-04-18T18:20:17Z +issue,75265,Kernel Dies while using conv2d layer / function,https://github.com/pytorch/pytorch/issues/75265,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-04-09T15:34:24Z +issue,2482,"""Shared memory manager connection has timed out""",https://github.com/pytorch/pytorch/issues/2482,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-04-09T07:32:36Z +issue,74670,Can't install Pytorch,https://github.com/pytorch/pytorch/issues/74670,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-04-01T17:49:06Z +issue,65027,CUDA out of memory,https://github.com/pytorch/pytorch/issues/65027,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-03-30T10:08:13Z +issue,69203,Error in SVD cusolver on Linux,https://github.com/pytorch/pytorch/issues/69203,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-03-24T23:32:33Z +issue,35754,pytorch 1.4.0 hangs when using with CUDA >= 10.1,https://github.com/pytorch/pytorch/issues/35754,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-02-22T11:56:24Z +issue,72105,ImportError: cannot import name '_VF' from partially initialized module 'torch',https://github.com/pytorch/pytorch/issues/72105,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-02-21T12:53:57Z +issue,72591,"Can't forward pass conv2d with kernel_size=1, and padding=1",https://github.com/pytorch/pytorch/issues/72591,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: convolution, module: macos",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-02-14T17:48:23Z +issue,69966,torch.nn.DataParallel caused inference failure with cpu set as device on NV machine,https://github.com/pytorch/pytorch/issues/69966,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: data parallel,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-02-13T22:02:39Z +issue,71704,M1 Pro Apple Silicon chip support.,https://github.com/pytorch/pytorch/issues/71704,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: macos, module: nnpack",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-01-25T03:11:08Z +issue,71272,"UserWarning: Seems like `optimizer.step()` has been overridden after learning rate scheduler initialization. Please, make sure to call `optimizer.step()` before `lr_scheduler.step()`. See more details at https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate warnings.warn(""Seems like `optimizer.step()` has been overridden after learning rate scheduler",https://github.com/pytorch/pytorch/issues/71272,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-01-20T16:33:17Z +issue,70038,comile error,https://github.com/pytorch/pytorch/issues/70038,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-01-08T00:34:33Z +issue,70519,RuntimeError: cublas runtime error,https://github.com/pytorch/pytorch/issues/70519,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cublas,confirm whether existing evidence is enough; otherwise wait for a reproducer,2022-01-06T13:57:42Z +issue,60341,C++ version 1.9.0 libtorch dynamic load fails -- GCC only,https://github.com/pytorch/pytorch/issues/60341,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: abi, module: binaries, module: cpp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-12-21T14:50:49Z +issue,24398,Significantly slower in latest version than in 0.4.0,https://github.com/pytorch/pytorch/issues/24398,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-12-20T00:54:32Z +issue,70008,Torch function runtime seemingly dependent on scipy call,https://github.com/pytorch/pytorch/issues/70008,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-12-17T15:04:44Z +issue,66781,lr_scheduler.py / list index out of range,https://github.com/pytorch/pytorch/issues/66781,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-12-13T15:25:36Z +issue,69448,RuntimeError: PyTorch convert function for op 'inverse' not implemented.,https://github.com/pytorch/pytorch/issues/69448,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-12-08T21:33:10Z +issue,68864,Incorrect accuracy pytorch(1.0.1) and libtorch because of nn.LSTM,https://github.com/pytorch/pytorch/issues/68864,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-11-30T01:37:22Z +issue,68144,Unstable buggy calculation.,https://github.com/pytorch/pytorch/issues/68144,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-11-25T00:53:03Z +issue,67740,`torch.utils.data.random_split` example broken in 1.10,https://github.com/pytorch/pytorch/issues/67740,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: data,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-11-10T19:01:37Z +issue,67835,It seems to be a bug！！！,https://github.com/pytorch/pytorch/issues/67835,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-11-05T20:16:28Z +issue,63726,a problem happened in torch.randperm,https://github.com/pytorch/pytorch/issues/63726,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: random,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-10-09T08:49:35Z +issue,64889,"""RuntimeError: CUDA error: out of memory"" for no reason",https://github.com/pytorch/pytorch/issues/64889,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-09-13T16:12:07Z +issue,64560,Private bytes consumption exception,https://github.com/pytorch/pytorch/issues/64560,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-09-10T22:00:33Z +issue,64804,RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation,https://github.com/pytorch/pytorch/issues/64804,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nn,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-09-10T12:38:42Z +issue,64143,AttributeError: module 'torchvision.models' has no attribute 'load_model',https://github.com/pytorch/pytorch/issues/64143,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-09-05T03:23:38Z +issue,64156,Get zero when using torch.matmul and torch.dot with 1-D tensor in torch-1.9.0-cp39-none-macosx_11_0_arm64.whl,https://github.com/pytorch/pytorch/issues/64156,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: arm, module: correctness (silent)",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-08-31T17:32:56Z +issue,61341,inerror: reference to ‘DeviceType’ is ambiguous,https://github.com/pytorch/pytorch/issues/61341,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-08-22T15:48:23Z +issue,63322,lstm's input h0 and c0 bug,https://github.com/pytorch/pytorch/issues/63322,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: numpy, module: rnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-08-17T18:05:54Z +issue,62644,waste too much time in first two pictures,https://github.com/pytorch/pytorch/issues/62644,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-08-11T03:13:14Z +issue,59628,what(): result type Float can't be cast to the desired output type Long,https://github.com/pytorch/pytorch/issues/59628,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-08-07T03:17:55Z +issue,61660,Pytorch model load failure in Gunicorn with Gevent workers,https://github.com/pytorch/pytorch/issues/61660,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: nn, module: numpy",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-08-03T22:21:05Z +issue,62152,Runtime Error: Expected a 'cuda' device type for generator but found 'cpu' in google colab,https://github.com/pytorch/pytorch/issues/62152,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-07-26T10:02:42Z +issue,61243,Pytorch 1.5+ is slower than pytorch 1.3,https://github.com/pytorch/pytorch/issues/61243,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: performance,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-07-26T08:35:24Z +issue,59626,multi scripted model on the same gpu on multi gpu machine,https://github.com/pytorch/pytorch/issues/59626,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multi-gpu,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-07-09T09:31:39Z +issue,60936,The usage of get_stoi() in 0.10.0 and stoi[] in previous versions,https://github.com/pytorch/pytorch/issues/60936,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-30T00:03:22Z +issue,60923,RuntimeError: derivative for im2col_backward is not implemented,https://github.com/pytorch/pytorch/issues/60923,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-29T06:13:34Z +issue,60094,bundled libiomp5 causing segfaults in other libraries that use libomp,https://github.com/pytorch/pytorch/issues/60094,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: macos, module: openmp",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-23T16:20:31Z +issue,56928,torch-1.8.1 wheel verification fails with distlib,https://github.com/pytorch/pytorch/issues/56928,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-22T13:40:05Z +issue,59883,"matmul causing cuDNN error: CUDNN_STATUS_INTERNAL_ERROR at future, unrelated Conv2d",https://github.com/pytorch/pytorch/issues/59883,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-14T20:56:27Z +issue,36649,"RuntimeError: Expected cuda::check_device({sparse_, r_, t, dense}) to be true, but got false.",https://github.com/pytorch/pytorch/issues/36649,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: data parallel,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-06-07T22:36:29Z +issue,54975,CUBLAS_STATUS_EXECUTION_FAILED error on torch >= 1.8.0 and CUDA 11.1,https://github.com/pytorch/pytorch/issues/54975,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-05-19T04:38:03Z +issue,58135,Segmentation fault need help,https://github.com/pytorch/pytorch/issues/58135,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-05-13T23:31:32Z +issue,57794,CUDA error: an illegal memory access was encountered,https://github.com/pytorch/pytorch/issues/57794,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-05-08T02:24:48Z +issue,57796,BatchNorm grad calculation is imprecise,https://github.com/pytorch/pytorch/issues/57796,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: numerical-stability",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-05-07T22:13:28Z +issue,57491,setting all seed still get different result.,https://github.com/pytorch/pytorch/issues/57491,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: random",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-05-05T23:40:05Z +issue,54799,"Vertices=torch.matmul(vertices.unsqueeze(0), rotations_init), RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling `cublasSgemmStridedBatched in CentOS",https://github.com/pytorch/pytorch/issues/54799,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numpy",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-04-22T18:22:05Z +issue,49832,RuntimeError: CUDA error: unspecified launch failure,https://github.com/pytorch/pytorch/issues/49832,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-04-21T15:50:14Z +issue,56019,when building apk with libtorch,https://github.com/pytorch/pytorch/issues/56019,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-04-15T10:35:42Z +issue,12641,Install Jetson TX2 Max Regcount Error,https://github.com/pytorch/pytorch/issues/12641,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: jetson",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-04-15T10:32:43Z +issue,55035,DataLoader performance drop on 4-channel images?,https://github.com/pytorch/pytorch/issues/55035,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-04-07T14:31:57Z +issue,54716,Segmentation fault in PyTorch dataloader,https://github.com/pytorch/pytorch/issues/54716,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-04-01T18:14:05Z +issue,54827,Import error,https://github.com/pytorch/pytorch/issues/54827,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-30T16:58:01Z +issue,54487,"RuntimeError: iter.device(arg).is_cuda() INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1607370116979/work/aten/src/ATen/native/cuda/Loops.cuh"":94, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/54487,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tensoriterator",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-26T02:01:58Z +issue,54577,Question when I use Pytorch Mobile,https://github.com/pytorch/pytorch/issues/54577,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-24T20:33:38Z +issue,24397,SyncBatchNorm error when using model.eval() with DistributedDataParallel,https://github.com/pytorch/pytorch/issues/24397,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: distributed",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-20T11:38:07Z +issue,53464,libtorch reports bad_alloc when used together with darknet,https://github.com/pytorch/pytorch/issues/53464,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-15T17:17:58Z +issue,52989,lr_scheduler _triangular2_scale_fn calculation is overflowed,https://github.com/pytorch/pytorch/issues/52989,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-02T14:23:05Z +issue,10950,Cannot run torch in different sub-interpreters,https://github.com/pytorch/pytorch/issues/10950,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-03-01T21:41:32Z +issue,52515,torch.bmm incorrect with pytorch 1.7.1 and cuda 11,https://github.com/pytorch/pytorch/issues/52515,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-02-22T23:15:36Z +issue,52307,Model produces same outputs for different inputs when using GPU,https://github.com/pytorch/pytorch/issues/52307,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-02-20T20:54:05Z +issue,51275,Error of running,https://github.com/pytorch/pytorch/issues/51275,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-01-29T07:56:16Z +issue,40133,Why does Libtorch use more memory than Pytorch does in Python？,https://github.com/pytorch/pytorch/issues/40133,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-01-21T12:38:13Z +issue,46806,Compile Pytorch with MAGMA Issue,https://github.com/pytorch/pytorch/issues/46806,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-01-20T15:14:54Z +issue,50261,ModuleNotFoundError: No module named '__torch__',https://github.com/pytorch/pytorch/issues/50261,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: serialization",confirm whether existing evidence is enough; otherwise wait for a reproducer,2021-01-08T18:13:08Z +issue,49669,"While training the model, GPU util reaches 100% but no progress happends.",https://github.com/pytorch/pytorch/issues/49669,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: deadlock",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-12-30T03:11:35Z +issue,49870,Tensorboard Error,https://github.com/pytorch/pytorch/issues/49870,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-12-29T20:31:57Z +issue,48579,test_fs_pool fails,https://github.com/pytorch/pytorch/issues/48579,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: tests",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-12-21T09:50:23Z +issue,47832,Error during distributed training,https://github.com/pytorch/pytorch/issues/47832,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-12-04T08:37:37Z +issue,32584,Segmentation Fault (core dumped) with 1.4.0,https://github.com/pytorch/pytorch/issues/32584,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-10-27T20:19:39Z +issue,46153,Getting this error while installing torch and torchvision!,https://github.com/pytorch/pytorch/issues/46153,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-10-12T21:24:32Z +issue,28943,The `@` operation uses too much memory on GPU,https://github.com/pytorch/pytorch/issues/28943,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: memory usage,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-10-09T21:32:55Z +issue,44855,RuntimeError: Caught RuntimeError in replica 1 on device 1.,https://github.com/pytorch/pytorch/issues/44855,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: data parallel,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-09-17T03:49:10Z +issue,42581,Build script complains that gcc is < 6 even though gcc version is 8,https://github.com/pytorch/pytorch/issues/42581,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-09-14T00:41:06Z +issue,44024,"When the pytorch training data reaches a certain epoch, the memory remains unchanged?",https://github.com/pytorch/pytorch/issues/44024,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-09-05T03:01:25Z +issue,43859,Multi-machine multi-gpu training won't start on CNGrid,https://github.com/pytorch/pytorch/issues/43859,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-09-03T04:57:55Z +issue,35870,CUDNN_STATUS_INTERNAL_ERROR with GPU RTX 8000,https://github.com/pytorch/pytorch/issues/35870,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-06-25T23:50:02Z +issue,4334,torch.cuda.device_count() returns 1 using 4 TitanX setup.,https://github.com/pytorch/pytorch/issues/4334,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-06-25T15:31:14Z +issue,39832,Torch Distributed Asynch docs with invalid ENV args,https://github.com/pytorch/pytorch/issues/39832,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rpc",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-06-15T18:09:13Z +issue,3863,Considerable slowdown in Adam.step after a number of epochs with multiple losses,https://github.com/pytorch/pytorch/issues/3863,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-05-26T16:19:25Z +issue,26400,RuntimeError: tensor.ndimension() == static_cast(expected_size.size()) INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/26400,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multi-gpu",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-05-15T05:47:59Z +issue,38155,"Python script name with ""profile.py"" will run twice",https://github.com/pytorch/pytorch/issues/38155,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-05-13T00:39:15Z +issue,36721,[1.4.1] Cuda build fails,https://github.com/pytorch/pytorch/issues/36721,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-04-24T10:19:18Z +issue,37057,CUDA error: device-side assert triggered @ model.cuda(),https://github.com/pytorch/pytorch/issues/37057,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-04-22T18:23:23Z +issue,37066,Saved model behaves differently on same data,https://github.com/pytorch/pytorch/issues/37066,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: serialization,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-04-22T18:07:42Z +issue,37067,can't wrap two models in the same class,https://github.com/pytorch/pytorch/issues/37067,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-04-22T18:04:17Z +issue,35990,"[JIT] Tensor method API behavior discrepancy, Tensor.detach(..)",https://github.com/pytorch/pytorch/issues/35990,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-04-07T18:09:53Z +issue,34937,[JIT]torch.jit.export invaild for pytorch1.4,https://github.com/pytorch/pytorch/issues/34937,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-03-20T18:31:57Z +issue,25904,Python hang after using torch.exp(),https://github.com/pytorch/pytorch/issues/25904,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: deadlock,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-03-10T14:53:52Z +issue,30900,CUDA error: initialization error (multiprocessing) with Python 3.7,https://github.com/pytorch/pytorch/issues/30900,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-02-12T19:43:37Z +issue,22597,Autograd profiler memory leak when use_cuda=True,https://github.com/pytorch/pytorch/issues/22597,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: autograd,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-02-07T22:52:41Z +issue,30830,Torch getting stuck transfering model to GPU in multiple GPU setting,https://github.com/pytorch/pytorch/issues/30830,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-02-04T07:48:07Z +issue,32662,Second-order gradient cause segfault over time,https://github.com/pytorch/pytorch/issues/32662,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-01-30T18:07:59Z +issue,32443,Segmentation fault in lazyInitCUDA -> CUDAHooks::initCUDA -> THCMagma_init -> magma_init,https://github.com/pytorch/pytorch/issues/32443,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: third_party",confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-01-21T22:52:20Z +issue,31660,Cuda error 59 : device-side assert triggered,https://github.com/pytorch/pytorch/issues/31660,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2020-01-02T15:00:07Z +issue,31004,The inference speed of the torch compiled manually is slower than the torch build from official binaries?,https://github.com/pytorch/pytorch/issues/31004,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: performance",confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-12-10T01:24:56Z +issue,30955,error C3203: “templated_iterator”: 未专用化的 类 模板 不能用作 模板 变量，该变量属于 模板 参数“_Ty1”，应为 real 类型,https://github.com/pytorch/pytorch/issues/30955,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: internals,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-12-09T18:28:21Z +issue,29891,Get wrong precision when multi nodes run in docker,https://github.com/pytorch/pytorch/issues/29891,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-11-21T20:31:25Z +issue,28956,RuntimeError: CUDA error: unknown error,https://github.com/pytorch/pytorch/issues/28956,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cuda,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-10-31T13:31:21Z +issue,14352,[libtorch] Catkin_make compilation error,https://github.com/pytorch/pytorch/issues/14352,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-10-27T12:33:01Z +issue,27694,Deployment training model at C + + end,https://github.com/pytorch/pytorch/issues/27694,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs, oncall: jit",confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-10-22T02:14:58Z +issue,22924,[data loader] Graceful data loader threads exit on KeyboardInterrupt,https://github.com/pytorch/pytorch/issues/22924,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: dataloader,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-10-01T17:27:35Z +issue,16654,pytorch was blocked at loss.backward,https://github.com/pytorch/pytorch/issues/16654,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-09-17T13:54:47Z +issue,20134,"ERROR: Command ""python setup.py egg_info"" when dockerfile build",https://github.com/pytorch/pytorch/issues/20134,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-06-26T13:52:53Z +issue,20805,LibTorch :About torch.jit.trace generate model.pt,https://github.com/pytorch/pytorch/issues/20805,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: oncall: jit,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-06-21T03:35:46Z +issue,21179,Failed to install pytorch from source on ubuntu.,https://github.com/pytorch/pytorch/issues/21179,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-06-19T00:08:08Z +issue,19128,pytorch.version.cuda is None when compiling with CUDA support,https://github.com/pytorch/pytorch/issues/19128,needs reproduction,14,"needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-05-28T09:47:17Z +issue,19029,C++ custom module not thread safe,https://github.com/pytorch/pytorch/issues/19029,needs reproduction,14,needs reproduction label present; old update age or low-urgency/stale label; component/oncall labels: module: cpp,confirm whether existing evidence is enough; otherwise wait for a reproducer,2019-04-11T08:06:40Z +pr,189297,[CI] Pin quack to torch's vendored SHA to fix B200 smoke cutlass import crash,https://github.com/pytorch/pytorch/pull/189297,draft/noise,13,"PR is draft; CI lane labels: ciflow/b200, ciflow/trunk",wait for author to mark ready,2026-07-09T00:20:52Z +pr,188739,flex_gemm: add quant lowerings,https://github.com/pytorch/pytorch/pull/188739,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-08T23:12:51Z +pr,188112,flex_gemm: feed local reductions into epilogues,https://github.com/pytorch/pytorch/pull/188112,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-08T23:12:51Z +pr,188470,flex_gemm: extend local reduce aux groups,https://github.com/pytorch/pytorch/pull/188470,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-08T23:12:50Z +pr,189290,Harden Inductor SubprocPool against compile-worker death and teardown hangs,https://github.com/pytorch/pytorch/pull/189290,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-08T21:49:50Z +pr,189230,Support multiple pools in cudagraph_trees,https://github.com/pytorch/pytorch/pull/189230,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-08T21:35:35Z +pr,189304,Use freeze_rng_state in distribution rsample gradient tests,https://github.com/pytorch/pytorch/pull/189304,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-08T21:31:26Z +pr,185424,Add CPython Dynamo agentic loop plan,https://github.com/pytorch/pytorch/pull/185424,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-08T20:53:26Z +pr,189258,[inductor] Replace ExternKernel/ComputedBuffer plumbing tuples with dataclasses + Protocol,https://github.com/pytorch/pytorch/pull/189258,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for author to mark ready,2026-07-08T19:12:16Z +pr,186302,Install Rust toolchain in PyTorch CI/CD build images (#186302),https://github.com/pytorch/pytorch/pull/186302,draft/noise,13,"PR is draft; CI lane labels: ciflow/docker, ciflow/inductor",wait for author to mark ready,2026-07-08T17:49:26Z +pr,188682,Limit ROCm build targets to architectures covered by testing,https://github.com/pytorch/pytorch/pull/188682,draft/noise,13,"PR is draft; CI lane labels: ciflow/rocm-mi300, ciflow/rocm-mi350, ciflow/trunk",wait for author to mark ready,2026-07-08T17:25:27Z +pr,187859,[functorch] Add torch._functorch.aot_autograd.compile_to_python,https://github.com/pytorch/pytorch/pull/187859,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-07-08T16:57:39Z +pr,189253,[TEST] Refactor test_c10d_spawn_ucc.py with hw_classification,https://github.com/pytorch/pytorch/pull/189253,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-08T12:01:34Z +pr,188653,Add test decorator for device-type-specific RNG seed setting.,https://github.com/pytorch/pytorch/pull/188653,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-08T10:51:18Z +pr,188388,[WIP][XPU][Test]Port 12 UT test suites to Intel GPU,https://github.com/pytorch/pytorch/pull/188388,draft/noise,13,PR is draft; CI lane labels: ciflow/xpu,wait for author to mark ready,2026-07-08T08:28:23Z +pr,183041,[FSDP1] Test forward overlap with use_orig_params,https://github.com/pytorch/pytorch/pull/183041,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-08T08:18:44Z +pr,189238,[XPU] Register linalg._powsum and _foreach_powsum.Scalar for XPU,https://github.com/pytorch/pytorch/pull/189238,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-08T06:31:23Z +pr,185816,[WIP] [PoC] Enable CPU FlexAttention on AArch64,https://github.com/pytorch/pytorch/pull/185816,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan, ciflow/trunk",wait for author to mark ready,2026-07-07T17:14:11Z +pr,189097,[RFC] Expose Kineto custom logger registration,https://github.com/pytorch/pytorch/pull/189097,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-07T13:56:30Z +pr,188195,Adding OpenReg allocator stats tests,https://github.com/pytorch/pytorch/pull/188195,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-07T11:52:44Z +pr,187318,[xpu]Implement scaled_mm for XPU - combined,https://github.com/pytorch/pytorch/pull/187318,draft/noise,13,"PR is draft; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300",wait for author to mark ready,2026-07-07T08:21:54Z +pr,189020,Reorder condition to handle empty tensor the same for file_system & file_descriptor,https://github.com/pytorch/pytorch/pull/189020,draft/noise,13,PR is draft; CI lane labels: ciflow/trunk,wait for author to mark ready,2026-07-07T07:04:43Z +pr,182379,"[DTensor] Custom algorithm dispatch mechanism for DTensor, one-sided GEMM algorithm",https://github.com/pytorch/pytorch/pull/182379,draft/noise,13,"PR is draft; CI lane labels: ciflow/dtensor, ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-07T01:08:45Z +pr,189099,[Profiler][NOT FOR LAND] CI test debugging,https://github.com/pytorch/pytorch/pull/189099,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-06T23:37:02Z +pr,182841,[CI] No-op docker change to test current build state,https://github.com/pytorch/pytorch/pull/182841,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-06T22:51:36Z +pr,182838,"[CI] Fix docker builds: pin miniforge, use --extra-index-url",https://github.com/pytorch/pytorch/pull/182838,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-06T22:51:34Z +pr,189081,Add tp_name property to Variable Tracker,https://github.com/pytorch/pytorch/pull/189081,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-06T20:16:23Z +pr,181659,[FSDP2] Add recurse= to set_symm_mem_for_comm and set_force_sum_reduction_for_comms,https://github.com/pytorch/pytorch/pull/181659,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-06T08:02:22Z +pr,188918,[dynamo] Model list subclass __new__/__init__ arg handling to match CPython,https://github.com/pytorch/pytorch/pull/188918,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-03T21:13:49Z +pr,187540,"[DRAFT] Fix forward AD inconsistency for pow(abs(complex), complex)",https://github.com/pytorch/pytorch/pull/187540,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-03T16:56:53Z +pr,188888,Add Reusable Event support to XPUEvent,https://github.com/pytorch/pytorch/pull/188888,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-03T11:07:34Z +pr,188782,[CI] Benchmark TorchInductor perf against 2.13.0 RC,https://github.com/pytorch/pytorch/pull/188782,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-02T22:50:16Z +pr,188781,[CI] Benchmark TorchInductor perf against 2.12.1 release (baseline),https://github.com/pytorch/pytorch/pull/188781,draft/noise,13,PR is draft,wait for author to mark ready,2026-07-02T10:36:31Z +pr,186282,Fix DTensor view-unflatten when one output dim is sharded by 2+ mesh axes,https://github.com/pytorch/pytorch/pull/186282,draft/noise,13,"PR is draft; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-07-02T04:49:37Z +pr,188593,"[ROCm] Add skipIfRocmVersionAtLeast([7, 14]) skips for ROCm 7.14 known failures",https://github.com/pytorch/pytorch/pull/188593,draft/noise,13,"PR is draft; CI lane labels: ciflow/dtensor, ciflow/inductor",wait for author to mark ready,2026-07-01T01:23:01Z +pr,188565,[ROCm] Add initial GFX1250 architecture support,https://github.com/pytorch/pytorch/pull/188565,draft/noise,13,"PR is draft; CI lane labels: ciflow/b200, ciflow/docker, ciflow/h100, ciflow/h100-symm-mem",wait for author to mark ready,2026-06-30T20:07:06Z +pr,186980,Move header-only c10/core headers to torch/headeronly,https://github.com/pytorch/pytorch/pull/186980,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-30T16:04:35Z +pr,178305,[XPU][oneDNN] Enforce const/mutable pointer semantics for mkldnn xpu detail,https://github.com/pytorch/pytorch/pull/178305,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-30T12:13:40Z +pr,188447,Add DTensor shard propagation skill,https://github.com/pytorch/pytorch/pull/188447,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-29T19:31:11Z +pr,181145,[CuteDSL][Native Ops] group norm kernel,https://github.com/pytorch/pytorch/pull/181145,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-28T18:52:57Z +pr,183688,[MPS] Add torch.mps.MPSGraph capture/replay via MTLIndirectCommandBuffer,https://github.com/pytorch/pytorch/pull/183688,draft/noise,13,PR is draft; CI lane labels: ciflow/mps,wait for author to mark ready,2026-06-27T22:52:51Z +pr,182749,[MPS] Zero-copy CPU->MPS via memcpy on Apple Silicon UMA,https://github.com/pytorch/pytorch/pull/182749,draft/noise,13,PR is draft; CI lane labels: ciflow/mps,wait for author to mark ready,2026-06-27T22:52:51Z +pr,182736,[MPS] Zero-copy MPS->CPU transfer via unified memory in copy_from_mps_,https://github.com/pytorch/pytorch/pull/182736,draft/noise,13,PR is draft; CI lane labels: ciflow/mps,wait for author to mark ready,2026-06-27T22:52:51Z +pr,182731,[MPS] Zero-copy scalar extraction via unified memory in _local_scalar_dense,https://github.com/pytorch/pytorch/pull/182731,draft/noise,13,PR is draft; CI lane labels: ciflow/mps,wait for author to mark ready,2026-06-27T22:52:51Z +pr,182791,[MPS] UMA fast path for fill/zero via CPU memset/std::fill,https://github.com/pytorch/pytorch/pull/182791,draft/noise,13,PR is draft; CI lane labels: ciflow/mps,wait for author to mark ready,2026-06-27T22:52:50Z +pr,180573,Avoid unnecessary refcount bump in inplace op Python bindings,https://github.com/pytorch/pytorch/pull/180573,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-26T22:01:48Z +pr,188257,[RFC] S3-over-RDMA checkpoint storage backends via NVIDIA cuObject,https://github.com/pytorch/pytorch/pull/188257,draft/noise,13,PR is draft; CI lane labels: ciflow/torchtitan,wait for author to mark ready,2026-06-26T18:17:07Z +pr,181528,sdp: warn when enable_gqa=True silently falls back to MATH,https://github.com/pytorch/pytorch/pull/181528,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-25T22:59:58Z +pr,188120,[FSDP2] Per-FQN offload policy map and cpu_offload_by_budget helper,https://github.com/pytorch/pytorch/pull/188120,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-25T00:31:33Z +pr,188065,[AOTI] measure performance with AOTI_RUNTIME_CHECK_INPUTS=1,https://github.com/pytorch/pytorch/pull/188065,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-24T19:03:06Z +pr,181557,[XPU] Enable grouped_mm and scaled_grouped_mm with OneDNN,https://github.com/pytorch/pytorch/pull/181557,draft/noise,13,"PR is draft; CI lane labels: ciflow/b200, ciflow/h100, ciflow/inductor, ciflow/rocm-mi300",wait for author to mark ready,2026-06-22T15:09:20Z +pr,187835,[profiler] Enable OVERHEAD activity type for the XPU backend,https://github.com/pytorch/pytorch/pull/187835,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-22T14:21:55Z +pr,181187,[spmd types] fully_shard but in DTensor land,https://github.com/pytorch/pytorch/pull/181187,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-22T03:48:43Z +pr,180317,[xpu][inductor] Add Dual GEMM SiLU-Mul fusion pass for XPU,https://github.com/pytorch/pytorch/pull/180317,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-19T06:18:01Z +pr,187685,Name AOTAutograd backward gradient outputs grad_ (#110700),https://github.com/pytorch/pytorch/pull/187685,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-18T18:36:46Z +pr,187647,Remove XPU skips on test_partitioner_unbacked_symints and test_sdpa_s…,https://github.com/pytorch/pytorch/pull/187647,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-18T11:25:08Z +pr,187612,docker-builds: move to default runner group (regular OSDC pool),https://github.com/pytorch/pytorch/pull/187612,draft/noise,13,PR is draft; CI lane labels: ciflow/docker,wait for author to mark ready,2026-06-17T23:59:57Z +pr,187620,[FSDP2] Add PartialOffloadPolicy for fractional CPU parameter offload,https://github.com/pytorch/pytorch/pull/187620,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-17T22:49:51Z +pr,187564,"Register `linalg_qr`, `geqrf`, and `linalg_householder_product` for XPU",https://github.com/pytorch/pytorch/pull/187564,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-17T09:16:14Z +pr,177649,BLAS sparse ops: dense arguments -- extend support beyond just dense and non-overlapping inputs.,https://github.com/pytorch/pytorch/pull/177649,draft/noise,13,PR is draft; CI lane labels: ciflow/trunk,wait for author to mark ready,2026-06-16T12:24:10Z +pr,186982,Move header-only c10/util concurrency/lifetime utils to torch/headeronly,https://github.com/pytorch/pytorch/pull/186982,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-16T06:06:41Z +pr,186983,Move header-only c10/util math utils to torch/headeronly,https://github.com/pytorch/pytorch/pull/186983,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-16T06:06:39Z +pr,186981,Move header-only c10/util metaprogramming/type utils to torch/headeronly,https://github.com/pytorch/pytorch/pull/186981,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-16T06:06:39Z +pr,186985,Move header-only c10/util low-level utils to torch/headeronly,https://github.com/pytorch/pytorch/pull/186985,draft/noise,13,PR is draft; CI lane labels: ciflow/inductor,wait for author to mark ready,2026-06-16T06:06:31Z +pr,176393,Make avgpool3d support fp16/bf16,https://github.com/pytorch/pytorch/pull/176393,draft/noise,13,PR is draft; CI lane labels: ciflow/trunk,wait for author to mark ready,2026-06-15T03:33:18Z +pr,180460,Decompose all_gather+matmul into P2P+tiled mm for fine-grained overlap,https://github.com/pytorch/pytorch/pull/180460,draft/noise,13,"PR is draft; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-14T16:59:39Z +pr,187213,Issue-106571: Lint Fixes - Rule B007 - gb_registry_linter - Replace …,https://github.com/pytorch/pytorch/pull/187213,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-12T21:17:36Z +pr,180087,[DTensor] Error on random ops with device mismatch,https://github.com/pytorch/pytorch/pull/180087,draft/noise,13,"PR is draft; CI lane labels: ciflow/dtensor, ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-06-10T07:55:25Z +pr,186852,[profiler][cupti] ProfilerObserver + record_function annotation routing,https://github.com/pytorch/pytorch/pull/186852,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-10T03:28:56Z +pr,186851,[profiler][cupti] ProfilerObserver + record_function annotation routing,https://github.com/pytorch/pytorch/pull/186851,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-10T02:16:15Z +pr,185217,Fix Debug/RelWithDebInfo PyTorch+XPU on Windows,https://github.com/pytorch/pytorch/pull/185217,draft/noise,13,PR is draft,wait for author to mark ready,2026-06-01T12:10:10Z +pr,185178,[FSDP][AC] Bypass SAC tracking for all-gather internals,https://github.com/pytorch/pytorch/pull/185178,draft/noise,13,"PR is draft; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-29T07:01:24Z +pr,179223,[Test] Refactor test_higher_order_ops.py to be device-generic,https://github.com/pytorch/pytorch/pull/179223,draft/noise,13,"PR is draft; CI lane labels: ciflow/dynamo, ciflow/inductor",wait for author to mark ready,2026-05-25T06:33:10Z +pr,185036,[inductor] Prototype channel-resident reduction suffix,https://github.com/pytorch/pytorch/pull/185036,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T06:14:35Z +pr,185037,[inductor] Use persistent path for small fused online softmax,https://github.com/pytorch/pytorch/pull/185037,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T06:14:34Z +pr,185027,[inductor] Fold CE backward one-hot row sums,https://github.com/pytorch/pytorch/pull/185027,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T06:14:30Z +pr,185028,[inductor] Reassociate cat under surviving reductions,https://github.com/pytorch/pytorch/pull/185028,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T06:14:26Z +pr,184941,[inductor] Enable reduction epilogue fusion by deferring split decision,https://github.com/pytorch/pytorch/pull/184941,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T04:53:20Z +pr,185030,[inductor] Fold masked invariant expand reductions,https://github.com/pytorch/pytorch/pull/185030,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T04:53:15Z +pr,185032,[inductor] Reuse expensive broadcast factors in pointwise tiling,https://github.com/pytorch/pytorch/pull/185032,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T04:53:14Z +pr,184932,[inductor] Use non-online softmax for huge inner reductions,https://github.com/pytorch/pytorch/pull/184932,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T03:27:57Z +pr,184936,[inductor] Avoid masks for zero-stride cat source loads,https://github.com/pytorch/pytorch/pull/184936,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T03:27:52Z +pr,184940,[inductor] Enable reduction+transpose fusion via loop reordering,https://github.com/pytorch/pytorch/pull/184940,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T02:15:28Z +pr,184929,[inductor] Sink single-use dropout randoms into fused epilogues,https://github.com/pytorch/pytorch/pull/184929,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T01:27:26Z +pr,184998,[inductor] Use one warp for tiny inner persistent reductions,https://github.com/pytorch/pytorch/pull/184998,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T01:25:53Z +pr,185003,[inductor] Tune small multi-output mix-order reductions,https://github.com/pytorch/pytorch/pull/185003,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-24T01:25:46Z +pr,184931,[inductor] Use persistent reduction for R=2048 online softmax,https://github.com/pytorch/pytorch/pull/184931,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-23T23:33:20Z +pr,184935,[inductor] Canonicalize duplicate mean and var reductions,https://github.com/pytorch/pytorch/pull/184935,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-23T23:02:53Z +pr,184777,Refactor test_dlpack for out of tree backend,https://github.com/pytorch/pytorch/pull/184777,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-23T14:55:29Z +pr,184724,[XPU][SDPA] Add XPU flash_attention_forward dispatch with dropout fallback,https://github.com/pytorch/pytorch/pull/184724,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-22T05:18:57Z +pr,182811,Reenable test_addcmul_alpha_one_fma_parity dtypes F32 and F64 on XPU.,https://github.com/pytorch/pytorch/pull/182811,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-21T14:14:04Z +pr,177636,Refactor platform support utilities,https://github.com/pytorch/pytorch/pull/177636,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/b200, ciflow/dtensor, ciflow/h100, ciflow/inductor",wait for author to mark ready,2026-05-21T09:06:13Z +pr,184318,[opaque obj] Check opaque object schema types at dispatch,https://github.com/pytorch/pytorch/pull/184318,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-20T20:10:14Z +pr,184498,[profiler] Propagate record_funtion_id from TorchOp ancestor to Kinet…,https://github.com/pytorch/pytorch/pull/184498,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-20T14:21:28Z +pr,180888,Update requirements-docs.txt,https://github.com/pytorch/pytorch/pull/180888,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-20T12:46:35Z +pr,182850,[inductor][symm_mem] Add low-contention allgather variants,https://github.com/pytorch/pytorch/pull/182850,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/h100-symm-mem, ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-18T14:13:51Z +pr,183903,[XPU] Relax Muon float32 tolerance in test_state_dict_with_cuda_params,https://github.com/pytorch/pytorch/pull/183903,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-18T02:33:45Z +pr,183594,TMA scatter_reduce min/max for sm_90+ (CUDA 12.8+),https://github.com/pytorch/pytorch/pull/183594,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-13T20:07:40Z +pr,183360,Prototype lazy activation checkpoint compile,https://github.com/pytorch/pytorch/pull/183360,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-13T03:28:06Z +pr,183359,Prototype lazy autograd.Function backward compile,https://github.com/pytorch/pytorch/pull/183359,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-13T02:15:17Z +pr,183354,[CI] Match MI355 ROCm shards with trunk,https://github.com/pytorch/pytorch/pull/183354,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2026-05-11T19:30:26Z +pr,183028,[Inductor] Generalize parallel Triton precompile to all multi-config kernels,https://github.com/pytorch/pytorch/pull/183028,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-11T18:22:26Z +pr,183253,[BE]: Improve torch utils checkpoint typing,https://github.com/pytorch/pytorch/pull/183253,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,wait for author to mark ready,2026-05-11T10:16:32Z +pr,183211,[FSDP] Break cross-language ref cycle keeping FlatParameter alive,https://github.com/pytorch/pytorch/pull/183211,draft/noise,5,"PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor, ciflow/torchtitan",wait for author to mark ready,2026-05-11T03:11:52Z +pr,178850,[xpu][feature] Add fork-safe is_available() via PYTORCH_ZES_BASED_XPU_CHECK env var,https://github.com/pytorch/pytorch/pull/178850,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/trunk,wait for author to mark ready,2026-03-31T13:27:45Z +pr,153557,"[PP] wip, allow grad to be None",https://github.com/pytorch/pytorch/pull/153557,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/torchtitan,wait for author to mark ready,2026-03-20T22:40:44Z +pr,153317,[associative_scan] Autograd for additional inputs,https://github.com/pytorch/pytorch/pull/153317,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for author to mark ready,2025-07-09T18:22:39Z +pr,147653,Bf16 fused adam(W),https://github.com/pytorch/pytorch/pull/147653,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2025-05-14T17:02:21Z +pr,142295,[inductor] Decide cooperative RSPLIT with same algorithm as split reductions,https://github.com/pytorch/pytorch/pull/142295,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for author to mark ready,2025-04-10T16:01:15Z +pr,143812,[inductor] Used fixed configs for contiguous reductions,https://github.com/pytorch/pytorch/pull/143812,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for author to mark ready,2025-04-10T15:54:03Z +pr,146593,[NOT FOR LANDING] experimental NVSHMEM integration,https://github.com/pytorch/pytorch/pull/146593,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2025-04-09T18:59:30Z +pr,139524,Make anomaly mode asynchronous,https://github.com/pytorch/pytorch/pull/139524,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2025-01-06T02:46:38Z +pr,125270,[do not review] Add tensor equality subsystem and get it working with compile,https://github.com/pytorch/pytorch/pull/125270,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for author to mark ready,2024-12-17T22:23:20Z +pr,125428,[do not review] support max_seqlen,https://github.com/pytorch/pytorch/pull/125428,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for author to mark ready,2024-10-07T03:35:08Z +pr,125326,[do not review] Use union find and new Python NestedInt with NJT,https://github.com/pytorch/pytorch/pull/125326,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for author to mark ready,2024-10-07T03:35:07Z +pr,126199,[POC] use a better default for num_threads,https://github.com/pytorch/pytorch/pull/126199,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2024-07-15T05:46:38Z +pr,88106,WIP: feat: LARS optimizer,https://github.com/pytorch/pytorch/pull/88106,draft/noise,5,PR is draft; PR has not updated for at least 45 days; CI lane labels: ciflow/inductor,wait for author to mark ready,2024-01-30T19:47:12Z +pr,84843,Support different NSE in batches of CSR and CSC tensors,https://github.com/pytorch/pytorch/pull/84843,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2023-12-12T17:41:33Z +pr,85699,Add vector-Jacobian-products for a subset of nvFuser-supported prims; add backward support for nvprims,https://github.com/pytorch/pytorch/pull/85699,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2023-04-17T00:36:00Z +pr,84545,Add nvfuser support for prims.copy_to,https://github.com/pytorch/pytorch/pull/84545,draft/noise,5,PR is draft; PR has not updated for at least 45 days,wait for author to mark ready,2023-04-17T00:36:00Z +issue,184036,"torch.nn.PairwiseDistance returns +Inf for moderately large p (overflow in fp32, no log-space rewrite)",https://github.com/pytorch/pytorch/issues/184036,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: distance functions, module: edge cases, module: linear algebra",scan when time allows,2026-06-22T22:12:43Z +issue,173225,[RFC] Renaming MPS backend as Metal backend,https://github.com/pytorch/pytorch/issues/173225,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2026-05-30T13:50:29Z +issue,177082,Stable Linux binaries of LibTorch have broken soname and versioning in libgomp,https://github.com/pytorch/pytorch/issues/177082,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2026-05-09T16:44:06Z +issue,111950,Operators that return dynamic-shape outputs that require_grad choke in AOTAutograd / unbacked SymInts are reallocated,https://github.com/pytorch/pytorch/issues/111950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: dynamic shapes, module: pt2-dispatcher, oncall: pt2",scan when time allows,2026-03-31T23:22:16Z +issue,94652,dynamo: handle contiguous graph breaks,https://github.com/pytorch/pytorch/issues/94652,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: graph breaks, oncall: pt2",scan when time allows,2026-03-11T22:17:58Z +issue,102898,After dynamo minifier generates repros that don't entirely match what we minified over,https://github.com/pytorch/pytorch/issues/102898,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamo,scan when time allows,2026-03-11T20:36:15Z +issue,174986,`torch.istft` returns an unhelpful internal error,https://github.com/pytorch/pytorch/issues/174986,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: fft",scan when time allows,2026-03-02T15:33:09Z +issue,169058,[MPS] Inconsistent overflow behavior when casting Float to Int8 compared to CPU/CUDA (Saturation vs Wraparound),https://github.com/pytorch/pytorch/issues/169058,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: mps",scan when time allows,2026-02-19T22:18:23Z +issue,165537,"`torch.linspace` with `inf` endpoint and `float16` dtype returns all `nan`,inconsistent with NumPy",https://github.com/pytorch/pytorch/issues/165537,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,scan when time allows,2026-02-19T18:52:35Z +issue,122290,"In a simple model, exporting onnx loses model operations",https://github.com/pytorch/pytorch/issues/122290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2026-01-16T15:09:25Z +issue,170573,"Silent Heap Buffer Overflow in standard ops (add, pinv) via implicit view resizing",https://github.com/pytorch/pytorch/issues/170573,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2026-01-10T00:01:07Z +issue,172085,Many test failures when GPU does not support BF16,https://github.com/pytorch/pytorch/issues/172085,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: tests",scan when time allows,2026-01-09T21:17:56Z +issue,170563,vLLM v1/distributed/test_eagle_dp.py Tests failing on PyTorch 2.10,https://github.com/pytorch/pytorch/issues/170563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vllm,scan when time allows,2026-01-09T18:05:14Z +issue,149187,Issue with Shared CUDA Tensor Reference Counting in Multi-Processing,https://github.com/pytorch/pytorch/issues/149187,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2026-01-09T10:18:54Z +issue,149062,Reserve customized modules in torch.compile's dynamo tracer,https://github.com/pytorch/pytorch/issues/149062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2026-01-09T07:44:18Z +issue,171697,"When the pinned_use_background_threads option is first set to True and then set to False, the background threads may not stop",https://github.com/pytorch/pytorch/issues/171697,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2026-01-09T07:06:49Z +issue,172034,Become compatible with CMAKE_CUDA_ARCHITECTURES,https://github.com/pytorch/pytorch/issues/172034,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions, module: cuda",scan when time allows,2026-01-09T00:26:36Z +issue,171534,Add proper FMA support for Pallas backend,https://github.com/pytorch/pytorch/issues/171534,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: pallas",scan when time allows,2026-01-08T20:27:58Z +issue,171956,[Inductor] Eliminate triton-related hardcode,https://github.com/pytorch/pytorch/issues/171956,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: privateuse1,scan when time allows,2026-01-08T07:16:24Z +issue,160787,[Intel Arc XPU] NaN Values in Gradients after SDPA with is_causal,https://github.com/pytorch/pytorch/issues/160787,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xpu,scan when time allows,2026-01-08T07:10:02Z +issue,168935,[ROCm] enabling compile should not cause signficiant accuracy loss on many important models,https://github.com/pytorch/pytorch/issues/168935,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2026-01-08T06:04:05Z +issue,166114,[BUG]interpolate operator triggers severe performance degradation when enabling deterministic computation,https://github.com/pytorch/pytorch/issues/166114,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2026-01-07T20:00:36Z +issue,169893,Investigate which submodules in third_party/ can be omitted from stable header hiding,https://github.com/pytorch/pytorch/issues/169893,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cpp-extensions",scan when time allows,2026-01-07T16:36:19Z +issue,144766,Matmul with int32 parameters on Intel GPU leads to errors,https://github.com/pytorch/pytorch/issues/144766,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xpu,scan when time allows,2026-01-07T07:32:06Z +issue,151549,multiprocessing.pool.MaybeEncodingError Reason: 'RuntimeError('_share_fd_: only available on CPU')',https://github.com/pytorch/pytorch/issues/151549,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: xpu",scan when time allows,2026-01-07T07:30:08Z +issue,151308,[Intel GPU][UT failure] Depthwise conv related UT failures,https://github.com/pytorch/pytorch/issues/151308,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xpu,scan when time allows,2026-01-07T07:26:00Z +issue,159232,Support installing Python bindings in CMake,https://github.com/pytorch/pytorch/issues/159232,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2026-01-07T00:16:27Z +issue,171771,[RFC] Optimize CUDA Allocator's Synchronization Behavior During OOM,https://github.com/pytorch/pytorch/issues/171771,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2026-01-06T22:56:52Z +issue,169073,Replace deprecated __args__ type access in typing libraries with proper typing.get_args,https://github.com/pytorch/pytorch/issues/169073,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2026-01-06T21:00:08Z +issue,168376,`TORCH_SHOW_CPP_STACKTRACES=1` often fails to display all function names,https://github.com/pytorch/pytorch/issues/168376,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: debug-build, module: error checking",scan when time allows,2026-01-06T19:40:01Z +issue,74627,Feature request: INT4 format support,https://github.com/pytorch/pytorch/issues/74627,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2026-01-06T13:15:18Z +issue,95100,"RuntimeError: view_as_complex is only supported for half, float and double tensors, but got a tensor of scalar type: BFloat16",https://github.com/pytorch/pytorch/issues/95100,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: complex",scan when time allows,2026-01-06T10:09:11Z +issue,149694,torch.det results in nans for torch.func.hessian,https://github.com/pytorch/pytorch/issues/149694,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: numerical-stability",scan when time allows,2026-01-05T23:58:26Z +issue,171302,Make ConcatOptim class for using multiple optimizers as one,https://github.com/pytorch/pytorch/issues/171302,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2026-01-05T18:56:36Z +issue,171263,"[CUDA Graphs] Capture failure leaves generator state corrupted, causing ""Offset increment outside graph capture encountered unexpectedly"" on subsequent RNG operations",https://github.com/pytorch/pytorch/issues/171263,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2026-01-05T18:54:40Z +issue,171401,Feedback about Flatten,https://github.com/pytorch/pytorch/issues/171401,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2026-01-05T18:53:41Z +issue,171323,Segmentation fault when calling `torch.tensor()` after importing xgboost,https://github.com/pytorch/pytorch/issues/171323,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, module: openmp",scan when time allows,2026-01-05T18:30:05Z +issue,171286,Memory Viz Y-Axis Uses SI Units (GB) Instead of Binary Units (GiB),https://github.com/pytorch/pytorch/issues/171286,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: profiler",scan when time allows,2026-01-05T18:17:07Z +issue,171536,`torch.nn.PixelUnshuffle` throws RuntimeError: Stride calculation overflowed on CUDA with empty input and large downscale factor (CPU succeeds),https://github.com/pytorch/pytorch/issues/171536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2026-01-05T15:02:18Z +issue,107352,dist.scatter is incompatible with transpose/permute operation,https://github.com/pytorch/pytorch/issues/107352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2026-01-05T07:54:13Z +issue,168696,Test: TestConvolutionNNDeviceType.test_conv3d_cudnn_broken,https://github.com/pytorch/pytorch/issues/168696,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: rocm",scan when time allows,2026-01-03T16:19:45Z +issue,155992,[feature request] Native checkpointing to/from `s3://` (DCP / `torch.load` / `torch.save`),https://github.com/pytorch/pytorch/issues/155992,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: serialization, oncall: distributed checkpointing",scan when time allows,2026-01-02T11:54:25Z +issue,91965,Proxy/cache server option/hooks for downloading model checkpoints and dataset archive files in cloud environment,https://github.com/pytorch/pytorch/issues/91965,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2026-01-02T10:36:49Z +issue,140130,[onnx] [njt] [feature request] Export NJT-enabled SDPA / MHA ops to ORT's PackingMode Attention,https://github.com/pytorch/pytorch/issues/140130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-12-30T21:20:37Z +issue,167210,[RFC] Introduce the Unified Memory Pool and Pluggable Allocator in torch.accelerator,https://github.com/pytorch/pytorch/issues/167210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: accelerator,scan when time allows,2025-12-30T10:48:36Z +issue,167170,[XPU][Tracker] Enable `torch._scaled_matmul`,https://github.com/pytorch/pytorch/issues/167170,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xpu,scan when time allows,2025-12-30T04:59:10Z +issue,171179,"NCCL error: /pytorch/torch/csrc/distributed/c10d/NCCLUtils.hpp:268, NCCL version 2.22.3 ncclSystemError: System call (e.g. socket, malloc) or external library call failed or device error. Last error: socketStartConnect: Connect to IP<51423> failed",https://github.com/pytorch/pytorch/issues/171179,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-12-30T03:26:14Z +issue,120587,Type hints on nn.Module.__setattr__ value,https://github.com/pytorch/pytorch/issues/120587,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: typing",scan when time allows,2025-12-29T13:37:04Z +issue,11936,torch.bmm doesn't support CUDA uint8 (byte) tensor,https://github.com/pytorch/pytorch/issues/11936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: cuda, module: linear algebra",scan when time allows,2025-12-29T02:39:54Z +issue,169870,Capturing ViewAndMutationMeta for training graphs for PyTorch 2.8,https://github.com/pytorch/pytorch/issues/169870,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping, oncall: pt2",scan when time allows,2025-12-28T22:09:55Z +issue,86060,[MPS] Pin_memory on MPS backend raises exception of setting wrong device storage,https://github.com/pytorch/pytorch/issues/86060,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-12-28T18:36:21Z +issue,169191,dim broadcast in masked tensor between mask and data.,https://github.com/pytorch/pytorch/issues/169191,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2025-12-28T16:08:23Z +issue,112377,"RuntimeError: NVML_SUCCESS == r INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1695392020201/work/c10/cuda/CUDACachingAllocator.cpp"":1154, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/112377,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-12-28T11:09:49Z +issue,170496,torch.func.grad does not support FlashAttention autograd function,https://github.com/pytorch/pytorch/issues/170496,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2025-12-28T04:38:40Z +issue,163504,[MPS] Unexpected behavior in torch.nn.functional.one_hot when target values exceed num_classes in MPS,https://github.com/pytorch/pytorch/issues/163504,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",scan when time allows,2025-12-27T18:17:53Z +issue,168722,Test: TestCudaMallocAsync.test_device_memory_used,https://github.com/pytorch/pytorch/issues/168722,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-12-27T06:24:29Z +issue,1468,avg_pool functions hold input for backward,https://github.com/pytorch/pytorch/issues/1468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2025-12-26T22:36:28Z +issue,166853,torch.randn on CPU is not accurate with dtype=bfloat16,https://github.com/pytorch/pytorch/issues/166853,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: correctness (silent), module: random",scan when time allows,2025-12-26T11:59:28Z +issue,88810,Return the attention weights using the Transformer Encoder class.,https://github.com/pytorch/pytorch/issues/88810,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-12-26T05:27:02Z +issue,167762,Conan Support,https://github.com/pytorch/pytorch/issues/167762,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-12-26T04:21:15Z +issue,81876,dtype mismatch when after using auto mixed precision,https://github.com/pytorch/pytorch/issues/81876,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2025-12-26T02:46:41Z +issue,171184,No hash in CachingAutotuner meta information for some kernels,https://github.com/pytorch/pytorch/issues/171184,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-12-25T18:52:15Z +issue,69610,[Question] How to extract/expose the complete PyTorch computation graph (forward and backward)?,https://github.com/pytorch/pytorch/issues/69610,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: visualization",scan when time allows,2025-12-24T06:43:52Z +issue,168774,Test: TestMatmulCuda.test_cublas_addmm,https://github.com/pytorch/pytorch/issues/168774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-12-24T06:15:07Z +issue,63138,Sign in slogdet is set to requires_grad = False even when using complex numbers.,https://github.com/pytorch/pytorch/issues/63138,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: linear algebra",scan when time allows,2025-12-24T00:43:35Z +issue,77230,fast `gradcheck` fails when outputs that do not require grad precede outputs that do,https://github.com/pytorch/pytorch/issues/77230,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra",scan when time allows,2025-12-24T00:40:49Z +issue,167825,extra mem copy for value to assign when dispatching `index_put` to `masked_fill` on PrivateUse1 backend.,https://github.com/pytorch/pytorch/issues/167825,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: privateuse1",scan when time allows,2025-12-23T08:42:09Z +issue,171100,Feedback about torch.bernoulli,https://github.com/pytorch/pytorch/issues/171100,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-12-23T05:29:40Z +issue,164041,[vLLM] AssertionError: out= ops aren't yet supported,https://github.com/pytorch/pytorch/issues/164041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: vllm, oncall: pt2",scan when time allows,2025-12-22T21:56:05Z +issue,171039,"When using torch.multiprocessing for CUDA IPC, memory cannot be released.",https://github.com/pytorch/pytorch/issues/171039,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2025-12-22T17:46:19Z +issue,170936,NaN-checking anomaly mode should work with DTensor,https://github.com/pytorch/pytorch/issues/170936,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2025-12-22T17:41:37Z +issue,68966,`torch.ldexp` generated tests fail on call to `torch.mul`,https://github.com/pytorch/pytorch/issues/68966,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2025-12-22T14:09:32Z +issue,168807,Test: TestNNDeviceType.test_upsamplingNearest2d_launch_fail,https://github.com/pytorch/pytorch/issues/168807,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-12-22T02:06:30Z +issue,170674,Flex attention cannot support page size/block mask size != 128 for prefill.,https://github.com/pytorch/pytorch/issues/170674,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-12-20T23:38:01Z +issue,170851,Add forward_by_train and forward_by_eval convenience methods to nn.Module,https://github.com/pytorch/pytorch/issues/170851,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-12-20T22:55:45Z +issue,76750,Bug in dataloader iterator found by mypy,https://github.com/pytorch/pytorch/issues/76750,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-12-20T22:35:36Z +issue,80256,[complex] dropout and it's variants should support complex tensors,https://github.com/pytorch/pytorch/issues/80256,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nn",scan when time allows,2025-12-20T12:29:37Z +issue,170168,torch.linspace with integer dtype gives inconsistent results on CPU vs CUDA,https://github.com/pytorch/pytorch/issues/170168,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: python frontend",scan when time allows,2025-12-20T09:21:36Z +issue,170694,Proposal to simplify `DeviceThreadHandlePool`,https://github.com/pytorch/pytorch/issues/170694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-12-20T02:58:29Z +issue,168450,Test: DistTensorOpsTest.test_index,https://github.com/pytorch/pytorch/issues/168450,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-19T21:42:03Z +issue,170542,BuildExtension(no_python_abi_suffix=True).get_ext_filename result is incorrect when ABI part is not present in the filename.,https://github.com/pytorch/pytorch/issues/170542,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-12-19T21:32:38Z +issue,168448,Test: DistMatrixOpsTest.test_grouped_mm,https://github.com/pytorch/pytorch/issues/168448,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-19T21:21:22Z +issue,170495,vLLM Blackwell Fusion and Compile Tests failing PyTorch 2.10,https://github.com/pytorch/pytorch/issues/170495,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vllm,scan when time allows,2025-12-19T19:48:35Z +issue,169479,Discussion on addition Spin commands that would be useful,https://github.com/pytorch/pytorch/issues/169479,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-12-19T16:02:52Z +issue,88380,cuDNN error (CUDNN_STATUS_NOT_SUPPORTED) for torch.nn.functional.grid_sample(),https://github.com/pytorch/pytorch/issues/88380,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2025-12-19T09:28:53Z +issue,147274,PaddleOCR PyTorch conflicts,https://github.com/pytorch/pytorch/issues/147274,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: windows",scan when time allows,2025-12-19T07:08:21Z +issue,170284,selective activation checkpoint compatibility to async activation offload,https://github.com/pytorch/pytorch/issues/170284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2025-12-19T05:35:46Z +issue,169146,bad numerics for variable length attention with cudnn,https://github.com/pytorch/pytorch/issues/169146,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: sdpa",scan when time allows,2025-12-19T01:34:06Z +issue,168858,Test: TestTesting.test_cuda_assert_should_not_stop_common_distributed_test_suite,https://github.com/pytorch/pytorch/issues/168858,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-12-18T23:17:09Z +issue,170691,[Feature Request] Add StateSpaceModel to torch.nn,https://github.com/pytorch/pytorch/issues/170691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-12-18T15:50:59Z +issue,170458,must_in_dispatch takes a long time in torch registered ops,https://github.com/pytorch/pytorch/issues/170458,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: torchbind,scan when time allows,2025-12-18T15:39:31Z +issue,170426,argmax over multiple axis,https://github.com/pytorch/pytorch/issues/170426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-12-18T15:37:31Z +issue,169154,AOTIModelPackageLoader occupies too much disk space when launching multiple instances,https://github.com/pytorch/pytorch/issues/169154,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",scan when time allows,2025-12-18T02:26:48Z +issue,167526,Missing documentation for CUTLASS backend,https://github.com/pytorch/pytorch/issues/167526,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2025-12-17T15:25:44Z +issue,154606,`torch.cumprod` for complex128 produces inconsistent results between CPU and CUDA,https://github.com/pytorch/pytorch/issues/154606,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: edge cases, module: nans and infs",scan when time allows,2025-12-17T12:39:20Z +issue,166512,Enable GitHub Actions app to support ppc64le CI jobs using IBM-hosted runners,https://github.com/pytorch/pytorch/issues/166512,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: power",scan when time allows,2025-12-17T11:29:42Z +issue,163984,[inductor] Convert complex operations to real-block matrices,https://github.com/pytorch/pytorch/issues/163984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, oncall: pt2",scan when time allows,2025-12-17T10:58:06Z +issue,170402,Write more test cases for Conv2d to check for 0 steps,https://github.com/pytorch/pytorch/issues/170402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2025-12-17T08:29:25Z +issue,170536,accept int padding value for F.pad,https://github.com/pytorch/pytorch/issues/170536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: padding,scan when time allows,2025-12-17T05:27:50Z +issue,170350,[feature request] Checked version of indexing/embedding ops,https://github.com/pytorch/pytorch/issues/170350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: embedding, module: nn",scan when time allows,2025-12-17T05:09:46Z +issue,170382,Feedback about torch.nn.functional.interpolate,https://github.com/pytorch/pytorch/issues/170382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-12-17T04:40:17Z +issue,162606,Tensorpipe - ROCm support,https://github.com/pytorch/pytorch/issues/162606,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: tensorpipe",scan when time allows,2025-12-17T02:56:09Z +issue,76557,Pip packaging and publishing improvements in pytorch wheels for better integration with poetry,https://github.com/pytorch/pytorch/issues/76557,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-12-16T23:15:01Z +issue,164878,Ban and remove plain asserts with no message in our python code,https://github.com/pytorch/pytorch/issues/164878,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: python frontend",scan when time allows,2025-12-16T20:02:43Z +issue,52596,pickle is a security issue,https://github.com/pytorch/pytorch/issues/52596,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: hub, module: pickle, module: serialization",scan when time allows,2025-12-16T16:57:23Z +issue,124018,Fix OOMs during parallel builds,https://github.com/pytorch/pytorch/issues/124018,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-12-16T15:17:02Z +issue,170345,[vllm] Pytorch failure lazy_imports.py is failing on torch._inductor.async_compile for release 2.9.0 and 2.10.0 rc1,https://github.com/pytorch/pytorch/issues/170345,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: vllm, oncall: pt2",scan when time allows,2025-12-16T00:16:40Z +issue,144810,Support torch.func.grad for Flex Attention,https://github.com/pytorch/pytorch/issues/144810,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flex attention, module: functorch, module: higher order operators, module: pt2-dispatcher",scan when time allows,2025-12-15T23:21:58Z +issue,136623,The MPS Backend sometimes samples outside of distribution domain with `multinomial`,https://github.com/pytorch/pytorch/issues/136623,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-12-15T16:09:24Z +issue,170365,[CI][CUDA][GH100] LinAlg Unit Test Failures,https://github.com/pytorch/pytorch/issues/170365,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",scan when time allows,2025-12-15T16:05:11Z +issue,165141,"[ROCm][RDNA3/GFX1100] MIOpen Conv3d catastrophically fails on large batches with bfloat16, causing vLLM to extremely slowdown",https://github.com/pytorch/pytorch/issues/165141,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: rocm",scan when time allows,2025-12-15T12:51:29Z +issue,71379,Leaky cmake cuda compile options,https://github.com/pytorch/pytorch/issues/71379,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: cpp-extensions",scan when time allows,2025-12-15T10:59:19Z +issue,169816,"Cannot load or export TorchVision quantized models on Windows - ""Quantized backend not supported""",https://github.com/pytorch/pytorch/issues/169816,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: onnx, module: windows, oncall: quantization",scan when time allows,2025-12-15T03:35:43Z +issue,168863,Test: TestTorchDeviceType.test_cov,https://github.com/pytorch/pytorch/issues/168863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-12-12T19:08:25Z +issue,169078,Custom Non-Integer Indexing,https://github.com/pytorch/pytorch/issues/169078,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-12-12T18:11:31Z +issue,89438,Allow `low` and `high` to be tensors in `torch.randint`,https://github.com/pytorch/pytorch/issues/89438,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2025-12-12T17:51:11Z +issue,170266,pytorchbot does not react to Copilot PRs,https://github.com/pytorch/pytorch/issues/170266,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-12-12T16:23:51Z +issue,167447,Memory related APIs in torch.accelerator are not supported on MPS,https://github.com/pytorch/pytorch/issues/167447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: accelerator, module: mps",scan when time allows,2025-12-12T11:21:14Z +issue,169294,Conv2d CPU performance 2-3x slower on macOS due to missing MKLDNN backend,https://github.com/pytorch/pytorch/issues/169294,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: macos, module: performance",scan when time allows,2025-12-12T10:37:14Z +issue,119215,[c10d] Workaround the ncclCommAbort() hang issues,https://github.com/pytorch/pytorch/issues/119215,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-12-12T08:52:58Z +issue,81405,Implement shape/size functions for nestedtensor,https://github.com/pytorch/pytorch/issues/81405,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-12-12T05:01:01Z +issue,151066,[ROCm] QR decomposition is much slower on MI300x than A100,https://github.com/pytorch/pytorch/issues/151066,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: rocm",scan when time allows,2025-12-11T17:50:16Z +issue,170175,"[MPS][Inconsistency] nonzero() incorrectly preserves non-contiguous strides from sliced inputs, causing .view() crash on CPU/CUDA,but pass on MPS",https://github.com/pytorch/pytorch/issues/170175,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-12-11T14:38:11Z +issue,132624,Add support for float8 dtypes for the MPS backend,https://github.com/pytorch/pytorch/issues/132624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8), module: mps",scan when time allows,2025-12-11T08:47:37Z +issue,144445,NotImplementedError: Output channels > 65536 not supported at the MPS device.,https://github.com/pytorch/pytorch/issues/144445,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: mps",scan when time allows,2025-12-11T07:12:42Z +issue,168743,Test: TestCudaPrimaryCtx.test_set_device_0,https://github.com/pytorch/pytorch/issues/168743,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-12-11T06:18:30Z +issue,168800,Test: TestNN.test_batchnorm,https://github.com/pytorch/pytorch/issues/168800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-12-11T05:11:51Z +issue,136833,Thread safety issue with torch.compile(),https://github.com/pytorch/pytorch/issues/136833,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: inductor, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-12-11T00:57:34Z +issue,140322,torch_dispatch x HOP handling,https://github.com/pytorch/pytorch/issues/140322,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-12-10T19:28:29Z +issue,106388,Better export story for autograd.Function?,https://github.com/pytorch/pytorch/issues/106388,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-12-10T08:28:37Z +issue,133751,"`torch.nn.LSTM` using cuDNN fails with sequences longer than 65,535",https://github.com/pytorch/pytorch/issues/133751,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: cudnn, module: nn",scan when time allows,2025-12-10T01:47:50Z +issue,169955,[Bug] Potential command injection risk (shell=True) in subprocess,https://github.com/pytorch/pytorch/issues/169955,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-12-10T00:24:54Z +issue,169934,Autocast cache not updated by grad enabled state,https://github.com/pytorch/pytorch/issues/169934,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: autograd",scan when time allows,2025-12-09T19:08:40Z +issue,169909,[CI] cudaErrorInvalidValue in test_std_cuda_kernel_launch_check_error_show_cpp_stacktraces,https://github.com/pytorch/pytorch/issues/169909,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2025-12-09T19:01:18Z +issue,169956,`TestLinalgCUDA.test__int_mm` fails false/false case on A100,https://github.com/pytorch/pytorch/issues/169956,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: tests",scan when time allows,2025-12-09T18:53:42Z +issue,169797,"When augment_with_fx_traces=True but the user has misconfigured the FX config, raise an error",https://github.com/pytorch/pytorch/issues/169797,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: profiler",scan when time allows,2025-12-09T18:23:40Z +issue,116350,RuntimeError: derivative for aten::_scaled_dot_product_flash_attention_backward is not implemented,https://github.com/pytorch/pytorch/issues/116350,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-12-09T16:34:09Z +issue,169184,[RFC] Speed-up checkout operations in CI with treeless checkouts,https://github.com/pytorch/pytorch/issues/169184,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-12-09T16:23:41Z +issue,36577,Add load_state_dict and state_dict() in C++,https://github.com/pytorch/pytorch/issues/36577,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-12-09T14:20:14Z +issue,169738,"[MPS][Inductor] AdaptiveMaxPool{1,2}d produces incorrect results (numerical correctness)",https://github.com/pytorch/pytorch/issues/169738,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps, oncall: pt2",scan when time allows,2025-12-09T12:07:58Z +issue,77962,Operating on boolean torch tensor and numpy array casts to `unit8`,https://github.com/pytorch/pytorch/issues/77962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: numpy, module: type promotion",scan when time allows,2025-12-09T11:04:56Z +issue,120710,Reduce duplication in number of places where we specify dependency pin,https://github.com/pytorch/pytorch/issues/120710,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: devx",scan when time allows,2025-12-09T09:20:45Z +issue,169763,Multiple torch.load() call on the same BinaryIO,https://github.com/pytorch/pytorch/issues/169763,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-12-09T04:47:10Z +issue,169469,AOTI infinite recursion from a scalar buffer,https://github.com/pytorch/pytorch/issues/169469,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",scan when time allows,2025-12-09T01:57:41Z +issue,67602,[feature request] Add no-op set_epoch on Sampler class,https://github.com/pytorch/pytorch/issues/67602,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-12-09T01:51:09Z +issue,168977,Easier way to correlate DebugMode with FX graph,https://github.com/pytorch/pytorch/issues/168977,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-12-08T23:52:11Z +issue,157648,RuntimeError: operator torchvision::nms does not exist,https://github.com/pytorch/pytorch/issues/157648,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2025-12-08T13:34:20Z +issue,117582,Add forward AD support for nn.functional.grid_sample,https://github.com/pytorch/pytorch/issues/117582,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, module: nn",scan when time allows,2025-12-08T12:21:23Z +issue,162346,The batching rule of aten::affine_grid_generator is not implemented,https://github.com/pytorch/pytorch/issues/162346,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2025-12-08T03:43:52Z +issue,144850,Batching rule for aten::_thnn_fused_gru_cell,https://github.com/pytorch/pytorch/issues/144850,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-12-08T03:25:29Z +issue,149821,`Aborted` error when using `torch.cuda.memory.caching_allocator_delete`,https://github.com/pytorch/pytorch/issues/149821,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-12-06T09:14:35Z +issue,169675,Header-hiding when TORCH_TARGET_VERSION is set is broken for editable torch installs,https://github.com/pytorch/pytorch/issues/169675,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions",scan when time allows,2025-12-06T00:25:05Z +issue,102963,torch.svd fails on large matrices,https://github.com/pytorch/pytorch/issues/102963,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: linear algebra",scan when time allows,2025-12-05T11:09:55Z +issue,169290,[MPS][Inductor] LayerNorm/BatchNorm/GroupNorm fail to compile with dynamic=True due to invalid Metal codegen (VLA & Min),https://github.com/pytorch/pytorch/issues/169290,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, oncall: pt2",scan when time allows,2025-12-05T02:50:23Z +issue,161105,UED guard check failure,https://github.com/pytorch/pytorch/issues/161105,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-12-05T01:56:35Z +issue,164116,qta and export onnx error,https://github.com/pytorch/pytorch/issues/164116,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: quantization",scan when time allows,2025-12-05T01:53:34Z +issue,163052,CPU dynamic quantization (qint8) cannot be used with multiprocessing,https://github.com/pytorch/pytorch/issues/163052,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, oncall: quantization",scan when time allows,2025-12-05T00:53:35Z +issue,160651,`dequantize_per_channel` in `quantized_decomposed` not compatible with Float8 types,https://github.com/pytorch/pytorch/issues/160651,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8), oncall: quantization",scan when time allows,2025-12-05T00:53:05Z +issue,168529,Test: AsyncTPTest.test_optimal_layout,https://github.com/pytorch/pytorch/issues/168529,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:20:01Z +issue,168528,Test: AsyncTPTest.test_fused_scaled_matmul_reduce_scatter,https://github.com/pytorch/pytorch/issues/168528,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:19:56Z +issue,168527,Test: AsyncTPTest.test_fused_matmul_reduce_scatter,https://github.com/pytorch/pytorch/issues/168527,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:19:51Z +issue,168526,Test: AsyncTPTest.test_fused_all_gather_scaled_matmul,https://github.com/pytorch/pytorch/issues/168526,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:19:45Z +issue,168525,Test: AsyncTPTest.test_multimem_all_gather_matmul,https://github.com/pytorch/pytorch/issues/168525,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:19:41Z +issue,168524,Test: AsyncTPTest.test_fused_all_gather_matmul_native,https://github.com/pytorch/pytorch/issues/168524,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:19:36Z +issue,168523,Test: AsyncTPTest.test_fused_all_gather_matmul,https://github.com/pytorch/pytorch/issues/168523,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:19:31Z +issue,168522,TestModule: test/distributed/test_symmetric_memory.py::AsyncTPTest,https://github.com/pytorch/pytorch/issues/168522,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-12-04T21:19:26Z +issue,168460,Test: ProcessGroupGlooTest.test_block_current_stream_cuda,https://github.com/pytorch/pytorch/issues/168460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:17:52Z +issue,168459,Test: ProcessGroupGlooTest.test_reduce_stress_cuda,https://github.com/pytorch/pytorch/issues/168459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:17:45Z +issue,168458,Test: ProcessGroupGlooTest.test_allgather_stress_cuda,https://github.com/pytorch/pytorch/issues/168458,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:17:38Z +issue,168457,Test: ProcessGroupGlooTest.test_gather_stress_cuda,https://github.com/pytorch/pytorch/issues/168457,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:17:29Z +issue,168456,Test: ProcessGroupGlooTest.test_scatter_stress_cuda,https://github.com/pytorch/pytorch/issues/168456,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:17:22Z +issue,168455,Test: ProcessGroupGlooTest.test_allreduce_stress_cuda,https://github.com/pytorch/pytorch/issues/168455,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:17:16Z +issue,168454,Test: ProcessGroupGlooTest.test_broadcast_stress_cuda,https://github.com/pytorch/pytorch/issues/168454,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:17:07Z +issue,168453,TestModule: test/distributed/test_c10d_gloo.py::ProcessGroupGlooTest,https://github.com/pytorch/pytorch/issues/168453,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rocm",scan when time allows,2025-12-04T21:16:56Z +issue,168415,Test: TestShardedTensorChunked.test_multiple_local_shards,https://github.com/pytorch/pytorch/issues/168415,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T21:16:35Z +issue,168414,Test: TestShardedTensorChunked.test_new_group,https://github.com/pytorch/pytorch/issues/168414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T21:16:28Z +issue,168413,Test: TestShardedTensorChunked.test_partial_world_size,https://github.com/pytorch/pytorch/issues/168413,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T21:16:20Z +issue,168412,Test: TestShardedTensorChunked.test_complete_world_size,https://github.com/pytorch/pytorch/issues/168412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T21:16:10Z +issue,152308,bizarre behavior with torch module's Attribute Error,https://github.com/pytorch/pytorch/issues/152308,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-12-04T21:05:17Z +issue,168431,Test: TestShardedTensorFromLocalShards.test_init_from_local_shards_and_global_metadata,https://github.com/pytorch/pytorch/issues/168431,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:54:03Z +issue,168430,Test: TestShardedTensorFromLocalShards.test_init_from_local_shards_and_global_metadata_with_local_view,https://github.com/pytorch/pytorch/issues/168430,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:53:59Z +issue,168429,Test: TestShardedTensorFromLocalShards.test_init_from_local_shards_and_global_metadata_with_all_zeros,https://github.com/pytorch/pytorch/issues/168429,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:53:54Z +issue,168428,Test: TestShardedTensorFromLocalShards.test_non_rw_sharded_recalc_for_metadata,https://github.com/pytorch/pytorch/issues/168428,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:53:50Z +issue,168427,Test: TestShardedTensorFromLocalShards.test_init_from_local_shards_with_different_glb_size,https://github.com/pytorch/pytorch/issues/168427,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:53:46Z +issue,168426,Test: TestShardedTensorFromLocalShards.test_recalc_for_metadata,https://github.com/pytorch/pytorch/issues/168426,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:53:40Z +issue,168425,Test: TestShardedTensorFromLocalShards.test_init_from_local_shards,https://github.com/pytorch/pytorch/issues/168425,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:53:35Z +issue,168423,Test: TestShardedTensorFromLocalTensor.test_init_from_local_tensor,https://github.com/pytorch/pytorch/issues/168423,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:53:07Z +issue,168421,Test: TestShardedTensorEnumerable.test_with_rpc_names,https://github.com/pytorch/pytorch/issues/168421,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:52:45Z +issue,168420,Test: TestShardedTensorEnumerable.test_multiple_local_shards,https://github.com/pytorch/pytorch/issues/168420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:52:40Z +issue,168419,Test: TestShardedTensorEnumerable.test_new_group,https://github.com/pytorch/pytorch/issues/168419,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:52:36Z +issue,168418,Test: TestShardedTensorEnumerable.test_partial_world_size,https://github.com/pytorch/pytorch/issues/168418,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:52:33Z +issue,168417,Test: TestShardedTensorEnumerable.test_grid_sharding,https://github.com/pytorch/pytorch/issues/168417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:52:29Z +issue,168411,TestModule: test/distributed/_shard/sharded_tensor/test_sharded_tensor.py::TestShardedTensorChunked,https://github.com/pytorch/pytorch/issues/168411,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:49:30Z +issue,168424,TestModule: test/distributed/_shard/sharded_tensor/test_sharded_tensor.py::TestShardedTensorFromLocalShards,https://github.com/pytorch/pytorch/issues/168424,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:47:32Z +issue,168422,TestModule: test/distributed/_shard/sharded_tensor/test_sharded_tensor.py::TestShardedTensorFromLocalTensor,https://github.com/pytorch/pytorch/issues/168422,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:47:23Z +issue,168416,TestModule: test/distributed/_shard/sharded_tensor/test_sharded_tensor.py::TestShardedTensorEnumerable,https://github.com/pytorch/pytorch/issues/168416,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-12-04T20:47:06Z +issue,161954,[RFC] CUDA code decoupling and directory refactoring,https://github.com/pytorch/pytorch/issues/161954,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: privateuse1,scan when time allows,2025-12-04T20:07:21Z +issue,169429,Error loading `torch\lib\torch_python.dll` due to DLL initialization failure on Windows from torch `2.9.0`,https://github.com/pytorch/pytorch/issues/169429,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: windows",scan when time allows,2025-12-04T08:02:21Z +issue,169119,"`torch.compile(backend=""cudagraphs"", fullgraph=True, cuda)` fails for sparse CSR → COO (`torch.compile does not support sparse Tensors`)",https://github.com/pytorch/pytorch/issues/169119,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sparse, oncall: pt2",scan when time allows,2025-12-04T02:11:22Z +issue,169371,C++ Generator API is platform dependent,https://github.com/pytorch/pytorch/issues/169371,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: accelerator, module: cpp",scan when time allows,2025-12-04T02:07:26Z +issue,169390,[CI] [CUDA] SIGIOT error with test_cuda.py test_graph_capture_reclaim_4_streams,https://github.com/pytorch/pytorch/issues/169390,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2025-12-04T00:18:12Z +issue,169080,Processing & Tiling Massive Images,https://github.com/pytorch/pytorch/issues/169080,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2025-12-03T18:46:14Z +issue,166167,"Proposal: include 'extras packages' for supported versions of PyTorch-adjacent tools (e.g., numpy, onnx)",https://github.com/pytorch/pytorch/issues/166167,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-12-03T18:27:04Z +issue,169458,[download.pytorch.org] obselete dangling `xformers` index on CPU only branch,https://github.com/pytorch/pytorch/issues/169458,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2025-12-03T16:01:47Z +issue,168624,Test: MinifierIsolateTests.test_after_aot_gpu_runtime_error,https://github.com/pytorch/pytorch/issues/168624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: rocm, oncall: pt2",scan when time allows,2025-12-03T14:21:42Z +issue,169237,Large discrepancy between CPU and GPU norms for identical float32 tensor,https://github.com/pytorch/pytorch/issues/169237,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: norms and normalization",scan when time allows,2025-12-03T12:20:46Z +issue,169163,Silent failure of torch.stack,https://github.com/pytorch/pytorch/issues/169163,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: correctness (silent), module: cuda",scan when time allows,2025-12-03T12:20:22Z +issue,169266,Translation units with thrust::pair should include thrust/pair.h,https://github.com/pytorch/pytorch/issues/169266,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2025-12-03T12:19:52Z +issue,169378,[rocm triton 3.6] rocm-periodic-dynamo-benchmarks-test : convnextv2_nano.fcmae_ft_in22k_in1k fail_accuracy,https://github.com/pytorch/pytorch/issues/169378,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-12-03T12:17:42Z +issue,169133,vLLM tlparse sometimes missing compilation_metrics,https://github.com/pytorch/pytorch/issues/169133,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vllm,scan when time allows,2025-12-03T12:17:10Z +issue,132132,Tests stopped working after merging #126905 on Arm64,https://github.com/pytorch/pytorch/issues/132132,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: ci, module: tests, oncall: pt2",scan when time allows,2025-12-03T01:31:21Z +issue,121967,[PT2.2] PyTorch dispacthing non core ATen ops in .compile mode for torch ops.,https://github.com/pytorch/pytorch/issues/121967,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-12-02T23:39:31Z +issue,159013,"[RFC] Support slicing of submesh and more flexible operations to one device mesh such as reshape/split, etc.",https://github.com/pytorch/pytorch/issues/159013,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-12-02T17:47:19Z +issue,138280,"Refactor FlexibleLayout to separate out ""this stride can be changed"" and ""how this buffer is allocated can be changed""",https://github.com/pytorch/pytorch/issues/138280,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-12-02T17:11:15Z +issue,151934,`torch.bmm` is slow on non-contiguous BF16 CPU tensors,https://github.com/pytorch/pytorch/issues/151934,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cpu, module: half, module: performance",scan when time allows,2025-12-02T07:30:13Z +issue,133629,RuntimeError: free_upper_bound + pytorch_used_bytes[device],https://github.com/pytorch/pytorch/issues/133629,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-12-02T06:09:49Z +issue,168976,DebugMode doesn't track input/output relationships,https://github.com/pytorch/pytorch/issues/168976,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-12-02T04:10:43Z +issue,137767,[CPU]Tracking complex datatype issues,https://github.com/pytorch/pytorch/issues/137767,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cpu",scan when time allows,2025-12-02T01:40:00Z +issue,169213,CPU bad-free in `torch.nn.utils.rnn.pad_packed_sequence` / `unpack_sequence`,https://github.com/pytorch/pytorch/issues/169213,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2025-12-02T01:32:55Z +issue,169318,[debug mode] disable debug mode temporarily,https://github.com/pytorch/pytorch/issues/169318,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-12-02T01:29:09Z +issue,169152,[debug mode] support pipeline parallelism,https://github.com/pytorch/pytorch/issues/169152,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-12-01T22:47:55Z +issue,169055,`[feature request][discussion][cuBLASLt]` fused backward support for `addmm` with epilogue fusions.,https://github.com/pytorch/pytorch/issues/169055,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda",scan when time allows,2025-12-01T19:27:51Z +issue,109098,Can't initializa NVML,https://github.com/pytorch/pytorch/issues/109098,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-12-01T15:28:14Z +issue,35713,libtorch for Windows. MNIST example does no work.,https://github.com/pytorch/pytorch/issues/35713,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-12-01T03:42:46Z +issue,151110,Update epsilon logic to improve numerical stability,https://github.com/pytorch/pytorch/issues/151110,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: numerical-stability",scan when time allows,2025-11-30T18:15:48Z +issue,139065,Cannot compile with latest LLVM-19,https://github.com/pytorch/pytorch/issues/139065,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-11-30T13:46:11Z +issue,147317,Tensorboard `add_video()` broken for `moviepy>=2.0`,https://github.com/pytorch/pytorch/issues/147317,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2025-11-29T03:08:34Z +issue,97395,Expanded weights tests broken,https://github.com/pytorch/pytorch/issues/97395,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-11-28T09:41:26Z +issue,152731,Inconsistent float16 overflow behavior between CPU and CUDA devices,https://github.com/pytorch/pytorch/issues/152731,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: half",scan when time allows,2025-11-28T07:17:51Z +issue,168967,implement __str__ for Schedular classes,https://github.com/pytorch/pytorch/issues/168967,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-11-27T20:53:45Z +issue,169170,Optimizer __repr__ Performance & Stability Improvement,https://github.com/pytorch/pytorch/issues/169170,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-11-27T20:29:19Z +issue,169072,Properly fix destructive hardware interference warnings,https://github.com/pytorch/pytorch/issues/169072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-11-27T20:11:45Z +issue,15821,No test coverage for kwargs of AvgPool2d and AvgPool3d,https://github.com/pytorch/pytorch/issues/15821,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: nn, module: pooling",scan when time allows,2025-11-27T17:48:47Z +issue,78656,Allow batch_norm_backward_elemt and batch_norm_gather_stats_with_counts handle 0 counts,https://github.com/pytorch/pytorch/issues/78656,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2025-11-27T10:33:43Z +issue,166131,Return a view from tensor(requires_grad=False) in autograd function may cause incorrect requires_grad attribute.,https://github.com/pytorch/pytorch/issues/166131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2025-11-27T09:06:44Z +issue,163400,"Add Flash Dynamic Mask Attention (FDMA) as an SDPA backend for sparse, extremely long-context attention",https://github.com/pytorch/pytorch/issues/163400,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sdpa, module: sparse",scan when time allows,2025-11-27T05:06:00Z +issue,165612,RFC: Optionally accept NumPy dtypes in all APIs where torch dtypes are accepted,https://github.com/pytorch/pytorch/issues/165612,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8), module: python frontend",scan when time allows,2025-11-27T00:42:21Z +issue,166378,Multihead Attention's in_proj_weight is initialized inconsistently,https://github.com/pytorch/pytorch/issues/166378,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sdpa",scan when time allows,2025-11-26T22:59:57Z +issue,65151,Registering a global fallback for all operators that defaults us to assuming that autograd is not implemented,https://github.com/pytorch/pytorch/issues/65151,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: bc-breaking, module: custom-operators, module: molly-guard",scan when time allows,2025-11-26T22:32:04Z +issue,167721,"Minimal, comprehensive test suite",https://github.com/pytorch/pytorch/issues/167721,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: infra, module: testing",scan when time allows,2025-11-26T21:55:31Z +issue,49284,TensorBoard SummaryWriter to remote storage unreasonably slow,https://github.com/pytorch/pytorch/issues/49284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2025-11-26T18:29:36Z +issue,167656,Lack of tensor identity tracking in DebugMode makes it difficult to interpret results,https://github.com/pytorch/pytorch/issues/167656,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: pt2",scan when time allows,2025-11-26T15:03:13Z +issue,168403,Test: TestFullyShardOverlap.test_fully_shard_training_overlap,https://github.com/pytorch/pytorch/issues/168403,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T19:58:08Z +issue,168405,Test: TestFullyShardNDTraining.test_2d_mlp_with_nd_mesh,https://github.com/pytorch/pytorch/issues/168405,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T19:58:07Z +issue,168407,Test: TestFullyShard2DTraining.test_train_parity_2d_mlp,https://github.com/pytorch/pytorch/issues/168407,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T19:58:03Z +issue,168476,Test: ComposabilityTest.test_pp_fsdp,https://github.com/pytorch/pytorch/issues/168476,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T19:53:13Z +issue,168452,Test: TestComputeCommReorderingMultiProc.test_grouped_scheduler_node,https://github.com/pytorch/pytorch/issues/168452,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-11-25T19:52:39Z +issue,168477,Test: ComposabilityTest.test_pp_fsdp_unshard_reshard_runtime,https://github.com/pytorch/pytorch/issues/168477,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T19:51:55Z +issue,168537,Test: LoweringTest.test_lowering_one_shot_all_reduce,https://github.com/pytorch/pytorch/issues/168537,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T19:49:20Z +issue,168568,Test: AOTInductorTestsTemplate.test__weight_int4pack_mm_with_scales_and_zeros,https://github.com/pytorch/pytorch/issues/168568,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",scan when time allows,2025-11-25T19:46:09Z +issue,168567,Test: AOTInductorTestsTemplate.test__weight_int4pack_mm,https://github.com/pytorch/pytorch/issues/168567,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",scan when time allows,2025-11-25T19:46:08Z +issue,168563,TestModule: test/inductor/test_aot_inductor.py::AOTInductorTestsTemplate,https://github.com/pytorch/pytorch/issues/168563,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",scan when time allows,2025-11-25T19:46:03Z +issue,168623,TestModule: test/inductor/test_minifier_isolate.py::MinifierIsolateTests,https://github.com/pytorch/pytorch/issues/168623,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: rocm, oncall: pt2",scan when time allows,2025-11-25T19:43:20Z +issue,168646,Test: MultiKernelTest.test_triton_relu_fused_gemm,https://github.com/pytorch/pytorch/issues/168646,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: rocm, oncall: pt2",scan when time allows,2025-11-25T19:42:26Z +issue,168645,Test: MultiKernelTest.test_triton_gemm,https://github.com/pytorch/pytorch/issues/168645,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: rocm, oncall: pt2",scan when time allows,2025-11-25T19:42:25Z +issue,168644,TestModule: test/inductor/test_multi_kernel.py::MultiKernelTest,https://github.com/pytorch/pytorch/issues/168644,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: rocm, oncall: pt2",scan when time allows,2025-11-25T19:42:24Z +issue,168651,TestModule: test/inductor/test_select_algorithm.py::TestSelectAlgorithm,https://github.com/pytorch/pytorch/issues/168651,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: rocm, oncall: pt2",scan when time allows,2025-11-25T19:42:06Z +issue,168692,TestModule: test/nn/test_convolution.py::TestConvolutionNNDeviceType,https://github.com/pytorch/pytorch/issues/168692,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: rocm",scan when time allows,2025-11-25T19:37:37Z +issue,168706,TestModule: test/test_cpp_extensions_aot.py::TestCppExtensionAOT,https://github.com/pytorch/pytorch/issues/168706,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",scan when time allows,2025-11-25T19:37:06Z +issue,168707,Test: TestCppExtensionAOT.test_cublas_extension,https://github.com/pytorch/pytorch/issues/168707,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",scan when time allows,2025-11-25T19:37:00Z +issue,168708,Test: TestCppExtensionAOT.test_cusolver_extension,https://github.com/pytorch/pytorch/issues/168708,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",scan when time allows,2025-11-25T19:36:58Z +issue,168709,Test: TestCppExtensionAOT.test_cuda_dlink_libs,https://github.com/pytorch/pytorch/issues/168709,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: rocm, oncall: export, oncall: pt2",scan when time allows,2025-11-25T19:36:55Z +issue,168819,TestModule: test/test_scaled_matmul_cuda.py::TestFP8Matmul,https://github.com/pytorch/pytorch/issues/168819,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8), module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:34:06Z +issue,168756,TestModule: test/test_linalg.py::TestLinalg,https://github.com/pytorch/pytorch/issues/168756,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:31:24Z +issue,168766,Test: TestLinalg.test__dyn_quant_pack_4bit_weight,https://github.com/pytorch/pytorch/issues/168766,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:31:11Z +issue,168767,Test: TestLinalg.test__dyn_quant_matmul_4bit,https://github.com/pytorch/pytorch/issues/168767,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:31:10Z +issue,168768,Test: TestLinalg.test_compile_dyn_quant_matmul_4bit,https://github.com/pytorch/pytorch/issues/168768,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:31:06Z +issue,168771,Test: TestLinalg.test_ldl_solve,https://github.com/pytorch/pytorch/issues/168771,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:55Z +issue,168772,Test: TestLinalg.test_ck_blas_library,https://github.com/pytorch/pytorch/issues/168772,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:52Z +issue,168773,TestModule: test/test_matmul_cuda.py::TestMatmulCuda,https://github.com/pytorch/pytorch/issues/168773,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:44Z +issue,168777,Test: TestMatmulCuda.test_grouped_gemm_compiled,https://github.com/pytorch/pytorch/issues/168777,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:42Z +issue,168778,Test: TestMatmulCuda.test_mm_bmm_dtype_overload,https://github.com/pytorch/pytorch/issues/168778,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:39Z +issue,168776,Test: TestMatmulCuda.test_cublas_batch_invariance_blackwell,https://github.com/pytorch/pytorch/issues/168776,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:39Z +issue,168780,TestModule: test/test_matmul_cuda.py::TestMixedDtypesLinearCuda,https://github.com/pytorch/pytorch/issues/168780,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:22Z +issue,168781,Test: TestMixedDtypesLinearCuda.test_mixed_dtypes_linear,https://github.com/pytorch/pytorch/issues/168781,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: rocm",scan when time allows,2025-11-25T19:30:17Z +issue,168831,TestModule: test/test_sparse_semi_structured.py::SparseSemiStructuredTensorCompileTest,https://github.com/pytorch/pytorch/issues/168831,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:52Z +issue,168832,Test: SparseSemiStructuredTensorCompileTest.test_mlp_contiguous_relu_compile_cusparselt,https://github.com/pytorch/pytorch/issues/168832,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:51Z +issue,168833,Test: SparseSemiStructuredTensorCompileTest.test_mlp_contiguous_relu_compile_cutlass,https://github.com/pytorch/pytorch/issues/168833,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:50Z +issue,168834,Test: SparseSemiStructuredTensorCompileTest.test_sp24_compile,https://github.com/pytorch/pytorch/issues/168834,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:49Z +issue,168835,TestModule: test/test_sparse_semi_structured.py::TestSparseSemiStructuredTraining,https://github.com/pytorch/pytorch/issues/168835,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:45Z +issue,168836,Test: TestSparseSemiStructuredTraining.test_prune_dense_static_sort,https://github.com/pytorch/pytorch/issues/168836,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:36Z +issue,168837,Test: TestSparseSemiStructuredTraining.test_pruning_algo_largest_abs_values_greedy,https://github.com/pytorch/pytorch/issues/168837,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:35Z +issue,168838,Test: TestSparseSemiStructuredTraining.test_pack_both_ways_meta_correctness,https://github.com/pytorch/pytorch/issues/168838,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:27Z +issue,168840,Test: TestSparseSemiStructuredTraining.test_pack_both_ways_edge_case1,https://github.com/pytorch/pytorch/issues/168840,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:26Z +issue,168839,Test: TestSparseSemiStructuredTraining.test_pack_both_ways_id,https://github.com/pytorch/pytorch/issues/168839,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:26Z +issue,168842,Test: TestSparseSemiStructuredTraining.test_sp24_apply_dense,https://github.com/pytorch/pytorch/issues/168842,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:25Z +issue,168841,Test: TestSparseSemiStructuredTraining.test_sp24_apply,https://github.com/pytorch/pytorch/issues/168841,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:25Z +issue,168843,Test: TestSparseSemiStructuredTraining.test_sp24_matmuls,https://github.com/pytorch/pytorch/issues/168843,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:24Z +issue,168844,Test: TestSparseSemiStructuredTraining.test_sp24_matmuls_mat_vec,https://github.com/pytorch/pytorch/issues/168844,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:23Z +issue,168845,Test: TestSparseSemiStructuredTraining.test_sp24_matmuls_bmm,https://github.com/pytorch/pytorch/issues/168845,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:20Z +issue,168846,TestModule: test/test_sparse_semi_structured.py::TestSparseSemiStructuredCUTLASS,https://github.com/pytorch/pytorch/issues/168846,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:19Z +issue,168848,Test: TestSparseSemiStructuredCUTLASS.test_sparse_semi_structured_ops_cutlass,https://github.com/pytorch/pytorch/issues/168848,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:16Z +issue,168847,Test: TestSparseSemiStructuredCUTLASS.test_linear_cutlass,https://github.com/pytorch/pytorch/issues/168847,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:16Z +issue,168850,Test: TestSparseSemiStructuredCUTLASS.test_conversions_all_patterns,https://github.com/pytorch/pytorch/issues/168850,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:15Z +issue,168849,Test: TestSparseSemiStructuredCUTLASS.test_conversions,https://github.com/pytorch/pytorch/issues/168849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:22:15Z +issue,168851,TestModule: test/test_sparse_semi_structured.py::TestSparseSemiStructuredCUSPARSELT,https://github.com/pytorch/pytorch/issues/168851,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:21:35Z +issue,168852,Test: TestSparseSemiStructuredCUSPARSELT.test_cslt_sparse_mm_alpha,https://github.com/pytorch/pytorch/issues/168852,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:21:32Z +issue,168853,Test: TestSparseSemiStructuredCUSPARSELT.test_cslt_sparse_mm_alpha_compile_autotune,https://github.com/pytorch/pytorch/issues/168853,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:21:31Z +issue,168854,Test: TestSparseSemiStructuredCUSPARSELT.test_cslt_sparse_mm_alpha_mixed_dtype,https://github.com/pytorch/pytorch/issues/168854,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sparse",scan when time allows,2025-11-25T19:21:29Z +issue,168446,Test: RingAttentionTest.test_ring_attention_sdpa,https://github.com/pytorch/pytorch/issues/168446,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sdpa",scan when time allows,2025-11-25T19:17:40Z +issue,168445,TestModule: test/distributed/tensor/test_attention.py::RingAttentionTest,https://github.com/pytorch/pytorch/issues/168445,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm, module: sdpa",scan when time allows,2025-11-25T19:17:37Z +issue,168788,Test: TestNestedTensorSubclass.test_sdpa_backwards,https://github.com/pytorch/pytorch/issues/168788,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sdpa",scan when time allows,2025-11-25T19:15:05Z +issue,168789,Test: TestNestedTensorSubclass.test_sdpa_autocast,https://github.com/pytorch/pytorch/issues/168789,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sdpa",scan when time allows,2025-11-25T19:15:04Z +issue,168876,Test: TestSDPACudaOnly.test_fused_kernels_nested_broadcasting,https://github.com/pytorch/pytorch/issues/168876,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sdpa",scan when time allows,2025-11-25T19:13:02Z +issue,168875,Test: TestSDPACudaOnly.test_fused_backwards_throws_determinism_warning,https://github.com/pytorch/pytorch/issues/168875,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sdpa",scan when time allows,2025-11-25T19:13:02Z +issue,168877,Test: TestSDPACudaOnly.test_fused_kernels_nested_broadcasting_query_dense,https://github.com/pytorch/pytorch/issues/168877,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: sdpa",scan when time allows,2025-11-25T19:13:01Z +issue,168481,Test: NCCLSymmetricMemoryTest.test_nccl_symmem_alloc,https://github.com/pytorch/pytorch/issues/168481,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:52:28Z +issue,168533,Test: SymmMemNegativeTest.test_barrier_timeout,https://github.com/pytorch/pytorch/issues/168533,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:52:18Z +issue,168534,Test: SymmMemNegativeTest.test_put_signal_timeout,https://github.com/pytorch/pytorch/issues/168534,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:52:17Z +issue,168535,Test: SymmMemNegativeTest.test_wait_signal_timeout,https://github.com/pytorch/pytorch/issues/168535,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:52:15Z +issue,168480,TestModule: test/distributed/test_nccl.py::NCCLSymmetricMemoryTest,https://github.com/pytorch/pytorch/issues/168480,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:45:58Z +issue,168447,TestModule: test/distributed/tensor/test_matrix_ops.py::DistMatrixOpsTest,https://github.com/pytorch/pytorch/issues/168447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-11-25T18:40:20Z +issue,168449,TestModule: test/distributed/tensor/test_tensor_ops.py::DistTensorOpsTest,https://github.com/pytorch/pytorch/issues/168449,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-11-25T18:40:04Z +issue,168451,TestModule: test/distributed/test_aten_comm_compute_reordering.py::TestComputeCommReorderingMultiProc,https://github.com/pytorch/pytorch/issues/168451,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-11-25T18:40:00Z +issue,168475,TestModule: test/distributed/test_composability.py::ComposabilityTest,https://github.com/pytorch/pytorch/issues/168475,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-11-25T18:39:32Z +issue,168532,TestModule: test/distributed/test_symmetric_memory.py::SymmMemNegativeTest,https://github.com/pytorch/pytorch/issues/168532,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:36:21Z +issue,168536,TestModule: test/distributed/test_symmetric_memory.py::LoweringTest,https://github.com/pytorch/pytorch/issues/168536,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:36:17Z +issue,168503,TestModule: test/distributed/test_nvshmem_triton.py::NVSHMEMTritonTest,https://github.com/pytorch/pytorch/issues/168503,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:33:45Z +issue,168515,Test: NVSHMEMTritonTest.test_triton_alltoall,https://github.com/pytorch/pytorch/issues/168515,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:33:35Z +issue,168516,Test: NVSHMEMTritonTest.test_triton_broadcast,https://github.com/pytorch/pytorch/issues/168516,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:33:34Z +issue,168518,Test: NVSHMEMTritonTest.test_triton_minmax_reduce,https://github.com/pytorch/pytorch/issues/168518,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:33:33Z +issue,168517,Test: NVSHMEMTritonTest.test_triton_sum_reduce,https://github.com/pytorch/pytorch/issues/168517,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:33:33Z +issue,168519,Test: NVSHMEMTritonTest.test_triton_prod_reduce,https://github.com/pytorch/pytorch/issues/168519,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, module: symm_mem",scan when time allows,2025-11-25T18:33:26Z +issue,153187,Inconsistent overflow handling in `torch.clamp_min` between CPU and CUDA for float16 tensors,https://github.com/pytorch/pytorch/issues/153187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2025-11-25T18:24:43Z +issue,168435,Test: DistributedUtilTest.test_port_already_in_use_on_worker,https://github.com/pytorch/pytorch/issues/168435,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-11-25T18:22:27Z +issue,168434,TestModule: test/distributed/elastic/utils/distributed_test.py::DistributedUtilTest,https://github.com/pytorch/pytorch/issues/168434,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rocm",scan when time allows,2025-11-25T18:22:18Z +issue,168406,TestModule: test/distributed/_composable/test_composability/test_2d_composability.py::TestFullyShard2DTraining,https://github.com/pytorch/pytorch/issues/168406,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T18:20:16Z +issue,168404,TestModule: test/distributed/_composable/fsdp/test_fully_shard_training.py::TestFullyShardNDTraining,https://github.com/pytorch/pytorch/issues/168404,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T01:03:31Z +issue,168402,TestModule: test/distributed/_composable/fsdp/test_fully_shard_overlap.py::TestFullyShardOverlap,https://github.com/pytorch/pytorch/issues/168402,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: rocm",scan when time allows,2025-11-25T01:03:19Z +issue,168861,TestModule: test/test_torch.py::TestTorchDeviceType,https://github.com/pytorch/pytorch/issues/168861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:51:42Z +issue,168855,TestModule: test/test_testing.py::TestTesting,https://github.com/pytorch/pytorch/issues/168855,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:51:34Z +issue,168806,Test: TestNNDeviceType.test_upsampling_64bit_indexing_channels_last,https://github.com/pytorch/pytorch/issues/168806,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:50:29Z +issue,168801,TestModule: test/test_nn.py::TestNNDeviceType,https://github.com/pytorch/pytorch/issues/168801,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:50:21Z +issue,168798,TestModule: test/test_nn.py::TestNN,https://github.com/pytorch/pytorch/issues/168798,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:50:17Z +issue,168792,Test: TestNestedTensorSubclass.test_compile_preserves_metadata_cache,https://github.com/pytorch/pytorch/issues/168792,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:50:08Z +issue,168791,Test: TestNestedTensorSubclass.test_dummy_mha_with_nt,https://github.com/pytorch/pytorch/issues/168791,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:50:07Z +issue,168784,TestModule: test/test_nestedtensor.py::TestNestedTensorSubclass,https://github.com/pytorch/pytorch/issues/168784,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:49:57Z +issue,168742,TestModule: test/test_cuda_primary_ctx.py::TestCudaPrimaryCtx,https://github.com/pytorch/pytorch/issues/168742,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:48:58Z +issue,168721,Test: TestCudaMallocAsync.test_memory_profiler_viz,https://github.com/pytorch/pytorch/issues/168721,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:48:29Z +issue,168720,TestModule: test/test_cuda.py::TestCudaMallocAsync,https://github.com/pytorch/pytorch/issues/168720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-25T00:48:28Z +issue,149016,DISABLED test_var_mean_tile_reduction_True_dynamic_shapes_cuda (__main__.DynamicShapesGPUTests),https://github.com/pytorch/pytorch/issues/149016,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: inductor, module: rocm, oncall: pt2",scan when time allows,2025-11-25T00:38:52Z +issue,103495,DISABLED test_mem_get_info (__main__.TestCuda),https://github.com/pytorch/pytorch/issues/103495,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: flaky-tests, module: rocm",scan when time allows,2025-11-25T00:37:55Z +issue,167970,Thread Sanitizer support is clunky right now.,https://github.com/pytorch/pytorch/issues/167970,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: sanitizers",scan when time allows,2025-11-24T18:56:32Z +issue,149990,[Async TP] Fuse all-gather-matmuls for float8 rowwise training,https://github.com/pytorch/pytorch/issues/149990,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-11-24T18:35:21Z +issue,148748,export deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B failed,https://github.com/pytorch/pytorch/issues/148748,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-11-24T17:22:52Z +issue,166409,`torch.polygamma` produces inconsistent results between CPU and CUDA for moderate orders,https://github.com/pytorch/pytorch/issues/166409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nans and infs",scan when time allows,2025-11-24T15:37:25Z +issue,168304,Finish pytree flattening path in create_native_op_schema,https://github.com/pytorch/pytorch/issues/168304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2025-11-24T13:40:21Z +issue,119968,"Error on trying to run detectron2 on Mac M1 with device set to ""mps"" (Metal Performance Shaders)",https://github.com/pytorch/pytorch/issues/119968,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-11-24T13:12:38Z +issue,125718,`torch.compile` and complex numbers,https://github.com/pytorch/pytorch/issues/125718,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: dynamo, oncall: pt2",scan when time allows,2025-11-24T09:00:00Z +issue,168158,A more efficient alternative to the Tensor.record_stream(...) api is needed,https://github.com/pytorch/pytorch/issues/168158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-11-24T07:50:02Z +issue,162489,BatchNorm1d exhibits NaN outputs on GPU due to extreme input values causing numerical instability.,https://github.com/pytorch/pytorch/issues/162489,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nn",scan when time allows,2025-11-24T05:28:09Z +issue,166234,"!handles_.at(i) INTERNAL ASSERT FAILED at ""/pytorch/c10/cuda/CUDACachingAllocator.cpp"":430",https://github.com/pytorch/pytorch/issues/166234,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: windows, module: wsl",scan when time allows,2025-11-23T19:59:51Z +issue,168907,make lint fails on Mac,https://github.com/pytorch/pytorch/issues/168907,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: macos",scan when time allows,2025-11-23T19:01:05Z +issue,150989,Add `pad_to_multiple_of` to `torch.nn.utils.rnn.pad_sequence`,https://github.com/pytorch/pytorch/issues/150989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-11-22T17:20:41Z +issue,168253,nestedtensor inconsistency in `torch.masked_select`,https://github.com/pytorch/pytorch/issues/168253,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-11-21T17:47:04Z +issue,122030,[MPS] Wrong Calculation on MPS,https://github.com/pytorch/pytorch/issues/122030,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-11-21T17:34:58Z +issue,166651,Make `_grouped_mm` public and add out_dtype,https://github.com/pytorch/pytorch/issues/166651,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2025-11-21T01:05:34Z +issue,167979,"Jupyter kernel restart during eval after first epoch (MNIST, CPU and CUDA, PyTorch 2.10.0.dev20251114+cu128 on Windows 11 + RTX 5080)",https://github.com/pytorch/pytorch/issues/167979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: openmp, module: windows",scan when time allows,2025-11-20T18:17:04Z +issue,165999,[RFC] Improve the performance of `associative_scan()` with `cuda.compute`,https://github.com/pytorch/pytorch/issues/165999,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-11-20T14:37:30Z +issue,161302,Support for nn.SyncBatchNorm in torch.compile,https://github.com/pytorch/pytorch/issues/161302,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2",scan when time allows,2025-11-19T23:55:45Z +issue,85604,"Multiple GPUs get ""errno: 98 - Address already in use""",https://github.com/pytorch/pytorch/issues/85604,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-11-19T23:04:35Z +issue,164493,[CI][CUDA][SymmMem] B200 Periodical Job Failure with test_triton_sum_reduce_int64,https://github.com/pytorch/pytorch/issues/164493,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",scan when time allows,2025-11-19T18:29:40Z +issue,167950,Insufficient documentation about the batching logic of `torch.linalg.solve`,https://github.com/pytorch/pytorch/issues/167950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: linear algebra",scan when time allows,2025-11-19T16:44:07Z +issue,139124,shift operators not supporting integral types,https://github.com/pytorch/pytorch/issues/139124,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: unsigned int,scan when time allows,2025-11-19T06:52:30Z +issue,168128,"[proposal] Make docs webpage layout leaner, reduce number of menu/nav bars",https://github.com/pytorch/pytorch/issues/168128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-11-19T04:54:20Z +issue,16706,Very poor Uniform() sampling near floating 0.0,https://github.com/pytorch/pytorch/issues/16706,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2025-11-19T00:11:12Z +issue,168010,Misleading error message for device mismatch in `conv2d`,https://github.com/pytorch/pytorch/issues/168010,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-11-18T21:58:44Z +issue,105448,Will nn.unfold support non-4D-tensor input in future version?,https://github.com/pytorch/pytorch/issues/105448,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-11-18T19:36:46Z +issue,167645,`torch.export` incompatible with `parametrize.cached()` - model 25x slower than expected,https://github.com/pytorch/pytorch/issues/167645,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-11-18T19:18:50Z +issue,167895,stable-diffusion 2 gone from huggingface,https://github.com/pytorch/pytorch/issues/167895,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-11-18T19:10:39Z +issue,167063,`torch.onnx.export` with `dynamo` breaks `FakeQuantize` exporting,https://github.com/pytorch/pytorch/issues/167063,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-11-18T14:43:06Z +issue,156792,backprop fails in HF SegformerForSemanticSegmentation on MPS backend with torch > 2.4,https://github.com/pytorch/pytorch/issues/156792,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-11-17T19:20:02Z +issue,150517,Pinned memory doubles memory usage for tensors slightly over 128MB,https://github.com/pytorch/pytorch/issues/150517,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2025-11-17T17:30:47Z +issue,145837,torch.nested.narrow() or torch.nested.to_padded_tensor() breaks backwards pass - invalid gradient,https://github.com/pytorch/pytorch/issues/145837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-11-17T13:41:41Z +issue,153536,`torch.utils.data.default_collate` raises misleading warning for read-only NumPy arrays,https://github.com/pytorch/pytorch/issues/153536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2025-11-17T04:44:10Z +issue,43259,RuntimeError: Expected to have finished reduction in the prior iteration before starting a new one.,https://github.com/pytorch/pytorch/issues/43259,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2025-11-16T22:14:14Z +issue,162591,"SequentialLR scheduler, bug with last_epoch = N",https://github.com/pytorch/pytorch/issues/162591,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2025-11-16T17:46:25Z +issue,165906,"`torch.bmm` returns wrong results for large tensors with torch.set_float32_matmul_precision(""medium"")",https://github.com/pytorch/pytorch/issues/165906,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2025-11-15T23:10:20Z +issue,167724,CUDA error: invalid configuration argument for tensor operation,https://github.com/pytorch/pytorch/issues/167724,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-11-15T03:56:00Z +issue,167765,[RFC] Extend MemPool support to SymmMem CUDA backend,https://github.com/pytorch/pytorch/issues/167765,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: symm_mem,scan when time allows,2025-11-15T03:55:04Z +issue,167754,Getting fallback when using scaled_dot_product_attention() despite CUDA kernels being installed and available,https://github.com/pytorch/pytorch/issues/167754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-11-15T03:54:51Z +issue,167283,[RFC] Pytorch compute platform quality levels proposal,https://github.com/pytorch/pytorch/issues/167283,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: accelerator, module: backend",scan when time allows,2025-11-14T22:28:35Z +issue,122250,Proposal to make `torch.nn.modules.utils._pair` public,https://github.com/pytorch/pytorch/issues/122250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-11-14T18:33:55Z +issue,146371,Torchrun does not handle worker failure gracefully,https://github.com/pytorch/pytorch/issues/146371,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-11-14T04:31:31Z +issue,167746,[xpu] Balance xpu ci test jobs workload,https://github.com/pytorch/pytorch/issues/167746,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: xpu",scan when time allows,2025-11-13T22:33:25Z +issue,152406,[DTensor] Calling .item() on DTensor with Partial placement results in local value,https://github.com/pytorch/pytorch/issues/152406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-11-13T19:35:24Z +issue,124686,Unbacked SymInts: Should backwards graph with unbacked SymInts be recompiled with hints,https://github.com/pytorch/pytorch/issues/124686,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-11-13T06:41:14Z +issue,167670,PyreFly results are dependent on local python version,https://github.com/pytorch/pytorch/issues/167670,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: lint",scan when time allows,2025-11-12T22:37:06Z +issue,117762,PyEval_AcquireThread can crash process if called from another thread at interpreter shutdown,https://github.com/pytorch/pytorch/issues/117762,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-11-12T16:26:29Z +issue,162592,Multi-Batching in F.scaled_dot_product_attention does not work.,https://github.com/pytorch/pytorch/issues/162592,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-11-12T10:36:10Z +issue,117949,aten::slow_conv3d_forward still missing from MPS,https://github.com/pytorch/pytorch/issues/117949,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-11-12T06:57:07Z +issue,167558,per_page=1000 doesn't work in hud.pytorch.org,https://github.com/pytorch/pytorch/issues/167558,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-11-11T19:30:25Z +issue,102334,There is a memory leak in torch.load,https://github.com/pytorch/pytorch/issues/102334,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: serialization",scan when time allows,2025-11-11T18:52:27Z +issue,167349,AOTAutograd cache bypasses on DTensor.to_local(),https://github.com/pytorch/pytorch/issues/167349,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: pt2",scan when time allows,2025-11-11T18:30:10Z +issue,167027,combine compiled vectorized function without recompiling already compiled part,https://github.com/pytorch/pytorch/issues/167027,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-11-11T18:15:06Z +issue,167254,Fine control of fp32 precison in backward phase,https://github.com/pytorch/pytorch/issues/167254,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2025-11-11T18:00:37Z +issue,166646,Allow kwargs in torch.utils.checkpoint.checkpoint().,https://github.com/pytorch/pytorch/issues/166646,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2025-11-11T15:27:56Z +issue,156701,Add generator argument to torch.rand_like,https://github.com/pytorch/pytorch/issues/156701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-11-11T09:07:50Z +issue,167520,improve selective decomposition tracing time,https://github.com/pytorch/pytorch/issues/167520,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-11-11T06:34:47Z +issue,167420,conv3d with dilation and bias = severe performance degradation in bf16/fp16,https://github.com/pytorch/pytorch/issues/167420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: performance",scan when time allows,2025-11-10T23:34:05Z +issue,109774,[DDP + Dynamo] Tracing DDP AllReduce (Compiled DDP),https://github.com/pytorch/pytorch/issues/109774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed, oncall: pt2",scan when time allows,2025-11-10T21:36:37Z +issue,150799,FSDP in hybrid mode throws _saved_grad_shard error when backward is called on cross-rank all-gathered loss,https://github.com/pytorch/pytorch/issues/150799,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-11-10T18:28:28Z +issue,167409,FFT operations hang indefinitely on RTX 5000 Ada (driver 570.86.15),https://github.com/pytorch/pytorch/issues/167409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: fft",scan when time allows,2025-11-10T18:05:33Z +issue,107800,[feature request] [discussion] Include basic `ctypes` bindings for `cudart`/`cublasLt`/`cublas`/`nvrtc`/`cudnn` with stock PyTorch,https://github.com/pytorch/pytorch/issues/107800,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2025-11-10T16:56:43Z +issue,157015,[WIP][RFC] Compilable flex_attention + Context Parallel,https://github.com/pytorch/pytorch/issues/157015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-11-10T01:19:21Z +issue,108948,Add Lambert W function as torch.special.lambertw,https://github.com/pytorch/pytorch/issues/108948,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2025-11-09T09:29:53Z +issue,147613,torch.nn.AvgPool2d fails with stride >= 2^31 on CUDA,https://github.com/pytorch/pytorch/issues/147613,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: nn, module: pooling",scan when time allows,2025-11-08T22:53:11Z +issue,163425,`torch._grouped_gemm` has poor performance for large batch sizes with dtype BF16,https://github.com/pytorch/pytorch/issues/163425,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: linear algebra, module: performance",scan when time allows,2025-11-08T20:14:42Z +issue,108175,Enable FlashAttentionV2 on Windows,https://github.com/pytorch/pytorch/issues/108175,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-11-08T17:38:08Z +issue,117321,[C++-polyfill] Pt2 - Discussion around user defined type->behavior dispatching,https://github.com/pytorch/pytorch/issues/117321,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-11-07T21:32:22Z +issue,111674,Dynamo Compile samples should record file/line that raised exception,https://github.com/pytorch/pytorch/issues/111674,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-11-07T21:30:57Z +issue,102063,[inlined-inbuilt-nn-modules][dynamo][BE] Revisit call_method of NNModuleVariable,https://github.com/pytorch/pytorch/issues/102063,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-11-07T21:21:21Z +issue,167351,move to gcc11 for CUDA 13 CI tests,https://github.com/pytorch/pytorch/issues/167351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",scan when time allows,2025-11-07T21:08:38Z +issue,118966,"[hierarchical compilation] A way to designate a portion of torch.compile as a noinline block that is compiled/guarded separately, but less disruptive than a graph break (e.g., for loops)",https://github.com/pytorch/pytorch/issues/118966,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, oncall: pt2",scan when time allows,2025-11-07T15:43:19Z +issue,65161,error: ‘_mm512_set_epi8’ was not declared in this scope when build from source on AMD 3600 CPU,https://github.com/pytorch/pytorch/issues/65161,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vectorization",scan when time allows,2025-11-07T15:29:47Z +issue,106608,ROCm & Windows Support,https://github.com/pytorch/pytorch/issues/106608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-07T03:45:44Z +issue,167229,"The quality of button ""Ask AI"" is very low. I propose removing it. The button gets too much area of webpage, made it inconvenient.",https://github.com/pytorch/pytorch/issues/167229,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-11-07T02:16:14Z +issue,167257,Support cuda event input/output in graph with streams,https://github.com/pytorch/pytorch/issues/167257,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-11-06T22:16:35Z +issue,159226,Add support sdpa kernels on gfx1101 (AMD RX7800XT or W7700),https://github.com/pytorch/pytorch/issues/159226,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-11-06T22:13:46Z +issue,126154,Optimizer classes not `dill` picklable after using `torch.compile`,https://github.com/pytorch/pytorch/issues/126154,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: serialization, oncall: pt2",scan when time allows,2025-11-06T21:57:54Z +issue,118884,[Export] [TorchVision] Export Support for TorchVision Detection Models,https://github.com/pytorch/pytorch/issues/118884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export, oncall: pt2",scan when time allows,2025-11-06T21:40:46Z +issue,106130,torch._subclasses.fake_tensor.DynamicOutputShapeException: aten.nonzero.default,https://github.com/pytorch/pytorch/issues/106130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: faketensor,scan when time allows,2025-11-06T15:09:57Z +issue,130282,[custom_ops] We should be able to mark an operator as non-differentiable.,https://github.com/pytorch/pytorch/issues/130282,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-11-06T14:37:08Z +issue,133268,Adafactor compile support tracker,https://github.com/pytorch/pytorch/issues/133268,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: pt2",scan when time allows,2025-11-06T05:12:14Z +issue,126566,`torch.compiler.allow_in_graph` does not create a `call_module` op in fx.Graph in torch 2.3.0,https://github.com/pytorch/pytorch/issues/126566,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-11-06T04:45:53Z +issue,137128,Context manager to skip checks inside decompositions,https://github.com/pytorch/pytorch/issues/137128,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, module: dynamic shapes, module: dynamo, oncall: pt2",scan when time allows,2025-11-06T02:47:19Z +issue,119128,[dynamo] Refactor switch statements to improve compile times,https://github.com/pytorch/pytorch/issues/119128,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, oncall: pt2",scan when time allows,2025-11-06T01:40:33Z +issue,97155,Custom recurrent network takes very long to compile for long sequences,https://github.com/pytorch/pytorch/issues/97155,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, module: python dispatcher, oncall: pt2",scan when time allows,2025-11-06T00:40:50Z +issue,167119,Pyrefly may regress type coverage - consider reenabling mypy,https://github.com/pytorch/pytorch/issues/167119,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-11-05T21:18:55Z +issue,138215,Dynamo x autograd.Function: leverage compiled autograd on graph break in backwards,https://github.com/pytorch/pytorch/issues/138215,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, oncall: pt2",scan when time allows,2025-11-05T21:08:03Z +issue,126846,propagate_real_tensors can cause deadlocks for programs with collectives,https://github.com/pytorch/pytorch/issues/126846,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-11-05T21:07:07Z +issue,118758,Selectively disable dynamic shapes in an inner compile region,https://github.com/pytorch/pytorch/issues/118758,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: dynamo, oncall: pt2",scan when time allows,2025-11-05T21:05:01Z +issue,58734,"Support for `uint16`, `uint32`, and `uint64`",https://github.com/pytorch/pytorch/issues/58734,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, oncall: pt2",scan when time allows,2025-11-05T19:48:10Z +issue,109074,torch.compile/triton holding GIL during compilation and CompiledKernel call results in deadlocks during distributed training,https://github.com/pytorch/pytorch/issues/109074,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: inductor, oncall: pt2",scan when time allows,2025-11-05T19:20:37Z +issue,103891,Docker images: faster linker for `torch.compile`,https://github.com/pytorch/pytorch/issues/103891,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, oncall: pt2",scan when time allows,2025-11-05T19:14:20Z +issue,118573,[Inductor] Unite all possible way to track mutation,https://github.com/pytorch/pytorch/issues/118573,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-11-05T19:02:22Z +issue,124427,"Find a common home for decompositions, perhaps outside of the obliquely named _refs directory",https://github.com/pytorch/pytorch/issues/124427,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-11-05T19:01:18Z +issue,93923,Dynamo uses CONSTANT_MATCH guards for string inputs,https://github.com/pytorch/pytorch/issues/93923,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-11-05T17:54:33Z +issue,124245,[RFC] Add new CPP builder for inductor on pytorch Windows,https://github.com/pytorch/pytorch/issues/124245,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: windows, oncall: cpu inductor, oncall: pt2",scan when time allows,2025-11-05T17:37:58Z +issue,163990,Add CI job to track typing changes made to PyTorch,https://github.com/pytorch/pytorch/issues/163990,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: infra, module: typing",scan when time allows,2025-11-05T15:45:27Z +issue,149301,Unexpected results w/ LayerNorm -- suspecting possible memory issue?,https://github.com/pytorch/pytorch/issues/149301,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: norms and normalization, module: numerical-stability",scan when time allows,2025-11-05T14:02:34Z +issue,135860,Lobpcg for complex gives wrong results.,https://github.com/pytorch/pytorch/issues/135860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: correctness (silent), module: linear algebra",scan when time allows,2025-11-05T02:23:17Z +issue,166611,CuTe DSL NVFP4 GEMM/Grouped GEMM kernels,https://github.com/pytorch/pytorch/issues/166611,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2025-11-04T19:25:35Z +issue,166511,RuntimeError when finetuning Gemma 3 using Unsloth on Databricks,https://github.com/pytorch/pytorch/issues/166511,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-11-04T09:42:40Z +issue,166696,`torch.accelerator.is_available()` fails when privateuseone is used,https://github.com/pytorch/pytorch/issues/166696,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: accelerator,scan when time allows,2025-11-04T07:54:51Z +issue,160420,CachingHostAllocator does not take into account NUMA affinity of GPU's when reusing freed blocks,https://github.com/pytorch/pytorch/issues/160420,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-11-04T07:03:52Z +issue,166823,Change tab order of docs to have main content before left sidebar,https://github.com/pytorch/pytorch/issues/166823,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-11-04T01:39:41Z +issue,153863,"cmake: add USE_SYSTEM_{KLEIDI,CUDNN_FRONTEND,CUTLASS,FMT} options to USE_SYSTEM_LIBS",https://github.com/pytorch/pytorch/issues/153863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-11-03T14:00:56Z +issue,165730,"autocast('cuda', enabled=False) silently ignored on NPU without warning, causing precision issues",https://github.com/pytorch/pytorch/issues/165730,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: error checking, module: privateuse1",scan when time allows,2025-11-03T03:08:33Z +issue,164707,Significant difference in performance between CUSPARSELT and CUTLASS when doing SparseSemiStructuredTensor matmul,https://github.com/pytorch/pytorch/issues/164707,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-11-02T06:37:18Z +issue,166733,benchmark and advertise torch.compile's ability to reduce triton kernel launch overhead,https://github.com/pytorch/pytorch/issues/166733,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: vllm, oncall: pt2",scan when time allows,2025-11-01T03:33:05Z +issue,164891,Better API or GuardRails for Distributed Optimizations with non-SPMD graphs,https://github.com/pytorch/pytorch/issues/164891,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: dynamo, oncall: pt2",scan when time allows,2025-11-01T02:31:01Z +issue,166401,torch.masked.argmax returns different results on CPU vs CUDA when multiple maximum values exist,https://github.com/pytorch/pytorch/issues/166401,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2025-11-01T00:53:45Z +issue,152595,"ROCm, 7900 XTX (gfx1100): Pytorch FLASH_ATTENTION SDPA is 2.5x slower than MATH (fp16, head_dim 256, seqlen 4360, 12 heads)",https://github.com/pytorch/pytorch/issues/152595,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: rocm, module: sdpa",scan when time allows,2025-10-31T23:00:32Z +issue,164860,Driver update on H100 and A100 instances,https://github.com/pytorch/pytorch/issues/164860,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-10-31T17:55:33Z +issue,166686,"pytorch can not load libtorch trained pt pth model ,",https://github.com/pytorch/pytorch/issues/166686,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-10-31T17:27:31Z +issue,166557,Two conflicting definitions of `torch.backends.cudnn.rnn` since #125888,https://github.com/pytorch/pytorch/issues/166557,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2025-10-31T15:39:09Z +issue,123505,RuntimeError: Unrecognized CachingAllocator option: expandable_segments,https://github.com/pytorch/pytorch/issues/123505,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-10-30T11:18:58Z +issue,147449,gradient checkpointing with use_reentrant=False cannot reduce peak memory,https://github.com/pytorch/pytorch/issues/147449,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2025-10-30T05:31:36Z +issue,164813,flex_attention not functional in NVIDIA RTX 2080Ti,https://github.com/pytorch/pytorch/issues/164813,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-10-30T02:34:12Z +issue,165409,[torch.compile][fullgraph][functionalization] In-place op on overlapped view from unfold fails: “tensor being mutated has internal overlap” (eager OK),https://github.com/pytorch/pytorch/issues/165409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-10-30T02:17:03Z +issue,163785,Revisit guarding on unbacked inputs !,https://github.com/pytorch/pytorch/issues/163785,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-10-29T22:58:35Z +issue,148681,foreach CUDA tests flaky on CUDA 12.6+ due to flaky profiler results,https://github.com/pytorch/pytorch/issues/148681,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-10-29T17:12:53Z +issue,129843,Concatenating CSR matrices fails,https://github.com/pytorch/pytorch/issues/129843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-10-29T14:24:34Z +issue,160248,"PyTorch 2.8: gloo-cuda linking issue, undefined symbol",https://github.com/pytorch/pytorch/issues/160248,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-10-29T06:35:22Z +issue,162790,torch.tensor and torch.as_tensor flatten (concatenates) list of 1d tensors instead of nesting them,https://github.com/pytorch/pytorch/issues/162790,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-10-29T04:41:11Z +issue,166116,[CCA] CUDACachingAllocator always release physical memory handle when the expandable segment unmaps.,https://github.com/pytorch/pytorch/issues/166116,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-10-29T02:57:00Z +issue,154074,UNSTABLE Build manywheel docker images for s390x / build-docker-cpu-s390x,https://github.com/pytorch/pytorch/issues/154074,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-10-29T02:37:31Z +issue,165231,_scaled_mm errors with row-major rhs matrix,https://github.com/pytorch/pytorch/issues/165231,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: floatx (formerly float8), module: linear algebra",scan when time allows,2025-10-28T18:50:40Z +issue,144564,Tabulate not official dependency of PyTorch but needed by features like FlopCounterMode,https://github.com/pytorch/pytorch/issues/144564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flop counter,scan when time allows,2025-10-28T12:53:08Z +issue,153996,UR Error when calling grid_sample,https://github.com/pytorch/pytorch/issues/153996,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: windows, module: xpu",scan when time allows,2025-10-28T12:42:45Z +issue,166186,We should have an env var to turn off all dynamic shapes,https://github.com/pytorch/pytorch/issues/166186,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: dynamo, oncall: pt2",scan when time allows,2025-10-28T03:54:22Z +issue,166233,"license: Is it possible to stop using Conda in the Dockerfile? Due to Conda’s licensing issues, many companies have already received legal warning letters.",https://github.com/pytorch/pytorch/issues/166233,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docker",scan when time allows,2025-10-28T03:42:37Z +issue,166243,[BugReport] Exponential distribution may generate zero values on CPU backend,https://github.com/pytorch/pytorch/issues/166243,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-10-28T00:54:41Z +issue,156434,Suggest to use the torch cmake target instead of ${TORCH_LIBRARIES} in the c++ docs,https://github.com/pytorch/pytorch/issues/156434,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-10-27T17:42:34Z +issue,29814,SGD fails on sparse matrix,https://github.com/pytorch/pytorch/issues/29814,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: sparse",scan when time allows,2025-10-27T17:34:16Z +issue,165811,[RFC] A Python backend registration API,https://github.com/pytorch/pytorch/issues/165811,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: python frontend",scan when time allows,2025-10-27T17:28:37Z +issue,165741,Ctx or at least Thread level switch for fp32 IEEE precision tolerance,https://github.com/pytorch/pytorch/issues/165741,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: multithreading",scan when time allows,2025-10-27T17:23:56Z +issue,64932,[RFC] TorchStore - A Shared-Memory Tensor Store,https://github.com/pytorch/pytorch/issues/64932,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-10-27T16:56:02Z +issue,165324,How to enable Bfloat16 when using torch.func.jvp,https://github.com/pytorch/pytorch/issues/165324,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2025-10-27T15:23:35Z +issue,166080,Feature Request: masked convolution,https://github.com/pytorch/pytorch/issues/166080,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-10-27T15:08:49Z +issue,152285,Error after successful build: No module named 'torch._C._distributed_c10d',https://github.com/pytorch/pytorch/issues/152285,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: distributed",scan when time allows,2025-10-27T09:03:18Z +issue,160849,Add new shuffle mechanism for SGD/Adam --- CorgiPile without full data shuffle,https://github.com/pytorch/pytorch/issues/160849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2025-10-27T03:19:59Z +issue,123465,lintrunner-noclang job may diverge from `lintrunner` locally,https://github.com/pytorch/pytorch/issues/123465,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: typing",scan when time allows,2025-10-27T03:04:02Z +issue,127271,ImportError: cannot import name 'triton_key' from 'triton.compiler.compiler' when using triton 2.3.x,https://github.com/pytorch/pytorch/issues/127271,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-10-27T02:44:20Z +issue,165626,Code Style Cleanup: Replace python3 -mpip with the more readable python3 -m pip,https://github.com/pytorch/pytorch/issues/165626,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-10-26T11:21:15Z +issue,157218,[Feature] Taylor expansion pruning,https://github.com/pytorch/pytorch/issues/157218,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pruning,scan when time allows,2025-10-26T08:05:36Z +issue,166219,Why are there so many warnings when building the C++ libtorch project? How to resolve it?,https://github.com/pytorch/pytorch/issues/166219,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: windows",scan when time allows,2025-10-25T15:39:20Z +issue,73359,"Pytorch Typing, for Tensor type annotations",https://github.com/pytorch/pytorch/issues/73359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-10-25T05:41:04Z +issue,139494,Mutating custom ops slower than non-mutating custom ops.,https://github.com/pytorch/pytorch/issues/139494,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, module: vllm, oncall: pt2",scan when time allows,2025-10-24T17:30:15Z +issue,133689,[torchbind x PT2] Handle vllm's ScalarType,https://github.com/pytorch/pytorch/issues/133689,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: torchbind, module: vllm",scan when time allows,2025-10-24T16:38:55Z +issue,150488,Lightweight CUDAGraph backend,https://github.com/pytorch/pytorch/issues/150488,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: inductor, module: vllm, oncall: pt2",scan when time allows,2025-10-24T16:38:40Z +issue,150484,API to specify cudagraph sizes,https://github.com/pytorch/pytorch/issues/150484,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: inductor, module: vllm, oncall: pt2",scan when time allows,2025-10-24T16:38:40Z +issue,150482,API to specify static input indices for cudagraph,https://github.com/pytorch/pytorch/issues/150482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: inductor, module: vllm, oncall: pt2",scan when time allows,2025-10-24T16:38:39Z +issue,150190,"auto_functionalize HOPs are cacheable, also we should check that we allow for manual overrides for cacheability",https://github.com/pytorch/pytorch/issues/150190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vllm,scan when time allows,2025-10-24T16:38:38Z +issue,148983,inference_mode Tensors do not always need to be guarded on,https://github.com/pytorch/pytorch/issues/148983,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vllm,scan when time allows,2025-10-24T16:38:36Z +issue,163428,[vLLM] Crash during inferece in torch dynamo on 2xRTX2080Ti,https://github.com/pytorch/pytorch/issues/163428,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, module: vllm",scan when time allows,2025-10-24T16:38:07Z +issue,162402,Issue with torch.compile - sm110a(Nvidia Jetson Thor),https://github.com/pytorch/pytorch/issues/162402,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: vllm, oncall: pt2",scan when time allows,2025-10-24T16:38:06Z +issue,158699,reports of torch.compile'd rmsnorm numerics different from eager,https://github.com/pytorch/pytorch/issues/158699,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: vllm, oncall: pt2",scan when time allows,2025-10-24T16:38:00Z +issue,161500,nn.TransformerEncoder NaNs with float mask and float src_key_padding_mask only when in both eval() and torch.no_grad(),https://github.com/pytorch/pytorch/issues/161500,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: nn",scan when time allows,2025-10-24T09:51:45Z +issue,166101,Compiling libtorch requires GLIBC_2.35,https://github.com/pytorch/pytorch/issues/166101,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: rocm",scan when time allows,2025-10-24T05:27:17Z +issue,139349,torch.nansum does not work with complex numbers on CPU,https://github.com/pytorch/pytorch/issues/139349,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",scan when time allows,2025-10-23T18:54:10Z +issue,165982,LELU Activation Function: Proposal for PyTorch,https://github.com/pytorch/pytorch/issues/165982,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-10-23T12:33:36Z +issue,101666,Unexpected behavior of fmod op in some float32 input,https://github.com/pytorch/pytorch/issues/101666,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2025-10-23T06:37:14Z +issue,165992,AttributeError: module 'triton.language.math' has no attribute 'tanh',https://github.com/pytorch/pytorch/issues/165992,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-10-23T01:33:28Z +issue,166054,Add XLA to `at::accelerator::getAccelerator` and `at::accelerator::isAccelerator`,https://github.com/pytorch/pytorch/issues/166054,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: accelerator,scan when time allows,2025-10-22T23:47:56Z +issue,165912,Runtime assert on unbacked input inside torch.cond improperly gets lifted to top level,https://github.com/pytorch/pytorch/issues/165912,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes,scan when time allows,2025-10-22T23:44:19Z +issue,85791,nn.CrossEntropyLoss overflow with FP16 and minibatch,https://github.com/pytorch/pytorch/issues/85791,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: loss, module: nn",scan when time allows,2025-10-22T20:20:36Z +issue,166064,Add submodul for `six`,https://github.com/pytorch/pytorch/issues/166064,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2025-10-22T16:16:30Z +issue,166072,UNSTABLE trunk / linux-jammy-py3-clang12-executorch / test (executorch),https://github.com/pytorch/pytorch/issues/166072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-10-22T16:15:20Z +issue,105912,Batching rule for aten::bincount.,https://github.com/pytorch/pytorch/issues/105912,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-10-22T11:28:36Z +issue,128785,"Any plans for a ""torch.minmax"" (min-max normalization) function?",https://github.com/pytorch/pytorch/issues/128785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-10-22T08:02:04Z +issue,166011,cusparse deprecation warning in CUDA 13.0 inductor build,https://github.com/pytorch/pytorch/issues/166011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: sparse",scan when time allows,2025-10-21T19:28:25Z +issue,166014,Make Inductor Fallback Nodes Less Reliant on Invariants from Functionalization / AOT Autograd,https://github.com/pytorch/pytorch/issues/166014,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-10-21T18:59:31Z +issue,165976,Internal Assert Failure with Specific Large Padding Value in torch.nn.Conv3d,https://github.com/pytorch/pytorch/issues/165976,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: error checking",scan when time allows,2025-10-21T14:27:55Z +issue,165826,“Request: Add CUDA support for NVIDIA RTX 5070 Ti (Ada Lovelace Refresh)”,https://github.com/pytorch/pytorch/issues/165826,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-10-20T15:30:01Z +issue,165880,torch.vmap doesn't work on most of the models,https://github.com/pytorch/pytorch/issues/165880,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: functorch",scan when time allows,2025-10-20T14:29:08Z +issue,165902,torchcodec in pytorch url,https://github.com/pytorch/pytorch/issues/165902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-10-20T14:27:16Z +issue,162235,Inconsistent handling of `-0.0` between CPU and CUDA for many operators,https://github.com/pytorch/pytorch/issues/162235,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: nans and infs",scan when time allows,2025-10-20T01:36:26Z +issue,165421,[torch.compile][fullgraph][dynamic shapes] aten.masked_select.default rejected as “Dynamic shape operator” (eager OK),https://github.com/pytorch/pytorch/issues/165421,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-10-19T23:10:31Z +issue,165799,`torch.where` does not accept scalar argument when `out=` is passed,https://github.com/pytorch/pytorch/issues/165799,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-10-19T19:21:34Z +issue,148552,Return type annotation of `Tensor.long()` etc is not narrowed down to dtype-specific names `LongTensor` etc,https://github.com/pytorch/pytorch/issues/148552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-10-19T15:10:34Z +issue,145337,[libTorch] Model initialization on multi-device is slow. It seems to run sequentially in multi-thread,https://github.com/pytorch/pytorch/issues/145337,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-10-19T07:50:16Z +issue,165715,Missing spans in CPU section of profile likely caused by memory allocation with expandable segments,https://github.com/pytorch/pytorch/issues/165715,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: profiler",scan when time allows,2025-10-19T03:23:28Z +issue,165170,"A distributed test ""test_cupy_as_tensor"" fails on B200.",https://github.com/pytorch/pytorch/issues/165170,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2025-10-17T23:58:12Z +issue,58779,C++ Extensions that use ninja should enable colors by default,https://github.com/pytorch/pytorch/issues/58779,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-10-17T18:06:14Z +issue,65083,THPVariable_subclass_dealloc (i.e. tp_dealloc for torch.Tensor) should handle Python error state gracefully,https://github.com/pytorch/pytorch/issues/65083,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: internals",scan when time allows,2025-10-17T18:06:00Z +issue,64185,PyTorch public API cleanup,https://github.com/pytorch/pytorch/issues/64185,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-10-17T17:48:54Z +issue,29137,"torch.sum(tensor, dim=()) is different from np.sum(arr, axis=())",https://github.com/pytorch/pytorch/issues/29137,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: deprecation, module: numpy, module: reductions",scan when time allows,2025-10-17T17:48:03Z +issue,27780,torch.cuda.default_generators documentation referenced but don't exist.,https://github.com/pytorch/pytorch/issues/27780,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2025-10-17T17:42:29Z +issue,3619,Multiprocessing with torch.solve hangs,https://github.com/pytorch/pytorch/issues/3619,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: multiprocessing",scan when time allows,2025-10-17T17:38:50Z +issue,165251,out_dtype for einsum,https://github.com/pytorch/pytorch/issues/165251,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: type promotion",scan when time allows,2025-10-17T15:55:36Z +issue,165744,Inconsistency in `conv_transpose2d` for float16 with inf weights between CPU and CUDA,https://github.com/pytorch/pytorch/issues/165744,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nans and infs",scan when time allows,2025-10-17T14:42:16Z +issue,165739,Inconsistent in `torch.fft.irfft` and `torch.fft.ifftn` for complex64 inputs containing `Inf`,https://github.com/pytorch/pytorch/issues/165739,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: nans and infs",scan when time allows,2025-10-17T14:41:15Z +issue,165685,"`PYTORCH_CUDA_ALLOC_CONF=""expandable_segments:True""` triggers `RuntimeError: pidfd_getfd: Operation not permitted` for CUDA IPC transfer",https://github.com/pytorch/pytorch/issues/165685,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-10-17T04:32:48Z +issue,165634,make_fx pre_dispatch=True produces incorrect graph when BatchedFallback is triggered,https://github.com/pytorch/pytorch/issues/165634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-10-17T04:30:22Z +issue,165578,Out of tree backend documentation does not seem accurate,https://github.com/pytorch/pytorch/issues/165578,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: privateuse1",scan when time allows,2025-10-17T04:30:01Z +issue,165570,Add support for element-wise operations to flop counter,https://github.com/pytorch/pytorch/issues/165570,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flop counter,scan when time allows,2025-10-17T04:29:30Z +issue,114629,[DeviceMesh] `init_device_mesh` does not support CPU-only,https://github.com/pytorch/pytorch/issues/114629,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-10-17T00:42:43Z +issue,165684,Python version in test CI job doesn't align with the python version in the docker build log,https://github.com/pytorch/pytorch/issues/165684,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docker, module: infra",scan when time allows,2025-10-16T22:51:16Z +issue,160838,backward on 5090(sm_120) much slower than on 4090,https://github.com/pytorch/pytorch/issues/160838,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2025-10-16T14:45:17Z +issue,165621,Feedback about RNN,https://github.com/pytorch/pytorch/issues/165621,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rnn",scan when time allows,2025-10-16T14:18:23Z +issue,155671,torch.clamp throws overflow error on CPU but not on CUDA,https://github.com/pytorch/pytorch/issues/155671,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: error checking",scan when time allows,2025-10-15T23:21:38Z +issue,114605,vmap: Transform single-element tensor to integer,https://github.com/pytorch/pytorch/issues/114605,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-10-15T19:03:16Z +issue,107873,"[BE] Consolidation of SymNode methods constant_int, maybe_as_int, etc",https://github.com/pytorch/pytorch/issues/107873,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-10-15T18:54:57Z +issue,106614,Case study of torch.compile / cpp inductor on CPU: min_sum / mul_sum with 1d / matmul-like with static / dynamic shapes,https://github.com/pytorch/pytorch/issues/106614,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: inductor, oncall: pt2",scan when time allows,2025-10-15T18:35:31Z +issue,94167,"Faster `pad_sequence` and `tensor_split` function with CUDA kernel, are they possible?",https://github.com/pytorch/pytorch/issues/94167,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: rnn, oncall: pt2",scan when time allows,2025-10-15T18:28:10Z +issue,80372,[BE] Update ProcessGroupWrapper tests to test other collective message,https://github.com/pytorch/pytorch/issues/80372,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-10-15T16:52:25Z +issue,48959,[Poll] Support DistributedDataParallel (DDP) in PyTorch C++ API (libtorch),https://github.com/pytorch/pytorch/issues/48959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2025-10-15T15:49:55Z +issue,165038,Print float tensors on devices that do not support fp64,https://github.com/pytorch/pytorch/issues/165038,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: privateuse1, module: python frontend",scan when time allows,2025-10-15T07:16:53Z +issue,147044,Clean up backend_type_map from distributed_c10d,https://github.com/pytorch/pytorch/issues/147044,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-10-15T06:28:35Z +issue,129649,Low GPU precision,https://github.com/pytorch/pytorch/issues/129649,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability",scan when time allows,2025-10-15T05:34:55Z +issue,122274,CUDA: Performance issues with `torch.unique`,https://github.com/pytorch/pytorch/issues/122274,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2025-10-15T01:53:39Z +issue,146817,BF16 linear(matmul) operator 100x slower on odd matrix dimension sizes on A100,https://github.com/pytorch/pytorch/issues/146817,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: linear algebra, module: performance",scan when time allows,2025-10-15T01:40:57Z +issue,147134,`torch.mul` uses `OpMathType` for computation.,https://github.com/pytorch/pytorch/issues/147134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: half, module: numerical-stability",scan when time allows,2025-10-14T23:48:00Z +issue,151433,The compare interface of the memory snapshot visualization tool “_memory_viz.py” has a bug,https://github.com/pytorch/pytorch/issues/151433,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, oncall: profiler",scan when time allows,2025-10-14T23:45:04Z +issue,165425,Support FFT operations in FLOP counter,https://github.com/pytorch/pytorch/issues/165425,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: flop counter",scan when time allows,2025-10-14T23:15:49Z +issue,164732,Fake tensor leakage in aot_autograd by itself,https://github.com/pytorch/pytorch/issues/164732,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: functorch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-10-14T18:21:11Z +issue,165244,torch.tensor slicing with negative step raises ValueError,https://github.com/pytorch/pytorch/issues/165244,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: python frontend",scan when time allows,2025-10-14T15:05:25Z +issue,165277,"autowrap_modules should be tuple[ModuleType, ...]",https://github.com/pytorch/pytorch/issues/165277,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-10-14T13:56:12Z +issue,165312,TransformerEncoder crashes with IndexError when num_layers=0,https://github.com/pytorch/pytorch/issues/165312,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-10-14T13:34:02Z +issue,119574,Cannot configure dist_timeout when using device_mesh,https://github.com/pytorch/pytorch/issues/119574,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-10-14T08:42:59Z +issue,154226,NotImplementedError when computing JVP of Attention,https://github.com/pytorch/pytorch/issues/154226,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, module: sdpa",scan when time allows,2025-10-14T08:16:43Z +issue,156062,torch.nn.functional.conv_transpose3d return inconsistent results when weight containing inf between CPU and GPU,https://github.com/pytorch/pytorch/issues/156062,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-10-14T08:12:55Z +issue,165319,Memory leak when converting from numpy array,https://github.com/pytorch/pytorch/issues/165319,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: numpy",scan when time allows,2025-10-14T08:06:37Z +issue,164925,add support for distributed backends to specify default backend in multiple backends format,https://github.com/pytorch/pytorch/issues/164925,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-10-14T03:17:00Z +issue,165122,RPC: abort when RRef from remote() is mistakenly called with wait() due to unhandled exception in thread pool,https://github.com/pytorch/pytorch/issues/165122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-10-13T17:33:50Z +issue,164799,Check behavior of multicast object release when using clang to compile PyTorch,https://github.com/pytorch/pytorch/issues/164799,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2025-10-13T17:19:21Z +issue,164843,`set_model_state_dict()` causes `AssertionError: found no DeviceMesh from dtensor args for c10d.broadcast_.default!`,https://github.com/pytorch/pytorch/issues/164843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-10-13T17:19:01Z +issue,164122,[CUDA][Sparse][Linalg] torch.sparse.spsolve is unavailable,https://github.com/pytorch/pytorch/issues/164122,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: linear algebra, module: sparse",scan when time allows,2025-10-13T13:38:08Z +issue,132292,Can we avoid RecordStream for P2P ops?,https://github.com/pytorch/pytorch/issues/132292,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-10-13T08:21:44Z +issue,130073,☂️ TORCH_INTERNAL_ASSERT failures while running CI,https://github.com/pytorch/pytorch/issues/130073,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: ci, module: dynamic shapes",scan when time allows,2025-10-13T06:34:57Z +issue,126738,FakeTensor `forward()` on `bert` attention when using FakeTensor returns a DataDependentOutputException,https://github.com/pytorch/pytorch/issues/126738,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: faketensor",scan when time allows,2025-10-13T05:48:52Z +issue,160438,torch.autocast with no_grad() on aliased model disables gradients for subsequent forward on same instance,https://github.com/pytorch/pytorch/issues/160438,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2025-10-13T00:52:18Z +issue,82053,Inconsistent implementation of quant_utils:: ChooseQuantizationParams compared with fbgemm:: ChooseQuantizationParams,https://github.com/pytorch/pytorch/issues/82053,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2025-10-12T22:40:48Z +issue,31394,torch.scatter_logsumexp,https://github.com/pytorch/pytorch/issues/31394,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2025-10-12T21:32:22Z +issue,140696,inconsistency in ```torch.nn.functional.adaptive_avg_pool3d``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140696,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2025-10-12T20:31:18Z +issue,114569,inconsistency on torch.clamp,https://github.com/pytorch/pytorch/issues/114569,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nans and infs",scan when time allows,2025-10-12T02:09:49Z +issue,142232,torch.tensor.to(‘cuda’) so slow in jetson orin,https://github.com/pytorch/pytorch/issues/142232,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: jetson",scan when time allows,2025-10-11T23:45:29Z +issue,153497,NotImplementedError: Could not run 'aten::log' with arguments from the 'SparseCUDA' backend.,https://github.com/pytorch/pytorch/issues/153497,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-10-11T23:35:58Z +issue,161381,[XPU] ​​Incorrect free_memory reported by torch.xpu.mem_get_info()​,https://github.com/pytorch/pytorch/issues/161381,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xpu,scan when time allows,2025-10-11T01:58:32Z +issue,165173,Add CUDA 13.0 in CI,https://github.com/pytorch/pytorch/issues/165173,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",scan when time allows,2025-10-10T23:52:46Z +issue,150021,[ued] I want to see the full sizes/strides in TORCH_LOGS=recompiles,https://github.com/pytorch/pytorch/issues/150021,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-10-10T16:23:21Z +issue,121286,"Assertion error (`assert op == ""==""`) in symbolic_shapes.prettify_results for operator `<=`",https://github.com/pytorch/pytorch/issues/121286,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",scan when time allows,2025-10-10T16:09:38Z +issue,114414,[BE] remove SmallVector optimization in PyInterpreter.cpp when storing custom sizes,https://github.com/pytorch/pytorch/issues/114414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: dynamic shapes",scan when time allows,2025-10-10T15:43:00Z +issue,165100,Header files not found during build,https://github.com/pytorch/pytorch/issues/165100,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-10-10T13:43:50Z +issue,165042,constant ones/zeros tensor,https://github.com/pytorch/pytorch/issues/165042,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-10-10T13:42:00Z +issue,144824,[MPS] Indexing Returns 0 if OOB,https://github.com/pytorch/pytorch/issues/144824,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",scan when time allows,2025-10-10T12:43:08Z +issue,153327,FPE when calling `torch.pixel_shuffle()` with empty tensors and large upscale_factor,https://github.com/pytorch/pytorch/issues/153327,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: empty tensor, module: error checking",scan when time allows,2025-10-10T09:51:26Z +issue,123119,`nn.Embedding` behaves differently on CPU (fails) and MPS (fails not),https://github.com/pytorch/pytorch/issues/123119,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: mps, module: nn",scan when time allows,2025-10-10T09:31:20Z +issue,165040,DISABLED rocm / linux-jammy-rocm-py3_10-gfx1100 / test (default),https://github.com/pytorch/pytorch/issues/165040,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: rocm",scan when time allows,2025-10-09T23:59:25Z +issue,132067,Pickle loads torch tensor memory leak,https://github.com/pytorch/pytorch/issues/132067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-10-09T21:31:20Z +issue,164302,Mixing cuda graph and non cuda graph code works poorly in the presence of autograd.,https://github.com/pytorch/pytorch/issues/164302,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda graphs",scan when time allows,2025-10-09T18:08:40Z +issue,131283,"`input` argument of `nan_to_num()` works with `complex` type but `nan`, `posinf` and `neginf` argument don't work with `complex` type",https://github.com/pytorch/pytorch/issues/131283,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2025-10-09T13:27:09Z +issue,67158,Make streams used for NCCL operations configurable,https://github.com/pytorch/pytorch/issues/67158,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2025-10-09T03:32:19Z +issue,164062,[strict-export][install_free_tensors] Tracker for issues,https://github.com/pytorch/pytorch/issues/164062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export, oncall: pt2",scan when time allows,2025-10-08T23:54:49Z +issue,163013,Compiled HOPS do not support compilation when run under a custom DispatchMode,https://github.com/pytorch/pytorch/issues/163013,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-10-08T20:09:19Z +issue,164872,Tensor subclass slice during `inference_mode` fails,https://github.com/pytorch/pytorch/issues/164872,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, oncall: quantization",scan when time allows,2025-10-08T19:33:54Z +issue,164951,Docker checkouts take 30+ min on H100 runners,https://github.com/pytorch/pytorch/issues/164951,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-10-08T17:12:25Z +issue,164759,Bug : Side effect in th.func.vjp of th.func.jvp,https://github.com/pytorch/pytorch/issues/164759,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2025-10-07T21:34:08Z +issue,164787,Add static build for NCCL in CI,https://github.com/pytorch/pytorch/issues/164787,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",scan when time allows,2025-10-07T21:30:38Z +issue,164862,Fix incomplete exception handling in validate_input_col function,https://github.com/pytorch/pytorch/issues/164862,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2025-10-07T21:30:05Z +issue,158782,[BE] linter for base docker image configurations with docker_builds.yml,https://github.com/pytorch/pytorch/issues/158782,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docker, module: lint",scan when time allows,2025-10-07T20:33:52Z +issue,118842,CUDA Extension Bug for Pytorch 2.2.0,https://github.com/pytorch/pytorch/issues/118842,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda",scan when time allows,2025-10-07T20:03:43Z +issue,95408,Parallel Associative Scan,https://github.com/pytorch/pytorch/issues/95408,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-10-07T19:21:22Z +issue,154311,torch.inf.to(torch.int32) produces different values on CPU vs CUDA,https://github.com/pytorch/pytorch/issues/154311,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nans and infs",scan when time allows,2025-10-07T18:17:05Z +issue,164756,[PP] Unshard and reshard not working with FSDP,https://github.com/pytorch/pytorch/issues/164756,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pipelining, oncall: distributed",scan when time allows,2025-10-07T16:57:43Z +issue,25570,Remote memory access similar to MPI one-sided in pytorch,https://github.com/pytorch/pytorch/issues/25570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mpi, oncall: distributed",scan when time allows,2025-10-07T16:18:04Z +issue,98836,PyTorch's packaged libgomp causes significant performance penalties on CPU when used together with other Python packages,https://github.com/pytorch/pytorch/issues/98836,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: multithreading",scan when time allows,2025-10-07T15:27:59Z +issue,164786,How should we handle PyTorch build flags in torch/headeronly for custom ops?,https://github.com/pytorch/pytorch/issues/164786,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-10-07T15:26:28Z +issue,100249,"[discussion] ""TensorList"" as first-class python-world abstraction (to mirror existing dispatcher's `TensorList` and `torch::autograd::tensor_list`) and as key for dispatch for merging `torch._foreach_*` into regular `torch.*` functions",https://github.com/pytorch/pytorch/issues/100249,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-10-07T14:02:40Z +issue,164331,Anomaly mode (detect_anomaly) should populate backward stacks with torch.profiler.profile(with_stack=True),https://github.com/pytorch/pytorch/issues/164331,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: profiler",scan when time allows,2025-10-06T19:32:59Z +issue,137392,segfault when using DTensor with nonblocking nccl comm,https://github.com/pytorch/pytorch/issues/137392,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: nccl, oncall: distributed",scan when time allows,2025-10-06T10:28:38Z +issue,152717,Cannot mask a DTensor,https://github.com/pytorch/pytorch/issues/152717,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-10-05T13:01:54Z +issue,47052,Activation functions for complex tensors,https://github.com/pytorch/pytorch/issues/47052,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nn",scan when time allows,2025-10-04T12:24:51Z +issue,77049,Peak GPU-memory usage extremely huge when sorting with torch.sort,https://github.com/pytorch/pytorch/issues/77049,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: sorting and selection",scan when time allows,2025-10-03T17:16:47Z +issue,155472,[RFC][c10d] Make it easier to act on group,https://github.com/pytorch/pytorch/issues/155472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-10-03T01:41:09Z +issue,125984,`torch.export` code that includes `torch.autograd.grad` RuntimeError: element 0 of tensors does not require grad and does not have a grad_fn,https://github.com/pytorch/pytorch/issues/125984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: onnx, oncall: export, oncall: pt2",scan when time allows,2025-10-02T21:29:48Z +issue,164249,Accuracy issue with sum-reduction on A100,https://github.com/pytorch/pytorch/issues/164249,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability, module: tests",scan when time allows,2025-10-02T14:57:43Z +issue,162719,linalg.eig does not get parallelized on CPU,https://github.com/pytorch/pytorch/issues/162719,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: linear algebra, module: performance",scan when time allows,2025-10-02T12:03:49Z +issue,101241,Different results with vmap when using torch.jit.script,https://github.com/pytorch/pytorch/issues/101241,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, oncall: jit",scan when time allows,2025-10-02T12:03:26Z +issue,153976,MPS (Metal) CTC Loss Implementation (Gist link),https://github.com/pytorch/pytorch/issues/153976,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-10-02T02:58:52Z +issue,77724,FSDP: enhanced shared parameter support,https://github.com/pytorch/pytorch/issues/77724,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-10-01T17:44:24Z +issue,163709,Inconsistent handling of backend floating point precision settings (related to PR #125888),https://github.com/pytorch/pytorch/issues/163709,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance, module: python frontend",scan when time allows,2025-10-01T17:40:17Z +issue,102109,Can't vmap over a slice expression,https://github.com/pytorch/pytorch/issues/102109,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-09-30T20:44:12Z +issue,158127,"A ""Merge-when-ready"" label when ci finishes?",https://github.com/pytorch/pytorch/issues/158127,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-09-30T01:27:08Z +issue,71465,torch.nn.LayerNorm support for arbitrary axis in order to allow NCHW application,https://github.com/pytorch/pytorch/issues/71465,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-09-29T23:18:37Z +issue,27647,🚀 Graceful RPCAgent termination in multi-driver scenario,https://github.com/pytorch/pytorch/issues/27647,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2025-09-29T23:16:37Z +issue,61417,Reductions tracking issue,https://github.com/pytorch/pytorch/issues/61417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: reductions",scan when time allows,2025-09-29T23:15:54Z +issue,22402,PyTorch Tensor subclasses and protocols for NumPy interoperability,https://github.com/pytorch/pytorch/issues/22402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-09-29T23:14:13Z +issue,22815,TensorImpl de-virtualization,https://github.com/pytorch/pytorch/issues/22815,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dependency bug, module: internals",scan when time allows,2025-09-29T23:12:08Z +issue,25785,[Feature request] modified Cholesky decomposition,https://github.com/pytorch/pytorch/issues/25785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-09-29T23:11:08Z +issue,31895,"torch.masked_select out argument can easily be misused, because output shape is dynamically computed",https://github.com/pytorch/pytorch/issues/31895,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: safe resize",scan when time allows,2025-09-29T23:10:54Z +issue,24243,hasSideEffects INTERNAL ASSERT FAILED when using .split method with JIT,https://github.com/pytorch/pytorch/issues/24243,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2025-09-29T23:04:37Z +issue,20367,Overhead performance regression over time umbrella issue.,https://github.com/pytorch/pytorch/issues/20367,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: internals, module: performance",scan when time allows,2025-09-29T23:02:34Z +issue,18631,FP32 depthwise convolution is slow in GPU,https://github.com/pytorch/pytorch/issues/18631,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn, module: dependency bug",scan when time allows,2025-09-29T23:01:25Z +issue,18182,Update weight initialisations to current best practices,https://github.com/pytorch/pytorch/issues/18182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: nn",scan when time allows,2025-09-29T23:01:08Z +issue,11390,Views created in no_grad block still have requires_grad=True,https://github.com/pytorch/pytorch/issues/11390,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2025-09-29T23:00:48Z +issue,15829,computing entropy of a tensor,https://github.com/pytorch/pytorch/issues/15829,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-09-29T22:56:42Z +issue,158428,Inconsistent torch.cholesky_inverse results on complex128 between CPU and CUDA,https://github.com/pytorch/pytorch/issues/158428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-09-29T21:44:15Z +issue,163950,ImportError: cannot import name 'skip_code' from 'torch._C._dynamo.eval_frame' (unknown location),https://github.com/pytorch/pytorch/issues/163950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: pt2",scan when time allows,2025-09-29T16:48:15Z +issue,74540,No factory functions for strided quantized tensors,https://github.com/pytorch/pytorch/issues/74540,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2025-09-29T16:00:42Z +issue,164061,GPU Memory Leak due to distributions,https://github.com/pytorch/pytorch/issues/164061,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-09-29T14:54:42Z +issue,163821,Support for nn.BitLinear layer,https://github.com/pytorch/pytorch/issues/163821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-09-29T13:55:57Z +issue,154842,`torch.linalg.solve` does not raise an error for singular matrix on CPU.,https://github.com/pytorch/pytorch/issues/154842,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-09-29T13:48:33Z +issue,164089,Failing Operator Benchmarks,https://github.com/pytorch/pytorch/issues/164089,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-09-29T13:46:11Z +issue,67470,"integer (and possibly float16) matmul fails test_noncontiguous_samples on CPU (and all backward dtypes, too?)",https://github.com/pytorch/pytorch/issues/67470,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: linear algebra",scan when time allows,2025-09-27T03:31:41Z +issue,53178,Exceedingly different F.conv2d outputs on cuda/cpu,https://github.com/pytorch/pytorch/issues/53178,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: correctness (silent), module: cuda, module: cudnn",scan when time allows,2025-09-27T03:30:44Z +issue,48900,addmm with out= argument returns incorrect result,https://github.com/pytorch/pytorch/issues/48900,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: docs, module: linear algebra, module: partial aliasing",scan when time allows,2025-09-27T03:30:07Z +issue,68070,C++ at::Tensor's pinned_memory status is not printing out correctly.,https://github.com/pytorch/pytorch/issues/68070,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cpp",scan when time allows,2025-09-27T03:27:19Z +issue,56794,No tensor operations allowed inside at::parallel_for,https://github.com/pytorch/pytorch/issues/56794,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2025-09-26T23:50:42Z +issue,64412,Memory leak in multi-thread inference,https://github.com/pytorch/pytorch/issues/64412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: multithreading, module: performance",scan when time allows,2025-09-26T23:49:45Z +issue,68174,RFC: “What’s in a (NumPY) name?” — Changing PyTorch Operator Names,https://github.com/pytorch/pytorch/issues/68174,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-09-26T23:49:06Z +issue,71774,matmul returns uninitialized memory for int64 tensors with inner dimension of zero,https://github.com/pytorch/pytorch/issues/71774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: linear algebra",scan when time allows,2025-09-26T23:46:21Z +issue,75432,Rollup: Top forward-over-reverse formulas,https://github.com/pytorch/pytorch/issues/75432,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad",scan when time allows,2025-09-26T23:45:15Z +issue,74613,OpInfo request for `nn.functional` and `unbind`,https://github.com/pytorch/pytorch/issues/74613,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2025-09-26T23:44:05Z +issue,98456,README could use link to governance,https://github.com/pytorch/pytorch/issues/98456,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-09-26T23:37:12Z +issue,111384,[quant][pt2] Default batchnorm aten op has poor numerics during QAT,https://github.com/pytorch/pytorch/issues/111384,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2025-09-26T23:35:18Z +issue,117586,PyTorch 2.1.0 High memory utilization compared to PyTorch 2.0.1,https://github.com/pytorch/pytorch/issues/117586,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-09-26T23:33:59Z +issue,163789,[docs] instructions to locally build docs are underspecified,https://github.com/pytorch/pytorch/issues/163789,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-09-26T22:34:16Z +issue,163982,Need to update Magma version in Pytorch,https://github.com/pytorch/pytorch/issues/163982,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-09-26T19:23:09Z +issue,163902,Add banner for docker sevs to hud,https://github.com/pytorch/pytorch/issues/163902,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docker",scan when time allows,2025-09-26T02:30:09Z +issue,161713,[Proposal] Improving Type Signatures for Types Exported by torch,https://github.com/pytorch/pytorch/issues/161713,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-09-25T13:52:43Z +issue,163062,GIL is not released when creating a device tensor using a shape depending on data from another device tensor,https://github.com/pytorch/pytorch/issues/163062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: python frontend",scan when time allows,2025-09-25T02:23:29Z +issue,163659,Allow double in native_functions.yaml as a schema type,https://github.com/pytorch/pytorch/issues/163659,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: dispatch, module: library, module: pt2-dispatcher",scan when time allows,2025-09-24T18:45:19Z +issue,158831,Severe Performance Degradation in Batched SVD on A100-40G,https://github.com/pytorch/pytorch/issues/158831,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",scan when time allows,2025-09-24T09:42:01Z +issue,163048,[compile/aot-autograd] `torch.autograd.grad` w.r.t. an intermediate tensor works in eager but fails under `torch.compile` (“tensor not used in the graph”),https://github.com/pytorch/pytorch/issues/163048,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-09-24T03:34:55Z +issue,124645,Registering function that takes `const SymInt&` to op that accepts `SymInt` leads to cryptic error,https://github.com/pytorch/pytorch/issues/124645,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: dynamic shapes, module: pt2-dispatcher",scan when time allows,2025-09-24T02:12:03Z +issue,162971,[CD] Reasonable time constraint for binary builds,https://github.com/pytorch/pytorch/issues/162971,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: ci",scan when time allows,2025-09-23T20:23:03Z +issue,163302,[vLLM CI] Periodically run B200 tests,https://github.com/pytorch/pytorch/issues/163302,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-09-23T20:13:26Z +issue,163577,Lintrunner failures on files that only manifest on CI,https://github.com/pytorch/pytorch/issues/163577,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-09-23T20:05:06Z +issue,159674,OOM when using use_mem_pool due to restriction on retry,https://github.com/pytorch/pytorch/issues/159674,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-09-23T17:58:14Z +issue,163657,UNSTABLE periodic / linux-jammy-cuda12.4-py3.10-gcc11 / test (legacy_nvidia_driver),https://github.com/pytorch/pytorch/issues/163657,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-09-23T16:18:42Z +issue,163624,[aoti] [xpu] [null-pointer-deference] potential npt issue in `sycl_runtime_wrappers.h`,https://github.com/pytorch/pytorch/issues/163624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",scan when time allows,2025-09-23T16:02:18Z +issue,163574,Bfloat16 large numerical discrepancy on Blackwell,https://github.com/pytorch/pytorch/issues/163574,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: correctness (silent), module: cuda, module: numerical-stability",scan when time allows,2025-09-23T15:50:27Z +issue,144464,"[ONNX] Trying to export QAT model with torch.onnx.export, running into Problems that workarounds cant fix. Is ONNX export planned for QAT?",https://github.com/pytorch/pytorch/issues/144464,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-09-23T13:34:16Z +issue,140520,Enable autograd for padded dense forward / backward operators.,https://github.com/pytorch/pytorch/issues/140520,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nestedtensor",scan when time allows,2025-09-22T22:27:06Z +issue,162820,[CI][CUDA][Distributed] test_ring_flex_attention failed on 8xB200 Runner,https://github.com/pytorch/pytorch/issues/162820,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: flex attention, module: higher order operators, module: pt2-dispatcher",scan when time allows,2025-09-22T22:23:48Z +issue,163502,at::Tensor::to_sparse_csr() does not produce correct result in Debug configuration when tensor is big sparse CSC on CUDA device in LibTorch 2.8.0,https://github.com/pytorch/pytorch/issues/163502,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cuda, module: debug-build, module: sparse",scan when time allows,2025-09-22T19:45:59Z +issue,90607,"Export to ONNX of `as_strided()` hard codes stride in the graph, although it should be dynamic",https://github.com/pytorch/pytorch/issues/90607,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-09-22T18:39:40Z +issue,162449,optimum-cli export onnx for model intfloat/multilingual-e5-large-instruct fails with sufficient accuracy,https://github.com/pytorch/pytorch/issues/162449,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-09-22T18:38:48Z +issue,126774,[ONNX] Implement aten._trilinear and aten.bilinear.default,https://github.com/pytorch/pytorch/issues/126774,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-09-22T18:38:24Z +issue,159380,`torch.nested.nested_tensor_from_jagged` does not automatically compute `max_seqlen`,https://github.com/pytorch/pytorch/issues/159380,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-09-22T17:53:31Z +issue,162918,`torch.embedding` should accept `indices` of dtype `torch.uint8`,https://github.com/pytorch/pytorch/issues/162918,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: nn",scan when time allows,2025-09-22T16:29:45Z +issue,119108,[MPS] Tracking issue for ModuleInfo failures when enabling testing for `torch.float16`,https://github.com/pytorch/pytorch/issues/119108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-09-22T14:41:09Z +issue,162178,[CUDA] Umbrella Issues/Failures on B200 Runner,https://github.com/pytorch/pytorch/issues/162178,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-09-22T00:29:33Z +issue,153737,SystemError: PY_SSIZE_T_CLEAN macro must be defined for '#' formats - PyTorch compile fails with Python 3.12,https://github.com/pytorch/pytorch/issues/153737,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: python version, oncall: pt2",scan when time allows,2025-09-21T19:16:22Z +issue,161330,better tool to repro issues in 1 dynamo graph,https://github.com/pytorch/pytorch/issues/161330,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-09-20T01:35:49Z +issue,162782,Is `torch.nn.functional.gumbel_softmax` going to be deprecated?,https://github.com/pytorch/pytorch/issues/162782,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: nn",scan when time allows,2025-09-19T17:36:22Z +issue,40319,Too many labels in the repo,https://github.com/pytorch/pytorch/issues/40319,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2025-09-19T01:25:50Z +issue,163142,`compile_kernel`: fast inline compilation with nvrtc tracker,https://github.com/pytorch/pytorch/issues/163142,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-09-18T22:43:32Z +issue,163290,Tracker for migrating xformers to torch stable ABI,https://github.com/pytorch/pytorch/issues/163290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: abi,scan when time allows,2025-09-18T20:43:43Z +issue,160816,torch._scaled_mm crash with cublas error when out_dtype is not specified,https://github.com/pytorch/pytorch/issues/160816,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: floatx (formerly float8)",scan when time allows,2025-09-18T14:38:51Z +issue,29961,Add support for integer matrix multiplication (particularly for dtype = torch.int8 ),https://github.com/pytorch/pytorch/issues/29961,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-09-18T13:50:01Z +issue,157714,AssertHandler::printMessage On Intel GPU,https://github.com/pytorch/pytorch/issues/157714,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xpu,scan when time allows,2025-09-18T07:12:23Z +issue,90585,[threaded pg] All threads share one Random Number Generator,https://github.com/pytorch/pytorch/issues/90585,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: random, module: threaded pg",scan when time allows,2025-09-18T03:15:02Z +issue,162964,[feat] OpInfo-like concept for subclass edge conditions to handle,https://github.com/pytorch/pytorch/issues/162964,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2025-09-18T01:14:03Z +issue,162786,[MSVC] LINK : fatal error LNK1181: cannot open input file 'sleef\src\libm\CMakeFiles\mkrename.dir\mkrename.c.obj',https://github.com/pytorch/pytorch/issues/162786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2025-09-18T01:12:51Z +issue,153502,Add MXFP8 Support to scaled_grouped_gemm,https://github.com/pytorch/pytorch/issues/153502,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: floatx (formerly float8)",scan when time allows,2025-09-17T20:55:53Z +issue,162907,Poor bf16 precision in backward torch.asin,https://github.com/pytorch/pytorch/issues/162907,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpu",scan when time allows,2025-09-17T20:30:36Z +issue,162911,`torch._foreach_...` functions error on zero-length inputs,https://github.com/pytorch/pytorch/issues/162911,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mta,scan when time allows,2025-09-17T20:28:54Z +issue,102894,"BCELoss and BCEWithLogitsLoss differ when one of the input logits is float(""inf"")",https://github.com/pytorch/pytorch/issues/102894,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nn",scan when time allows,2025-09-17T18:13:47Z +issue,140514,Error with fused AdamW and FSDP,https://github.com/pytorch/pytorch/issues/140514,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: optimizer",scan when time allows,2025-09-17T16:00:32Z +issue,103580,Support ByteTensor and ShortTensor for nn.Embedding and nn.EmbeddingBag,https://github.com/pytorch/pytorch/issues/103580,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-09-17T05:46:10Z +issue,69532,[feature request] A rank-revealing SVD for better stability in backward.,https://github.com/pytorch/pytorch/issues/69532,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra, module: numerical-stability",scan when time allows,2025-09-17T04:42:36Z +issue,106667,[docs] Idea collection of examples of custom ops / inline torch extensions,https://github.com/pytorch/pytorch/issues/106667,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: docs",scan when time allows,2025-09-17T04:40:01Z +issue,36524,"Drop _stacklevel from argspecs of F.softmax, F.softmin, F.log_softmax (for implicit dim has been long deprecated)",https://github.com/pytorch/pytorch/issues/36524,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-09-17T03:51:03Z +issue,83163,torch.nn.functional.log_softmax parameter '_stacklevel' undocumented,https://github.com/pytorch/pytorch/issues/83163,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-09-17T03:48:59Z +issue,156618,THPStorage_shareCuda zeroes CUDA tensor,https://github.com/pytorch/pytorch/issues/156618,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dependency bug, module: multiprocessing, module: wsl",scan when time allows,2025-09-16T21:53:17Z +issue,40497,Mixed precision causes NaN loss,https://github.com/pytorch/pytorch/issues/40497,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: nans and infs",scan when time allows,2025-09-16T13:41:35Z +issue,162870,[RFC] library function with 64+ arguments,https://github.com/pytorch/pytorch/issues/162870,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: library",scan when time allows,2025-09-15T23:25:57Z +issue,151030,Unexpected memory usage in FSDP 2 Hybrid Sharding (HSDP),https://github.com/pytorch/pytorch/issues/151030,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed, oncall: distributed checkpointing",scan when time allows,2025-09-15T21:44:30Z +issue,162700,"RuntimeError: The tensor has a non-zero number of elements, but its data is not allocated yet during torch.backward with InternVL",https://github.com/pytorch/pytorch/issues/162700,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-09-15T20:45:48Z +issue,162154,Linear algebra operations hang indefinitely doing nothing when run in subprocesses,https://github.com/pytorch/pytorch/issues/162154,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: linear algebra, module: multiprocessing",scan when time allows,2025-09-15T15:12:54Z +issue,158861,Different behavior between sparse and dense tensors with broadcasting multiplication.,https://github.com/pytorch/pytorch/issues/158861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-09-15T06:07:44Z +issue,94294,CUBLAS_STATUS_NOT_SUPPORTED when calling cublasDgemv,https://github.com/pytorch/pytorch/issues/94294,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2025-09-14T06:20:10Z +issue,162837,XPU backend is falling back to run on the CPU.,https://github.com/pytorch/pytorch/issues/162837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xpu,scan when time allows,2025-09-14T00:21:50Z +issue,58054,PyTorch master failed to build with glog 0.5.0.,https://github.com/pytorch/pytorch/issues/58054,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-09-13T00:17:24Z +issue,124505,"[feature request] allow torch.compile calls with compiler options, but without modifying global dynamo options",https://github.com/pytorch/pytorch/issues/124505,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-09-12T21:28:50Z +issue,131263,An API to re-enable dynamo-disabled functions,https://github.com/pytorch/pytorch/issues/131263,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-09-12T21:26:01Z +issue,133944,"torch.compile ""run at compile time"" decorator.",https://github.com/pytorch/pytorch/issues/133944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-09-12T21:19:44Z +issue,156308,Composition of nested `torch.compile` calls is not well defined,https://github.com/pytorch/pytorch/issues/156308,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-09-12T20:11:51Z +issue,161886,Directly monkey patching torch.Tensor.__getitem__ breaks tensor construction,https://github.com/pytorch/pytorch/issues/161886,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-09-12T16:35:13Z +issue,136622,torch.onnx.errors.UnsupportedOperatorError: Exporting the operator 'aten::linalg_inv' to ONNX opset version 18 is not supported,https://github.com/pytorch/pytorch/issues/136622,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-09-12T14:03:48Z +issue,126881,[DCP] DCP does not support objects which are lazy initialized.,https://github.com/pytorch/pytorch/issues/126881,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed checkpointing,scan when time allows,2025-09-12T03:41:48Z +issue,90560,"[discussion, idea] Batched, vectorized base64 decoding / encoding + maybe RLE decoding / encoding",https://github.com/pytorch/pytorch/issues/90560,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: vision",scan when time allows,2025-09-12T01:20:45Z +issue,162711,`scatter` / `gather` index type unclear,https://github.com/pytorch/pytorch/issues/162711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-09-11T22:04:48Z +issue,162712,FPE in torch.range,https://github.com/pytorch/pytorch/issues/162712,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2025-09-11T21:44:42Z +issue,127500,Avoid Having to Register Op For ExternKernelChoice of Aten Refs,https://github.com/pytorch/pytorch/issues/127500,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-09-11T17:34:03Z +issue,162352,Hiding C/C++ symbols in `version_script.lds`,https://github.com/pytorch/pytorch/issues/162352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-09-11T15:39:54Z +issue,162362,the result is wrong for sdpa with sliding window attention,https://github.com/pytorch/pytorch/issues/162362,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: sdpa",scan when time allows,2025-09-11T15:36:15Z +issue,162482,[sparse] `torch._sparse_broadcast_to` creates false-positive coalesced tensors with inputs of shape `()`,https://github.com/pytorch/pytorch/issues/162482,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-09-11T15:34:52Z +issue,162582,calling _record_memory_history(enabled=None) before _dump_snapshot would clear the timestamp information,https://github.com/pytorch/pytorch/issues/162582,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-09-11T15:33:53Z +issue,162685,Prevent py::class_ with no pointer type from being used on structs that inherit from intrusive_ptr_target,https://github.com/pytorch/pytorch/issues/162685,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2025-09-11T15:14:22Z +issue,162674,"NotImplementedError: ""addmm_sparse_cuda"" not implemented for 'Half'",https://github.com/pytorch/pytorch/issues/162674,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: sparse",scan when time allows,2025-09-11T14:52:17Z +issue,77737,"torchrun leads to `ModuleNotFoundError: No module named 'tensorboard'`, but python -m torch.distributed.launch is ok",https://github.com/pytorch/pytorch/issues/77737,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: r2p,scan when time allows,2025-09-11T07:04:50Z +issue,160697,[DTensor] nn.Embedding Compile - Empty Masked Buffer from AOT Autograd,https://github.com/pytorch/pytorch/issues/160697,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: dynamo, oncall: distributed",scan when time allows,2025-09-10T22:09:46Z +issue,160240,Migrate test.sh into python based tests,https://github.com/pytorch/pytorch/issues/160240,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-09-10T17:03:21Z +issue,152335,pin_memory crashes for big tensors and leaks page locked memory,https://github.com/pytorch/pytorch/issues/152335,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2025-09-10T15:31:34Z +issue,162375,`torch.compile` fails with `pack_padded_sequence` + `pad_packed_sequence` (FakeTensor error),https://github.com/pytorch/pytorch/issues/162375,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-09-10T07:56:36Z +issue,145656,Stable C bindings for libtorch,https://github.com/pytorch/pytorch/issues/145656,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-09-10T03:55:11Z +issue,162458,torch.float8_e5m2 unique() functionality support,https://github.com/pytorch/pytorch/issues/162458,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2025-09-09T21:01:37Z +issue,162182,CI jobs for x86_64 and x86_zen (AMD CPU) are overwriting each other,https://github.com/pytorch/pytorch/issues/162182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: correctness (silent)",scan when time allows,2025-09-09T20:20:22Z +issue,141984,CI feature mirroring or caching for Docker Hub images,https://github.com/pytorch/pytorch/issues/141984,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-09-09T17:44:24Z +issue,141235,Enable CI and Build Support for PyTorch on PPC64LE Architecture,https://github.com/pytorch/pytorch/issues/141235,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2025-09-09T13:20:34Z +issue,113222,EPOCH_DEPRECATION_WARNING in ChainedScheduler.step,https://github.com/pytorch/pytorch/issues/113222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2025-09-09T02:47:47Z +issue,67958,The `SequentialLR` scheduler uses a deprecated pattern,https://github.com/pytorch/pytorch/issues/67958,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer, module: tests",scan when time allows,2025-09-09T02:45:31Z +issue,162398,Unexpected variable type when parsing YAML in torchgen,https://github.com/pytorch/pytorch/issues/162398,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2025-09-08T23:19:13Z +issue,162280,libcufile symbol mismatch error in CUDA 13.0 libtorch build,https://github.com/pytorch/pytorch/issues/162280,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-09-08T22:50:36Z +issue,101930,[feature request] [minor] Inplace torch.flip_,https://github.com/pytorch/pytorch/issues/101930,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-09-08T20:50:13Z +issue,162259,[Triton] Support multi-GPU with different types of GPUs (4090s + 5090s on same machine),https://github.com/pytorch/pytorch/issues/162259,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-09-08T19:07:46Z +issue,156846,Can't alter the learning rate though loading optimizer states,https://github.com/pytorch/pytorch/issues/156846,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2025-09-08T14:34:38Z +issue,152992,conv2d with int8 on CUDA: GET was unable to find an engine to execute this computation,https://github.com/pytorch/pytorch/issues/152992,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda",scan when time allows,2025-09-08T14:17:20Z +issue,2569,Add a [strong] hash function for tensor data,https://github.com/pytorch/pytorch/issues/2569,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-09-08T05:33:39Z +issue,69687,Performance improvement in Autograd Forward AD using ZeroTensors,https://github.com/pytorch/pytorch/issues/69687,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, module: performance",scan when time allows,2025-09-07T19:54:13Z +issue,161970,Confusion behavior of PyTorch when combining boolean indexing with integer indexing,https://github.com/pytorch/pytorch/issues/161970,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: numpy",scan when time allows,2025-09-07T05:08:10Z +issue,160937,Module produced by export.unflatten is not runnable for Qwen3 and Llama3,https://github.com/pytorch/pytorch/issues/160937,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-09-06T04:43:33Z +issue,162283,"Failure after building torch with ""pip install -e . -v --no-build-isolation""",https://github.com/pytorch/pytorch/issues/162283,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: functorch",scan when time allows,2025-09-05T19:53:40Z +issue,157139,Distribute ARM Images,https://github.com/pytorch/pytorch/issues/157139,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: cuda, module: docker",scan when time allows,2025-09-05T18:24:06Z +issue,24005,Using `torch.utils.checkpoint.checkpoint_sequential` and `torch.autograd.grad` breaks when used in combination with `DistributedDataParallel`,https://github.com/pytorch/pytorch/issues/24005,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, oncall: distributed",scan when time allows,2025-09-05T08:20:54Z +issue,162127,[CUDA] Need support for Cuda 13.0,https://github.com/pytorch/pytorch/issues/162127,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-09-04T23:44:10Z +issue,159667,"Non-tensor auxiliary types don't work in torch.func.{grad, value_and_grad}",https://github.com/pytorch/pytorch/issues/159667,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functional ux, module: functorch",scan when time allows,2025-09-04T19:21:02Z +issue,162198,[Feature Request][Inductor] Support ops.store in ops.masked,https://github.com/pytorch/pytorch/issues/162198,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-09-04T18:53:30Z +issue,161621,evaluation forward() of Transformer model is slower than training forward() while autocast,https://github.com/pytorch/pytorch/issues/161621,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: performance",scan when time allows,2025-09-04T18:39:28Z +issue,161937,AOTAutograd runtime alias reconstruction applies view on non‑contiguous base under torch.compile (inductor + DDP): “Use .reshape(...) instead”. pre_grad_custom_pass and joint_custom_post_pass can’t override,https://github.com/pytorch/pytorch/issues/161937,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: distributed, oncall: pt2",scan when time allows,2025-09-04T18:06:59Z +issue,161509,Update install commands in c++ Debugging,https://github.com/pytorch/pytorch/issues/161509,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-09-04T13:25:01Z +issue,161778,RuntimeError: ptrDesc->finalize() with conv1d,https://github.com/pytorch/pytorch/issues/161778,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda",scan when time allows,2025-09-04T13:22:06Z +issue,161869,Add LazyGroupNorm,https://github.com/pytorch/pytorch/issues/161869,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2025-09-04T13:19:38Z +issue,162004,UBSAN appears to be active for ATen but not c10,https://github.com/pytorch/pytorch/issues/162004,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2025-09-04T13:19:11Z +issue,139761,[RFC] PyTorch - PyPi PEP-759 proposal (wheel-next),https://github.com/pytorch/pytorch/issues/139761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-09-03T15:41:24Z +issue,144931,"Error loading ""torch\lib\aoti_custom_ops.dll"" or one of its dependencies, when importing Torch, when building from Source on Windows 11 with cuDNN.",https://github.com/pytorch/pytorch/issues/144931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: windows",scan when time allows,2025-09-03T03:19:40Z +issue,161960,WeightsUnpickler error: Trying to call reduce for unrecognized function ,https://github.com/pytorch/pytorch/issues/161960,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-09-02T23:31:27Z +issue,155332,compiler cache not work,https://github.com/pytorch/pytorch/issues/155332,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-09-02T15:18:12Z +issue,158793,[RFC]: Iterations of DeviceMesh based on recent user requests/feedback,https://github.com/pytorch/pytorch/issues/158793,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-09-02T13:21:10Z +issue,159015,[RFC] Support Fake PG mode for the device mesh and provide an option to reuse PG under the hood,https://github.com/pytorch/pytorch/issues/159015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-09-02T13:21:08Z +issue,159014,[RFC] Make it easy to subclass DeviceMesh and also working in multi-threading setup,https://github.com/pytorch/pytorch/issues/159014,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-09-02T13:21:08Z +issue,159017,[RFC] Support the bookkeeping of different PGs beyond World_ and group_name,https://github.com/pytorch/pytorch/issues/159017,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-09-02T13:21:07Z +issue,159018,[RFC] Support more flexible pg option when it comes to submesh creation,https://github.com/pytorch/pytorch/issues/159018,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-09-02T13:21:06Z +issue,159019,[RFC] Support shorter print form for non-debugging mode,https://github.com/pytorch/pytorch/issues/159019,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-09-02T13:21:05Z +issue,155798,"`nn.RNN(...).to('cuda')` fails with `cuDNN error: CUDNN_STATUS_BAD_PARAM` on GPU, but works on CPU",https://github.com/pytorch/pytorch/issues/155798,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2025-09-02T06:27:58Z +issue,161853,torch.nn.FractionalMaxPool3d occurs bug with big kernel_size,https://github.com/pytorch/pytorch/issues/161853,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking",scan when time allows,2025-09-01T14:01:57Z +issue,131196,Missing float8 storage,https://github.com/pytorch/pytorch/issues/131196,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2025-09-01T03:39:04Z +issue,142840,Document spectral norm non-support on non-diagonalizable inputs.,https://github.com/pytorch/pytorch/issues/142840,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: nn.utils.parametrize",scan when time allows,2025-08-31T18:09:06Z +issue,159870,`torch.nn.functional.sigmoid` produces inconsistent results on different device for complex inputs,https://github.com/pytorch/pytorch/issues/159870,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cpu, module: nans and infs, module: nn",scan when time allows,2025-08-31T15:47:52Z +issue,119420,torch.stft sometimes raises RuntimeError: cuFFT error: CUFFT_INTERNAL_ERROR on low free memory,https://github.com/pytorch/pytorch/issues/119420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: fft, module: memory usage",scan when time allows,2025-08-31T02:19:50Z +issue,63291,[typing] new_ones has wrong signature,https://github.com/pytorch/pytorch/issues/63291,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: typing",scan when time allows,2025-08-30T23:01:20Z +issue,63305,Alternate Model Loading - Android,https://github.com/pytorch/pytorch/issues/63305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2025-08-29T14:27:31Z +issue,58393,"In Android, the memory used by the tensor or model cannot be recycled. Is there any way to solve it?",https://github.com/pytorch/pytorch/issues/58393,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2025-08-29T14:27:19Z +issue,129358,Pytorch build fail with GCC 14.1.0 due to third_party/fbgemm/src/UtilsAvx512.cc:970:35: error: ‘r’ may be used uninitialized [-Werror=maybe-uninitialized],https://github.com/pytorch/pytorch/issues/129358,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-08-29T12:01:51Z +issue,161283,Implement half-vs-float option for MPS output match tests,https://github.com/pytorch/pytorch/issues/161283,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: tests",scan when time allows,2025-08-28T20:02:12Z +issue,114299,[RFC] Per-Parameter-Sharding FSDP,https://github.com/pytorch/pytorch/issues/114299,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2025-08-28T02:35:24Z +issue,147001,Preload CUDA fails if CUDA libs in different PYTHONPATH,https://github.com/pytorch/pytorch/issues/147001,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: third_party",scan when time allows,2025-08-27T17:11:29Z +issue,161357,`reshape` support for sparse tensors (needed for `einsum`),https://github.com/pytorch/pytorch/issues/161357,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-08-27T16:20:37Z +issue,159906,New at::HostAllocator interface prevents using more than one allocator implementation for a device type,https://github.com/pytorch/pytorch/issues/159906,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-08-26T23:22:14Z +issue,161281,There should be an easy way in CI to disable individual failing inductor-periodic tests,https://github.com/pytorch/pytorch/issues/161281,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: pt2",scan when time allows,2025-08-26T20:16:40Z +issue,60466,PyTorch unfold could be faster,https://github.com/pytorch/pytorch/issues/60466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2025-08-26T19:17:21Z +issue,161108,AoTInductor gives segmentation fault with torchbind registration,https://github.com/pytorch/pytorch/issues/161108,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, module: torchbind, oncall: export, oncall: pt2",scan when time allows,2025-08-26T18:15:02Z +issue,161524,boolean index_add: Discrepancy between CPU and MPS backend,https://github.com/pytorch/pytorch/issues/161524,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: mps",scan when time allows,2025-08-26T18:11:35Z +issue,161287,Add partial shape inference support when creating a `view()` of a Nested Jagged Tensor,https://github.com/pytorch/pytorch/issues/161287,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-08-26T13:18:59Z +issue,161163,Dropout performance degradation after version 2.6.0,https://github.com/pytorch/pytorch/issues/161163,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: intel, module: performance, module: random",scan when time allows,2025-08-26T07:13:29Z +issue,161296,Make `torch.func.functional_call` compatible with params and buffer input as `tensordict.TensorDict`,https://github.com/pytorch/pytorch/issues/161296,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-08-26T01:54:04Z +issue,161365,orch.onnx.errors.UnsupportedOperatorError: Exporting the operator 'aten::linalg_inv' to ONNX opset version 16 is not supported.,https://github.com/pytorch/pytorch/issues/161365,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-08-25T22:48:07Z +issue,131947,`__torch_function__` does not work for functions that called within other overrided functions,https://github.com/pytorch/pytorch/issues/131947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2025-08-24T08:17:58Z +issue,160071,[TorchDynamo] excessive stack use: stack is 3286 deep on A100 with AMD EPYC CPU,https://github.com/pytorch/pytorch/issues/160071,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-08-23T13:30:03Z +issue,93982,`linalg.lstsq` fails the gradient computation in forward mode,https://github.com/pytorch/pytorch/issues/93982,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2025-08-23T09:52:20Z +issue,160508,`torch.compile` + `torch.vmap` + `torch.func.jacfwd` results in an error,https://github.com/pytorch/pytorch/issues/160508,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: functorch, module: vmap, oncall: pt2",scan when time allows,2025-08-23T05:48:59Z +issue,159914,`test_non_contiguous_input_mm_plus_mm` looks broken on B200,https://github.com/pytorch/pytorch/issues/159914,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tf32",scan when time allows,2025-08-22T23:02:14Z +issue,108567,Allow slicing of Nested Tensors along constant dimensions,https://github.com/pytorch/pytorch/issues/108567,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-08-22T19:17:04Z +issue,99556,torch.func.jacrev fails if model contains full_backward_hook,https://github.com/pytorch/pytorch/issues/99556,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2025-08-22T14:43:15Z +issue,143466,Support Dict Parameter Type for custom_op,https://github.com/pytorch/pytorch/issues/143466,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-08-22T13:40:07Z +issue,126913,Alternative access order for the same buffer can bring big perf win,https://github.com/pytorch/pytorch/issues/126913,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-08-22T13:17:24Z +issue,160359,Use `tf32x3` through Inductor,https://github.com/pytorch/pytorch/issues/160359,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-08-22T04:04:52Z +issue,161068,[ATen] Linkage error with TensorBase::data_ptr and TensorBase::mutable_data_ptr,https://github.com/pytorch/pytorch/issues/161068,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-08-21T03:38:10Z +issue,160891,Please add `torch._grouped_mm` on SM_120 (RTX 6000 Blackwell) hardware,https://github.com/pytorch/pytorch/issues/160891,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-08-20T22:47:35Z +issue,160629,torch.arange not implemented for complex numbers,https://github.com/pytorch/pytorch/issues/160629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2025-08-20T19:51:40Z +issue,161029,MPS does not support deterministic index operations,https://github.com/pytorch/pytorch/issues/161029,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: mps",scan when time allows,2025-08-20T02:36:16Z +issue,161001,Add some way to associate functional custom operators and their in-place out variants,https://github.com/pytorch/pytorch/issues/161001,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: inductor, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-08-20T02:33:44Z +issue,160856,Discrepancy between PyTorch and TensorRT when casting from FP32 to BF16 causes significant accuracy mismatch,https://github.com/pytorch/pytorch/issues/160856,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: onnx",scan when time allows,2025-08-19T20:39:49Z +issue,160167,"libtorch: CMake 4.1: Error evaluating generator expression: TARGET_PROPERTY:torch,INTERFACE_LINK_LIBRARIES",https://github.com/pytorch/pytorch/issues/160167,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions, module: windows",scan when time allows,2025-08-19T17:58:01Z +issue,157627,Regression: torch.distributed.gather_object segfaults,https://github.com/pytorch/pytorch/issues/157627,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-08-19T13:28:48Z +issue,125070,2.2.0+ regresses SDPA performance on Windows,https://github.com/pytorch/pytorch/issues/125070,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sdpa, module: windows",scan when time allows,2025-08-19T11:41:20Z +issue,158444,Proposal to introduce CUDA streams in BatchLinearAlgebraLib.cpp,https://github.com/pytorch/pytorch/issues/158444,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2025-08-19T07:26:09Z +issue,160337,"[RFC] Simplify bookkeeping of DeviceMesh slicing, (un)flattening, ...",https://github.com/pytorch/pytorch/issues/160337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2025-08-18T21:02:17Z +issue,159769,Inductor codegen for float8 dynamic quantization ops for scaled_grouped_mm backward pass is slow,https://github.com/pytorch/pytorch/issues/159769,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-08-18T19:39:25Z +issue,145471,Is there a PyTorch version that can work properly on the Thor platform based on the Blackwell architecture?,https://github.com/pytorch/pytorch/issues/145471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-08-18T07:31:00Z +issue,160517,Enhance output layouts of all-sparse matrix multiplications,https://github.com/pytorch/pytorch/issues/160517,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-08-16T11:34:08Z +issue,156873,[ONNX] Missing repeat_interleave for dynamo=True,https://github.com/pytorch/pytorch/issues/156873,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-08-16T00:38:23Z +issue,156745,[v.2.8.0] Release Tracker,https://github.com/pytorch/pytorch/issues/156745,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-08-14T17:56:02Z +issue,160405,[Expandable block] how to get the best-fit free block,https://github.com/pytorch/pytorch/issues/160405,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-08-14T05:24:01Z +issue,159781,MacOS Python 3.13t wheel is build for macOS 14.0+ ARM64 while Py3.9-3.13 wheels are for macOS 11.0+ ARM64,https://github.com/pytorch/pytorch/issues/159781,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, oncall: releng",scan when time allows,2025-08-13T18:24:11Z +issue,153919,FPE in `torch.remainder`,https://github.com/pytorch/pytorch/issues/153919,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: edge cases, module: python frontend",scan when time allows,2025-08-13T18:00:16Z +issue,128323,UnsupportedOperatorError: Exporting the operator 'quantized::batch_norm2d' to ONNX opset version 14 is not supported.,https://github.com/pytorch/pytorch/issues/128323,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-08-13T10:15:19Z +issue,159784,Immutable assignment (akin to `array.at[...]` in JAX),https://github.com/pytorch/pytorch/issues/159784,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: functorch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-08-13T07:15:51Z +issue,159947,Add official statically linked libtorch libraries,https://github.com/pytorch/pytorch/issues/159947,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docs, module: nativert",scan when time allows,2025-08-13T07:06:24Z +issue,117884,Foreach Where support,https://github.com/pytorch/pytorch/issues/117884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer, module: pt2 optimizer, oncall: pt2",scan when time allows,2025-08-13T03:12:20Z +issue,160308,Reserved memory is much bigger than allocated memory in multi-stream scenario,https://github.com/pytorch/pytorch/issues/160308,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-08-13T02:53:47Z +issue,159932,Cannot view dtype of tensor which is contiguous due to numel=1 but has non-standard striding,https://github.com/pytorch/pytorch/issues/159932,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-08-13T02:52:38Z +issue,160387,DISABLED RecordDebugHandles.Basic (__main__.test_jit),https://github.com/pytorch/pytorch/issues/160387,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: unknown",scan when time allows,2025-08-13T02:48:03Z +issue,160122,Add tensor-aware put_signal & wait_signal Triton wrappers to NVSHMEM Triton Kernels,https://github.com/pytorch/pytorch/issues/160122,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-08-13T02:47:28Z +issue,159866,BE: enable auto reformat List to list and similar typing,https://github.com/pytorch/pytorch/issues/159866,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: lint",scan when time allows,2025-08-12T20:26:31Z +issue,159680,[Benchmark Dashboard] Kernel-level benchmark,https://github.com/pytorch/pytorch/issues/159680,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: benchmark,scan when time allows,2025-08-12T20:15:59Z +issue,159676,[Benchmark Dashboard] Track quarterly change,https://github.com/pytorch/pytorch/issues/159676,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: benchmark,scan when time allows,2025-08-12T20:14:54Z +issue,152134,[RFC] Proposed Changes to Feature Tracking & Classification for PyTorch Releases starting Release 2.8,https://github.com/pytorch/pytorch/issues/152134,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-08-12T19:35:59Z +issue,160415,[MPS] Conv3d type mismatch with AMP,https://github.com/pytorch/pytorch/issues/160415,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: mps",scan when time allows,2025-08-12T19:10:28Z +issue,155011,Please support libtorch for XPU,https://github.com/pytorch/pytorch/issues/155011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: xpu",scan when time allows,2025-08-12T06:37:25Z +issue,72775,[vulkan] Vulkan backend fails creating tensor on x86_64 Linux,https://github.com/pytorch/pytorch/issues/72775,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2025-08-11T22:50:50Z +issue,102319,Re-enable Low Memory Dropout,https://github.com/pytorch/pytorch/issues/102319,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-08-11T16:53:28Z +issue,160142,"Add a ""max-performance"" mode to torch.compile for aggressive optimization",https://github.com/pytorch/pytorch/issues/160142,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-08-11T15:37:35Z +issue,159942,Is is possible for pytorch pipeline parallelism to support dynamic shapes(input/output),https://github.com/pytorch/pytorch/issues/159942,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: distributed, oncall: pt2",scan when time allows,2025-08-11T15:34:54Z +issue,160281,Improve CUDAGraph Tree heuristics on starting new generation,https://github.com/pytorch/pytorch/issues/160281,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2",scan when time allows,2025-08-10T22:06:27Z +issue,93859,Bitwise-perfect method for (de)serializing tensors in base64,https://github.com/pytorch/pytorch/issues/93859,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-08-10T18:53:10Z +issue,60832,State of Torch Named Tensors,https://github.com/pytorch/pytorch/issues/60832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2025-08-10T13:24:29Z +issue,119194,torch_function objects passed as non-Tensor args should trigger overrides,https://github.com/pytorch/pytorch/issues/119194,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: dynamic shapes",scan when time allows,2025-08-10T13:23:18Z +issue,156959,Gross mismatch in PDF between CUDA and CPU for multivariate Gaussian mixture models,https://github.com/pytorch/pytorch/issues/156959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: distributions",scan when time allows,2025-08-09T05:25:38Z +issue,68959,Deterministic implementation for grid_sampler_2d_backward_cuda,https://github.com/pytorch/pytorch/issues/68959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism",scan when time allows,2025-08-08T22:13:24Z +issue,160225,Building cuda kernels with debug information causes some cuda kernel launches to fail due to capacity constraints,https://github.com/pytorch/pytorch/issues/160225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: debug-build, module: third_party",scan when time allows,2025-08-08T21:54:45Z +issue,84530,`tensordot` not working for dtype int32 and lower when there is only 1 element in the given axis,https://github.com/pytorch/pytorch/issues/84530,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-08-08T08:40:20Z +issue,101188,Weird dataloader performance degradation caused by torch and numpy import order,https://github.com/pytorch/pytorch/issues/101188,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: openmp,scan when time allows,2025-08-08T05:35:00Z +issue,137389,non-deterministic issue of torch.einsum function on different GPU.,https://github.com/pytorch/pytorch/issues/137389,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: numerical-reproducibility",scan when time allows,2025-08-07T21:49:21Z +issue,159852,`torch.cond()` behaves inconsistently when using symbolic predicate,https://github.com/pytorch/pytorch/issues/159852,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-08-07T17:03:06Z +issue,159848,"`tree_flatten(some_pytree, lambda x: isinstance(x, Tensor))` yields `[None], _`",https://github.com/pytorch/pytorch/issues/159848,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pytree,scan when time allows,2025-08-07T13:57:50Z +issue,57185,Segmentation fault with ITIMER_REAL,https://github.com/pytorch/pytorch/issues/57185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: macos,scan when time allows,2025-08-07T05:09:56Z +issue,121969,"which Pythorch suports ""TORCH_USE_CUDA_DSA=1"" from the shell environment?",https://github.com/pytorch/pytorch/issues/121969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-08-06T20:44:03Z +issue,159622,Performance bug on `inplace` of `nn.ELU`,https://github.com/pytorch/pytorch/issues/159622,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: nn",scan when time allows,2025-08-06T19:19:14Z +issue,138831,Add mean and var operation for Nested Tensors,https://github.com/pytorch/pytorch/issues/138831,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-08-06T17:40:31Z +issue,157950,[RFC]: PyTorch Low-Precision GEMMs Public API,https://github.com/pytorch/pytorch/issues/157950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: floatx (formerly float8), module: rocm",scan when time allows,2025-08-06T17:39:27Z +issue,159886,torch._dynamo documentation,https://github.com/pytorch/pytorch/issues/159886,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: dynamo, oncall: pt2",scan when time allows,2025-08-06T15:10:43Z +issue,159838,C++23 forward declaration error :`invalid application of 'sizeof' to an incomplete type 'torch::_export::Graph'`,https://github.com/pytorch/pytorch/issues/159838,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: nativert",scan when time allows,2025-08-06T14:13:32Z +issue,158783,[BE] linter to detect unused docker images,https://github.com/pytorch/pytorch/issues/158783,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2025-08-05T20:08:22Z +issue,153238,[NJT] can only chunk if the 2nd dimension is ragged,https://github.com/pytorch/pytorch/issues/153238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-08-05T18:37:20Z +issue,159739,torch.empty should consider np.bool while parsing args,https://github.com/pytorch/pytorch/issues/159739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-08-05T18:09:56Z +issue,145094,Tracking issue: Incorrect Meta Strides / Turn On PyDispatcher in FakeTensor Mode,https://github.com/pytorch/pytorch/issues/145094,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, module: faketensor, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-08-05T16:46:30Z +issue,157445,DDP+TP composition does not work as expected,https://github.com/pytorch/pytorch/issues/157445,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-08-05T14:18:47Z +issue,155997,"ONNX export via Dynamo sets `dft_length = 1` in `DFT`, breaking shape-inference for `torch.fft.rfft`",https://github.com/pytorch/pytorch/issues/155997,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-08-05T09:29:11Z +issue,16375,Python-bound C++ frontend modules don't handle attributes well,https://github.com/pytorch/pytorch/issues/16375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-08-05T00:03:30Z +issue,117394,[RFC] Dynamo Single Step Graph,https://github.com/pytorch/pytorch/issues/117394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: pt2",scan when time allows,2025-08-04T23:21:04Z +issue,127973,Torch compile OpenMP error with simple nn.Module,https://github.com/pytorch/pytorch/issues/127973,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, module: openmp, oncall: pt2",scan when time allows,2025-08-04T19:41:52Z +issue,124749,Improve Dynamo (and other) flaky tests,https://github.com/pytorch/pytorch/issues/124749,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: tests, oncall: pt2",scan when time allows,2025-08-04T18:39:20Z +issue,128071,[user empathy day 2][based] torch.compile issues,https://github.com/pytorch/pytorch/issues/128071,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, oncall: pt2",scan when time allows,2025-08-04T18:32:03Z +issue,159301,#158878 seems to break `test_mempool_empty_cache`,https://github.com/pytorch/pytorch/issues/159301,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2025-08-04T17:59:54Z +issue,115526,Can PyTorch support using custom compilers for model's components and the Inductor compiler for the model?,https://github.com/pytorch/pytorch/issues/115526,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-08-04T16:46:21Z +issue,159564,Re-enable inactive_split_bytes when expandable_segments:True,https://github.com/pytorch/pytorch/issues/159564,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-08-04T15:51:07Z +issue,153933,PyTorch build fail with GCC 15.1.0 due to third_party/gloo/gloo/types.h:66:11: error: ‘uint8_t’ does not name a type,https://github.com/pytorch/pytorch/issues/153933,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-08-04T14:58:50Z +issue,159740,Better error message when requesting pinned memory on cpu-only system,https://github.com/pytorch/pytorch/issues/159740,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2025-08-04T01:40:19Z +issue,142290,RFC: Display Supported Versions in API Documentation#,https://github.com/pytorch/pytorch/issues/142290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-08-02T05:14:49Z +issue,141210,Addmm is not always better than add + mm,https://github.com/pytorch/pytorch/issues/141210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: dependency bug, module: performance",scan when time allows,2025-08-02T04:15:08Z +issue,159559,Support NCCL register userbuffer for P2P communication in NCCL backend,https://github.com/pytorch/pytorch/issues/159559,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2025-08-02T02:12:31Z +issue,129668,TensorDict - torch.compile issue tracker,https://github.com/pytorch/pytorch/issues/129668,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-08-01T23:34:43Z +issue,112901,Tracking: Dynamo Tracing Improvements,https://github.com/pytorch/pytorch/issues/112901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-08-01T21:38:32Z +issue,158187,"[torch.save] Integrate FastPersist by allowing f (a.k.a., file-like object) to save storage objects",https://github.com/pytorch/pytorch/issues/158187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-08-01T15:23:08Z +issue,105859,Report model flop utilization (mfu) in benchmark,https://github.com/pytorch/pytorch/issues/105859,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-08-01T02:44:57Z +issue,106269,Make our source attribution debug prints more useful for Compiler Explorer,https://github.com/pytorch/pytorch/issues/106269,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-08-01T02:44:46Z +issue,116575,Reland guarded config,https://github.com/pytorch/pytorch/issues/116575,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-08-01T02:42:30Z +issue,102853,pdb but for dynamo (and time travel debugging),https://github.com/pytorch/pytorch/issues/102853,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-08-01T02:39:21Z +issue,159536,Document NCCL Symmetric memory,https://github.com/pytorch/pytorch/issues/159536,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nccl",scan when time allows,2025-07-31T23:17:36Z +issue,31837,logsumexp: two little-impact perf suggestions (important because logsumexp is used for optimizing/fusing cross_entropy over large vocabs),https://github.com/pytorch/pytorch/issues/31837,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: reductions",scan when time allows,2025-07-31T21:47:45Z +issue,159510,Add more tests to h100 ciflow,https://github.com/pytorch/pytorch/issues/159510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: pt2",scan when time allows,2025-07-31T17:17:21Z +issue,153745,[RFC] [Feature] Intra-Device Heterogeneous Memory Allocation Support,https://github.com/pytorch/pytorch/issues/153745,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-31T16:19:57Z +issue,156694,"Vendored wheels on PyTorch pip repository are outdated (e.g., `cmake`, `certifi`)",https://github.com/pytorch/pytorch/issues/156694,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: releng",scan when time allows,2025-07-31T07:44:54Z +issue,159354,distributed/rpc/test_share_memory fails with timeouts on 64bit BAR systems,https://github.com/pytorch/pytorch/issues/159354,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-07-31T07:06:23Z +issue,157606,Both DTensor TP and SP are missing the last collective in the backward pass,https://github.com/pytorch/pytorch/issues/157606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-07-30T23:10:51Z +issue,159498,[FSDP2] provide profiler metadata for memory bandwidth calculation,https://github.com/pytorch/pytorch/issues/159498,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-07-30T18:39:52Z +issue,159353,torch.cond triggers shape mismatch with linear layer when nested in custom model,https://github.com/pytorch/pytorch/issues/159353,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, oncall: pt2",scan when time allows,2025-07-30T18:11:12Z +issue,159384,Graph partition performance,https://github.com/pytorch/pytorch/issues/159384,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: inductor, oncall: pt2",scan when time allows,2025-07-30T17:20:40Z +issue,159124,AOTInductor preallocate output tensors,https://github.com/pytorch/pytorch/issues/159124,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",scan when time allows,2025-07-30T15:39:25Z +issue,159333,torch._foreach_copy_() copy all nan as same number,https://github.com/pytorch/pytorch/issues/159333,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: mta",scan when time allows,2025-07-30T15:30:58Z +issue,106885,"[bug] Dynamo graph break when using pyton module `heapq` (manipulates with `list`s), although succeeds when placing `heapq.py` near the test script",https://github.com/pytorch/pytorch/issues/106885,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-30T13:52:29Z +issue,159149,"❌ Error generating speech: NVML_SUCCESS == r INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":995, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/159149,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-07-30T13:26:17Z +issue,159363,pad_sequence overflows when padding with larger value than int64.max - 512,https://github.com/pytorch/pytorch/issues/159363,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: python frontend",scan when time allows,2025-07-30T13:24:01Z +issue,158895,Update docs about running lint,https://github.com/pytorch/pytorch/issues/158895,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: lint",scan when time allows,2025-07-29T20:33:26Z +issue,158876,UNSTABLE pull / linux-jammy-py3_9-clang9-xla / test (xla),https://github.com/pytorch/pytorch/issues/158876,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: xla",scan when time allows,2025-07-29T20:30:26Z +issue,157379,[CI] Need better way to detect OOMs especially on pet instances,https://github.com/pytorch/pytorch/issues/157379,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: flaky-tests",scan when time allows,2025-07-29T20:05:38Z +issue,81565,__getitem__ is returned as an OverloadPacket instead of an OpOverload in __torch_dispatch__,https://github.com/pytorch/pytorch/issues/81565,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2025-07-29T14:19:25Z +issue,81717,PyTorch Embedding Op with max_norm is not working as expected,https://github.com/pytorch/pytorch/issues/81717,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: embedding, module: norms and normalization",scan when time allows,2025-07-29T14:19:24Z +issue,91251,Potential bug found with pybind11 dec_ref while gil released,https://github.com/pytorch/pytorch/issues/91251,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2025-07-29T14:19:21Z +issue,91508,Update map_nt to take into account size and strides,https://github.com/pytorch/pytorch/issues/91508,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-07-29T14:19:20Z +issue,93454,MMDet 3.x cannot run successfully in inductor mode,https://github.com/pytorch/pytorch/issues/93454,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:16Z +issue,93455,Performance regression on interpolation in Kornia,https://github.com/pytorch/pytorch/issues/93455,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:15Z +issue,93456,Way to run accuracy minifier on only one particular subgraph,https://github.com/pytorch/pytorch/issues/93456,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:14Z +issue,93463,Traceable tensor subclasses cannot actually be used with AOTAutograd,https://github.com/pytorch/pytorch/issues/93463,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:13Z +issue,93464,wav2vec2 model: error trying to do inference,https://github.com/pytorch/pytorch/issues/93464,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:12Z +issue,93465,Graph breaks with HuggingFace Stable Diffusion,https://github.com/pytorch/pytorch/issues/93465,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:11Z +issue,93480,Umbrella issue for only populate real_value_cache in export test suite fails,https://github.com/pytorch/pytorch/issues/93480,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-07-29T14:19:09Z +issue,93486,A Simple Function Causing Graph Break,https://github.com/pytorch/pytorch/issues/93486,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:08Z +issue,93488,"If minifier test fails, stderr/stdout of subprocess calls is not printed",https://github.com/pytorch/pytorch/issues/93488,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:07Z +issue,93580,AssertionError: Unknown expression s2,https://github.com/pytorch/pytorch/issues/93580,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:06Z +issue,93511,support setattr of arbitrary user provided types in tracing,https://github.com/pytorch/pytorch/issues/93511,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-07-29T14:19:06Z +issue,93592,Turning on minifier causes bug to go away (on DebertaForMaskedLM),https://github.com/pytorch/pytorch/issues/93592,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:05Z +issue,93596,Minifier doesn't work on DebertaForQuestionAnswering,https://github.com/pytorch/pytorch/issues/93596,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:04Z +issue,93602,Diffuser pipeline device attribute broken when using optimized model,https://github.com/pytorch/pytorch/issues/93602,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:03Z +issue,93604,TorchBench - moco - RuntimeError: Tensors must be CUDA and dense,https://github.com/pytorch/pytorch/issues/93604,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:02Z +issue,93609,Don't store example_value on FX node meta,https://github.com/pytorch/pytorch/issues/93609,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:19:01Z +issue,94183,Addition of batch CSR tensors produces incorrect and invalid CSR tensor,https://github.com/pytorch/pytorch/issues/94183,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: sparse",scan when time allows,2025-07-29T14:19:00Z +issue,94186,Addition of hybrid CSR tensors produces incorrect and invalid CSR tensor,https://github.com/pytorch/pytorch/issues/94186,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: sparse",scan when time allows,2025-07-29T14:18:59Z +issue,97268,Desync debugger encounters traceMap error,https://github.com/pytorch/pytorch/issues/97268,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-07-29T14:18:58Z +issue,99999,Runtime Error,https://github.com/pytorch/pytorch/issues/99999,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:18:56Z +issue,101314,Shared library loading logic breaks when CUDA packages are installed in a non-standard location,https://github.com/pytorch/pytorch/issues/101314,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bazel,scan when time allows,2025-07-29T14:18:54Z +issue,102673,Fix dynamo-related debug Python 3.11 failures,https://github.com/pytorch/pytorch/issues/102673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python version,scan when time allows,2025-07-29T14:18:53Z +issue,103800,Mis-annotated return for `F._no_grad_embedding_renorm_` (also JIT related),https://github.com/pytorch/pytorch/issues/103800,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: typing, oncall: jit",scan when time allows,2025-07-29T14:18:52Z +issue,107381,sparse_mask method ignores masked-in elements of sparse compressed input tensors,https://github.com/pytorch/pytorch/issues/107381,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: sparse",scan when time allows,2025-07-29T14:18:51Z +issue,107451,Conversion from COO with two sparse dimensions to CSR with dense_dim specified fails,https://github.com/pytorch/pytorch/issues/107451,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-07-29T14:18:50Z +issue,107478,Adding batched CSR tensors with different sparsities produces an invalid tensor,https://github.com/pytorch/pytorch/issues/107478,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: sparse",scan when time allows,2025-07-29T14:18:49Z +issue,108984,PPC64le: GCC 11.2.1 Linker Error in bin/torch_shm_manager,https://github.com/pytorch/pytorch/issues/108984,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:18:48Z +issue,111962,[pytree] `pytree.tree_map` does not respect type of `torch.Size`,https://github.com/pytorch/pytorch/issues/111962,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pytree,scan when time allows,2025-07-29T14:18:47Z +issue,114859,PT2 QAT flow fails to get reasonable accuracy with mobilenet_v3_large,https://github.com/pytorch/pytorch/issues/114859,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pt2 accuracy, oncall: pt2, oncall: quantization",scan when time allows,2025-07-29T14:18:45Z +issue,116013,[Dynamo] bytecode transformed by Dynamo is not serializable by marshal,https://github.com/pytorch/pytorch/issues/116013,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-07-29T14:18:44Z +issue,118642,Different outputs of `div` and `threshold` with a specific input,https://github.com/pytorch/pytorch/issues/118642,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, oncall: pt2",scan when time allows,2025-07-29T14:18:43Z +issue,121220,Misleading `log_abs_det_jacobian` shape for `StackTransform`?,https://github.com/pytorch/pytorch/issues/121220,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-07-29T14:18:40Z +issue,121439,copy_() produces wrong results for boolean tensors on mps,https://github.com/pytorch/pytorch/issues/121439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-07-29T14:18:39Z +issue,121725,torch.logsumexp throws RuntimeError instead of reducing all dimensions when dim=None,https://github.com/pytorch/pytorch/issues/121725,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: named tensor, module: python frontend",scan when time allows,2025-07-29T14:18:37Z +issue,122529,Distributed checkpoint state_dict load may report nccl error and hide the real root-cause exception,https://github.com/pytorch/pytorch/issues/122529,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",scan when time allows,2025-07-29T14:18:35Z +issue,133010,torch.multiprocessing.start_processes is blocking with large input arguments,https://github.com/pytorch/pytorch/issues/133010,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2025-07-29T14:18:31Z +issue,144530,[Pipelining] PP+DDP does not work for Zero Bubble,https://github.com/pytorch/pytorch/issues/144530,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pipelining, oncall: distributed",scan when time allows,2025-07-29T14:18:26Z +issue,146945,New CachingAutotuner pickling logic may be brittle to triton upgrades,https://github.com/pytorch/pytorch/issues/146945,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-29T14:18:23Z +issue,147814,unbind_copy opinformation cause exception while running test_dtensor_ops.py,https://github.com/pytorch/pytorch/issues/147814,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-07-29T14:18:19Z +issue,148106,NotImplementedError: FlexAttentionHigherOrderVariable() has no type,https://github.com/pytorch/pytorch/issues/148106,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: flex attention, module: higher order operators, module: pt2-dispatcher",scan when time allows,2025-07-29T14:18:18Z +issue,152190,HUD Dashboard sort by perf speedup doesn't do anything,https://github.com/pytorch/pytorch/issues/152190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-07-29T14:18:14Z +issue,157054,NVFp4 Cublas Error,https://github.com/pytorch/pytorch/issues/157054,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: floatx (formerly float8)",scan when time allows,2025-07-29T14:18:10Z +issue,159138,Some bugs that related to `ReflectionPadNd` on input dim and padding,https://github.com/pytorch/pytorch/issues/159138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-07-29T14:18:09Z +issue,159164,Improvements to CUDACachingAllocator When expandable_segments Is Enabled,https://github.com/pytorch/pytorch/issues/159164,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-07-29T03:01:27Z +issue,159234,Lingering reference to setup.py in CONTRIBUTING.md,https://github.com/pytorch/pytorch/issues/159234,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-07-28T22:06:56Z +issue,153534,`pca_lowrank` on CUDA is significantly slower than on CPU,https://github.com/pytorch/pytorch/issues/153534,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2025-07-28T15:56:39Z +issue,159122,CI lint unexpected failure,https://github.com/pytorch/pytorch/issues/159122,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: typing",scan when time allows,2025-07-28T14:15:17Z +issue,73656,torchrun: Hostname/endpoint mismatch not handled,https://github.com/pytorch/pytorch/issues/73656,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: c10d,scan when time allows,2025-07-28T10:39:47Z +issue,150186,`MaskedTensor` 10x slower than `Tensor` compared with `nn.functional.softmax`,https://github.com/pytorch/pytorch/issues/150186,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: masked operators, module: performance",scan when time allows,2025-07-28T10:27:22Z +issue,38671,Add CUDA callback to Python API,https://github.com/pytorch/pytorch/issues/38671,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: rpc",scan when time allows,2025-07-28T10:06:55Z +issue,159230,[Precompile] Investigate recompile in NanoGPT training,https://github.com/pytorch/pytorch/issues/159230,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-27T20:33:03Z +issue,159229,[Precompile] Assertion error when loading package in Nanogpt training with caching precompile fails after second load,https://github.com/pytorch/pytorch/issues/159229,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-27T20:24:23Z +issue,159228,[Precompile] mark_static_address behavior,https://github.com/pytorch/pytorch/issues/159228,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-27T20:10:32Z +issue,61582,Many reduction operators do not support reducing over multiple dimensions,https://github.com/pytorch/pytorch/issues/61582,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python array api, module: reductions",scan when time allows,2025-07-27T16:26:16Z +issue,159156,Fully remove the protoc dependency in core,https://github.com/pytorch/pytorch/issues/159156,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: onnx, module: third_party",scan when time allows,2025-07-25T17:14:54Z +issue,158738,searchsorted handling of nan,https://github.com/pytorch/pytorch/issues/158738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-07-25T09:56:38Z +issue,158873,Erroneous Recompile Warnings with Multiple Models,https://github.com/pytorch/pytorch/issues/158873,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-07-24T20:37:01Z +issue,158969,Scaled dot product attention accepts head_num for attn_mask,https://github.com/pytorch/pytorch/issues/158969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-07-24T19:33:14Z +issue,158199,GPT-2 compiled model perf/accuracy,https://github.com/pytorch/pytorch/issues/158199,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps, module: performance, oncall: pt2",scan when time allows,2025-07-24T16:26:34Z +issue,157919,tighten pt2 FX graph invariants,https://github.com/pytorch/pytorch/issues/157919,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-07-24T14:07:37Z +issue,158936,TorchAO FP8 training + HF transformer model + torch.compile + DDP does not work,https://github.com/pytorch/pytorch/issues/158936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: floatx (formerly float8), oncall: distributed, oncall: pt2",scan when time allows,2025-07-24T01:04:25Z +issue,155325,Excessively restrictive dependencies,https://github.com/pytorch/pytorch/issues/155325,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: releng",scan when time allows,2025-07-23T22:23:36Z +issue,155066,Add extensions `flash_attention` and `vllm` as test of new PyTorch releases for known issues of compat of their binaries and of possibility of compiling these from source,https://github.com/pytorch/pytorch/issues/155066,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2025-07-23T22:23:32Z +issue,158871,torch.classes module doesn't set all module attributes,https://github.com/pytorch/pytorch/issues/158871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-07-23T20:08:12Z +issue,51511,Python-3.9: Importing numba and torch triggers internal error,https://github.com/pytorch/pytorch/issues/51511,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-07-23T20:07:46Z +issue,158819,[FSDP2] return original unsharded nn.Module from FSDPModule,https://github.com/pytorch/pytorch/issues/158819,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-07-23T15:20:17Z +issue,158401,torch.compile hits internal tuple index out of range inside subclass_meta.unwrapped_idx,https://github.com/pytorch/pytorch/issues/158401,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-07-23T13:36:34Z +issue,97432,suspicious memory leak when increase DataLoader's prefetch_factor and enable pin_memory,https://github.com/pytorch/pytorch/issues/97432,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: memory usage",scan when time allows,2025-07-23T11:19:31Z +issue,90820,Support `divmod` for tensors,https://github.com/pytorch/pytorch/issues/90820,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-07-23T09:52:08Z +issue,158832,Lazy Layer Norm,https://github.com/pytorch/pytorch/issues/158832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-07-22T22:36:20Z +issue,158566,torch.compile fails with UncapturedHigherOrderOpError when using torch.ops.higher_order.scan,https://github.com/pytorch/pytorch/issues/158566,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, oncall: pt2",scan when time allows,2025-07-22T18:40:30Z +issue,153234,`torch.quantile` perform differently on cpu and cuda,https://github.com/pytorch/pytorch/issues/153234,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: reductions",scan when time allows,2025-07-22T16:20:19Z +issue,158643,torch.compile() produces inconsistent outputs with models containing dropout and in-place ReLU operations,https://github.com/pytorch/pytorch/issues/158643,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-07-22T15:43:31Z +issue,149393,Torch RPC examples from docs say usage is deprecated.,https://github.com/pytorch/pytorch/issues/149393,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rpc, oncall: distributed",scan when time allows,2025-07-22T15:14:07Z +issue,158785,[lint] Set_linter suggests wrong changes,https://github.com/pytorch/pytorch/issues/158785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2025-07-22T14:46:07Z +issue,158240,[Windows][inductor] Windows compilers building was slower 4x than on Linux.,https://github.com/pytorch/pytorch/issues/158240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: windows, oncall: pt2",scan when time allows,2025-07-22T14:34:18Z +issue,158664,DISABLED test_sum_all_cpu_float64 (__main__.TestReductionsCPU),https://github.com/pytorch/pytorch/issues/158664,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",scan when time allows,2025-07-22T14:29:57Z +issue,72821,"`{Batch,Instance}Norm{1,2,3}d` works when `num_features != C`!",https://github.com/pytorch/pytorch/issues/72821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-07-22T09:05:00Z +issue,158326,Support for etcd v3 api as rendezvous backend,https://github.com/pytorch/pytorch/issues/158326,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2025-07-21T20:44:41Z +issue,146330,DeepSeek: MLA attention,https://github.com/pytorch/pytorch/issues/146330,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-07-20T20:22:49Z +issue,158641,support checkPoolLiveAllocations for cudaMallocAsync,https://github.com/pytorch/pytorch/issues/158641,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-07-20T15:37:53Z +issue,92927,"Inplace fused (leaky)relu+(leaky)dropout for memory savings (I think, can be made fully allocation-less if never fully allocating random mask in FlashAttention style and recover the mask from the output)",https://github.com/pytorch/pytorch/issues/92927,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-07-19T20:51:32Z +issue,61819,"[discussion] Support other index dtypes for indexing, take, put, scatter, scatter_reduce and other indexing functions in addition to int64: int8, uint8, int16, int32, uint16 etc (without casting copies/reallocations)",https://github.com/pytorch/pytorch/issues/61819,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: scatter & gather ops, module: type promotion",scan when time allows,2025-07-19T15:58:51Z +issue,157976,support DTensor's local_map in compile,https://github.com/pytorch/pytorch/issues/157976,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: dynamo, oncall: pt2",scan when time allows,2025-07-19T00:33:37Z +issue,158610,vmap will always disable gradient calculation,https://github.com/pytorch/pytorch/issues/158610,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch, module: vmap",scan when time allows,2025-07-18T16:15:36Z +issue,103329,"RuntimeError: torch.vmap a function that includes in-place arithmetic operations on a zero-initialized tensor, an error ""vmap: inplace arithmetic(self, *extra_args) is not possible"" is raised.",https://github.com/pytorch/pytorch/issues/103329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-07-18T04:12:49Z +issue,158364,[feature request] [memory_viz] Flexible time/range memory event filtering controls to skip loading all memory events from pickle in the browser's memory,https://github.com/pytorch/pytorch/issues/158364,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,scan when time allows,2025-07-17T21:14:42Z +issue,138452,Cholesky factorization of sparse tensors,https://github.com/pytorch/pytorch/issues/138452,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2025-07-17T19:57:33Z +issue,152220,[C10D] Allow NCCL single P2P ops to use parent/collective communicator,https://github.com/pytorch/pytorch/issues/152220,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2025-07-17T19:34:50Z +issue,152756,Cuda-12.9 removed libnvToolsExt.so.* and is now purely header nvtx3,https://github.com/pytorch/pytorch/issues/152756,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-07-17T18:23:32Z +issue,31252,[feature request] Better handling for CUDA Out of Memory,https://github.com/pytorch/pytorch/issues/31252,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-07-17T16:57:47Z +issue,157465,Nightly NCCL builds are missing optional features from NCCL,https://github.com/pytorch/pytorch/issues/157465,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: nccl",scan when time allows,2025-07-17T16:54:18Z +issue,93716,End-to-End AMP training with GradScaler,https://github.com/pytorch/pytorch/issues/93716,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, oncall: pt2",scan when time allows,2025-07-17T14:20:20Z +issue,158412,Inconsistent torch.absolute results on complex128 between CPU and CUDA,https://github.com/pytorch/pytorch/issues/158412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nans and infs",scan when time allows,2025-07-17T12:39:18Z +issue,152032,RFC: The State of Custom CUDA extensions in PyTorch,https://github.com/pytorch/pytorch/issues/152032,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda",scan when time allows,2025-07-17T05:40:10Z +issue,158419,torch.copysign for float16 produces inconsistent results between CPU and CUDA,https://github.com/pytorch/pytorch/issues/158419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2025-07-17T04:28:27Z +issue,106664,[docs] URL and link format proposal to make function page URLs more concise,https://github.com/pytorch/pytorch/issues/106664,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-07-16T20:28:54Z +issue,113232,[Torch Inductor] Torch Inductor Better Support for GNN workload and Inductor Sparse Compiler,https://github.com/pytorch/pytorch/issues/113232,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: inductor, oncall: pt2",scan when time allows,2025-07-16T20:17:03Z +issue,153463,Feedback about Class Tensor,https://github.com/pytorch/pytorch/issues/153463,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2025-07-16T17:04:44Z +issue,102517,[feature request] PyTorch support for sub-interpreters with PEP 684 accepted and release in Python 3.12,https://github.com/pytorch/pytorch/issues/102517,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-07-16T16:07:38Z +issue,124714,`torch.distributed` hangs when using `torch.distributed.barrier` before any other communication primitives.,https://github.com/pytorch/pytorch/issues/124714,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-07-16T14:37:38Z +issue,37162,torch.distributions bug in RelaxedOneHotCategorical.log_prob,https://github.com/pytorch/pytorch/issues/37162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-07-16T09:26:36Z +issue,157624,file_name is not correctly read in here,https://github.com/pytorch/pytorch/issues/157624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: serialization, module: windows",scan when time allows,2025-07-16T09:08:47Z +issue,148807,Native Infinite Sampler for Datasets,https://github.com/pytorch/pytorch/issues/148807,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2025-07-16T00:28:09Z +issue,158169,Invalid `torch.linalg.lstsq` check during vmapped implementation,https://github.com/pytorch/pytorch/issues/158169,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: linear algebra, module: vmap",scan when time allows,2025-07-15T09:39:49Z +issue,158244,[QST] [DTensor] MoE + EP support,https://github.com/pytorch/pytorch/issues/158244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-07-15T02:26:58Z +issue,158125,index.Tensor doesn't properly account for boolean masks acting on multiple dimensions.,https://github.com/pytorch/pytorch/issues/158125,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-07-14T23:29:09Z +issue,111519,"[pt2+profiler] attach aot_id to CompiledFunction, _compiled_fn number without aot_autograd",https://github.com/pytorch/pytorch/issues/111519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-07-14T22:42:46Z +issue,112883,Torchbench inference failures,https://github.com/pytorch/pytorch/issues/112883,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-07-14T22:42:13Z +issue,155616,Additional Documentation About Size Checking with Custom Operators,https://github.com/pytorch/pytorch/issues/155616,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-07-14T21:05:59Z +issue,158080,loaded storage won't be cached during deserialization if location is meta,https://github.com/pytorch/pytorch/issues/158080,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: serialization",scan when time allows,2025-07-14T14:47:22Z +issue,158172,`torch.fmin` has inconsistent overflow behavior on CPU and GPU,https://github.com/pytorch/pytorch/issues/158172,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases",scan when time allows,2025-07-14T02:55:01Z +issue,108565,`bytes(...)` support of torch tensor does not match numpy + it would be nice to support tensor.tobytes() as alias,https://github.com/pytorch/pytorch/issues/108565,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-07-13T22:14:48Z +issue,1410,BatchNorm should use Bessel's correction consistently,https://github.com/pytorch/pytorch/issues/1410,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2025-07-13T21:31:56Z +issue,118715,Add capturable Adagrad implementation,https://github.com/pytorch/pytorch/issues/118715,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: optimizer, oncall: pt2",scan when time allows,2025-07-12T20:26:32Z +issue,157391,[FSDP2] document the contract for modifying DTensor model.parameters(),https://github.com/pytorch/pytorch/issues/157391,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-07-12T00:52:24Z +issue,157764,Drop SSE4 support in oneDNN,https://github.com/pytorch/pytorch/issues/157764,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: intel, module: mkldnn",scan when time allows,2025-07-12T00:16:11Z +issue,157495,[DTensor] Improve `tensor_metadata` and `redistribute_cost` coverage for op strategies.,https://github.com/pytorch/pytorch/issues/157495,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-07-11T22:31:37Z +issue,103990,[Inductor] Freezing Add support for Caching Parameter Conversions,https://github.com/pytorch/pytorch/issues/103990,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-07-11T18:17:07Z +issue,120667,Triton kernel unexpectedly gets 1.35x slower by more specializaiton,https://github.com/pytorch/pytorch/issues/120667,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: pt2",scan when time allows,2025-07-11T18:10:35Z +issue,121371,Match HuggingFace T5 SDPA pattern in Inductor,https://github.com/pytorch/pytorch/issues/121371,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: performance, oncall: pt2",scan when time allows,2025-07-11T18:04:49Z +issue,121162,opcheck should support TorchBind custom classes,https://github.com/pytorch/pytorch/issues/121162,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, module: torchbind, oncall: pt2",scan when time allows,2025-07-11T18:04:26Z +issue,120116,Use Incremental Fake Tensor Updater more uniformly across torch.compile compilation,https://github.com/pytorch/pytorch/issues/120116,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-07-11T18:03:39Z +issue,120079,Use sys.settrace or torch function mode to compute how much of a model was not covered by Dynamo,https://github.com/pytorch/pytorch/issues/120079,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: logging, oncall: pt2",scan when time allows,2025-07-11T18:03:17Z +issue,120057,Torch compile fx graph is not removing constant propagation,https://github.com/pytorch/pytorch/issues/120057,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-07-11T18:02:32Z +issue,118774,Tune whether to use mm or bmm for matmul in inductor max-autotune,https://github.com/pytorch/pytorch/issues/118774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: inductor, oncall: pt2",scan when time allows,2025-07-11T17:59:43Z +issue,128012,compile PixArt-sigma error,https://github.com/pytorch/pytorch/issues/128012,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-07-11T17:57:42Z +issue,41213,libTorch cpp docs missing for Tensor::item(),https://github.com/pytorch/pytorch/issues/41213,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2025-07-11T17:55:35Z +issue,124866,support side effects in HOPs?,https://github.com/pytorch/pytorch/issues/124866,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-07-11T16:45:50Z +issue,151303,[Feature Request] Experimental support to Moore Threads GPU MUSA,https://github.com/pytorch/pytorch/issues/151303,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: privateuse1",scan when time allows,2025-07-11T16:01:47Z +issue,29734,Tensor.nbytes() returns itemsize * numel for sparse tensors,https://github.com/pytorch/pytorch/issues/29734,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2025-07-11T12:49:38Z +issue,121631,[inductor] Incorrect handle of `autocast` results in type mismatch,https://github.com/pytorch/pytorch/issues/121631,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-07-11T00:58:00Z +issue,103173,[RFC] Emit better Telemetry in PyTorch,https://github.com/pytorch/pytorch/issues/103173,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: logging, oncall: pt2",scan when time allows,2025-07-10T23:58:53Z +issue,126644,[torch.compile]: Enhanced Error Reporting and Performance Canary Mode,https://github.com/pytorch/pytorch/issues/126644,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-07-10T23:53:05Z +issue,118175,Different behaviors in `torch.nn.functional.hinge_embedding_loss` between eagermode and torch.compile,https://github.com/pytorch/pytorch/issues/118175,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-07-10T23:30:06Z +issue,125672,[inductor][cpu]mobilenet_v2_quantized_qat float32 single thread static/dynamic shape CPP/default wrapper performance regression in 2024-04-28 nightly release,https://github.com/pytorch/pytorch/issues/125672,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",scan when time allows,2025-07-10T22:48:24Z +issue,123548,An error occurs when ‘max_split_size_mb ’and ‘expandable_segments ’ are enabled at the same time.,https://github.com/pytorch/pytorch/issues/123548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: docs, module: memory usage",scan when time allows,2025-07-10T20:28:25Z +issue,158041,pow_test fails on Aarch64,https://github.com/pytorch/pytorch/issues/158041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: correctness (silent), module: macos, module: nans and infs",scan when time allows,2025-07-10T17:57:59Z +issue,157334,Inefficient 2D convolution compared to JAX,https://github.com/pytorch/pytorch/issues/157334,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn, module: performance",scan when time allows,2025-07-10T16:31:33Z +issue,125217,MaxPool2D memory leakage on device MPS,https://github.com/pytorch/pytorch/issues/125217,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: mps",scan when time allows,2025-07-10T16:28:15Z +issue,133250,[torch.compile] tighten FX graph restrictions post-functionalization,https://github.com/pytorch/pytorch/issues/133250,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-07-10T13:20:16Z +issue,55261,Batched multi_dot / chain_matmul + let it accept a tensor instead of tuple,https://github.com/pytorch/pytorch/issues/55261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-07-10T07:52:02Z +issue,148289,Triton Kernel Rejects NamedTupleVariable Arguments,https://github.com/pytorch/pytorch/issues/148289,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: fx, module: user triton, oncall: pt2",scan when time allows,2025-07-10T00:17:56Z +issue,157502,Failure with cub::TransformInputIterator in 12.9 periodic CI test,https://github.com/pytorch/pytorch/issues/157502,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-07-09T22:02:30Z +issue,156258,[ONNX] Create a tutorial for exporting hf transformers model,https://github.com/pytorch/pytorch/issues/156258,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: onnx, oncall: pt2",scan when time allows,2025-07-09T18:05:12Z +issue,153472,Multihead Attention does not work with jagged tensors due to __torch_function__,https://github.com/pytorch/pytorch/issues/153472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-07-09T17:31:22Z +issue,150328,☂️ Update submodule dependencies to supported version of Cmake,https://github.com/pytorch/pytorch/issues/150328,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-07-09T17:03:25Z +issue,150118,PyTorch 2.6 License Issues,https://github.com/pytorch/pytorch/issues/150118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: third_party, oncall: releng",scan when time allows,2025-07-09T16:53:30Z +issue,150399,torch wheels are unusable if CUDA RPMs are installed on the system (was Import error in nvidia/cuda:12.6.3-cudnn-devel-rockylinux9),https://github.com/pytorch/pytorch/issues/150399,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: third_party",scan when time allows,2025-07-09T16:51:23Z +issue,147011,Deprecation of NVTX 2 (`nvToolsExt`): Recommended to move to NVTX 3,https://github.com/pytorch/pytorch/issues/147011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: profiler",scan when time allows,2025-07-09T16:43:04Z +issue,125198,[CUDA][Compex] `test_reference_numerics_large_jiterator_unary_cuda_complex64` broken after updating to `numpy >= 1.25.0`,https://github.com/pytorch/pytorch/issues/125198,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cuda, module: jiterator, module: numpy",scan when time allows,2025-07-09T15:43:49Z +issue,157918,[Precompile] Umbrella task,https://github.com/pytorch/pytorch/issues/157918,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-07-09T14:34:30Z +issue,155378,RuntimeError: Could not find libnvrtc.so. Please make sure CUDA is installed.,https://github.com/pytorch/pytorch/issues/155378,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-07-09T12:26:47Z +issue,107112,from_blob python api,https://github.com/pytorch/pytorch/issues/107112,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-07-09T12:19:15Z +issue,42654,RuntimeError: each element in list of batch should be of equal size,https://github.com/pytorch/pytorch/issues/42654,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-07-09T10:10:35Z +issue,156956,Allow clamp to work on absolute values and preserve sign,https://github.com/pytorch/pytorch/issues/156956,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-07-09T09:35:11Z +issue,150734,Training/Fine-tuning fails with PyTorch 2.8 + 4x 5090 GPUs using DDP/FSDP/DeepSpeed,https://github.com/pytorch/pytorch/issues/150734,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: fsdp, oncall: distributed",scan when time allows,2025-07-09T06:29:05Z +issue,157591,PT2E Quantization Migration Tracker,https://github.com/pytorch/pytorch/issues/157591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2025-07-08T22:31:37Z +issue,152212,[Release improvements] Have cherry-pick bot always add the current release to the PR,https://github.com/pytorch/pytorch/issues/152212,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",scan when time allows,2025-07-08T19:41:48Z +issue,148762,magma builds should be part of the docker image builds,https://github.com/pytorch/pytorch/issues/148762,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",scan when time allows,2025-07-08T19:28:56Z +issue,157808,sparse_csr bfloat16 matrix multiplication backward is 10x slower than float16,https://github.com/pytorch/pytorch/issues/157808,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: performance, module: sparse",scan when time allows,2025-07-08T16:05:56Z +issue,156607,[OSS tooling] pytorchbot fail to revert a PR,https://github.com/pytorch/pytorch/issues/156607,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",scan when time allows,2025-07-08T15:31:52Z +issue,68905,Shared `~/.cache/torch_extensions` needs to be pytorch version aware.,https://github.com/pytorch/pytorch/issues/68905,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-07-08T09:42:29Z +issue,155825,Running dispatch modes on compile-disabled regions of a compiled model,https://github.com/pytorch/pytorch/issues/155825,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-07-08T02:19:38Z +issue,157545,test_dtensor.py::test_dtensor_save_load_import conflicts with autoloader importing torch._dynamo,https://github.com/pytorch/pytorch/issues/157545,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-07-07T22:27:14Z +issue,156810,Add the XPU item to pytorch.org/get-started,https://github.com/pytorch/pytorch/issues/156810,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docs, oncall: releng",scan when time allows,2025-07-07T22:24:09Z +issue,153337,`torch.combinations` exhibits excessive memory usage and hangs for moderate `n` and `r` due to `n^r`,https://github.com/pytorch/pytorch/issues/153337,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2025-07-07T20:22:32Z +issue,157662,OffsetBasedRNGTracker's run_state_sync causes deadlock due to inconsistent broadcast order across ranks,https://github.com/pytorch/pytorch/issues/157662,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: random, oncall: distributed",scan when time allows,2025-07-07T19:56:43Z +issue,157721,[Precompile] Caching precompile and presence of packages in general should be safe re: guard serialization,https://github.com/pytorch/pytorch/issues/157721,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-07-07T18:34:41Z +issue,1529,[feature request] Caching allocator diagnostics and memory allocation tracing/visualization,https://github.com/pytorch/pytorch/issues/1529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2025-07-07T14:47:21Z +issue,156616,Add `is_outputs_batched` param to `autograd.grad`,https://github.com/pytorch/pytorch/issues/156616,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2025-07-06T12:05:05Z +issue,88802,ImportError: libcupti.so.11.2: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/88802,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-07-06T09:12:41Z +issue,145608,ROCm+gcc 15 asserts,https://github.com/pytorch/pytorch/issues/145608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-07-05T21:53:07Z +issue,107041,"I want to calculate the matrix multiplication of two Boolean matrices, but torch.mm will report an error. Is there any more efficient alternative?",https://github.com/pytorch/pytorch/issues/107041,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor,scan when time allows,2025-07-05T17:27:43Z +issue,69197,RendezvousConnectionError when use C10d on multi nodes,https://github.com/pytorch/pytorch/issues/69197,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: r2p,scan when time allows,2025-07-05T17:01:16Z +issue,151591,Device Error on vmap,https://github.com/pytorch/pytorch/issues/151591,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2025-07-04T04:41:49Z +issue,153531,Documentation: explaining the STFT formula,https://github.com/pytorch/pytorch/issues/153531,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: fft",scan when time allows,2025-07-03T22:43:29Z +issue,157548,Nightly cu128 aarch64 wheels haven't been built for weeks,https://github.com/pytorch/pytorch/issues/157548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: docs",scan when time allows,2025-07-03T19:59:33Z +issue,40375,Add batched torch.combinations,https://github.com/pytorch/pytorch/issues/40375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: batching,scan when time allows,2025-07-03T14:25:46Z +issue,126753,[cudagraph] simplify usage of how cudagraph dumps debug file,https://github.com/pytorch/pytorch/issues/126753,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2025-07-03T02:18:47Z +issue,110858,Broadcasting matmul is much slower than corresponding einsum,https://github.com/pytorch/pytorch/issues/110858,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2025-07-02T23:47:37Z +issue,124487,PyTorch Memory Management in GPU-to-CPU Transfers issue,https://github.com/pytorch/pytorch/issues/124487,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2025-07-02T18:15:29Z +issue,157431,``torch.quantile`` edge case,https://github.com/pytorch/pytorch/issues/157431,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2025-07-02T16:21:07Z +issue,156085,Idea: Add SBOM Generation (and optional vuln scan) for better supply chain insight,https://github.com/pytorch/pytorch/issues/156085,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2025-07-02T15:03:03Z +issue,156680,torch compile does not support SyncBatchNorm with fullgraph=True,https://github.com/pytorch/pytorch/issues/156680,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: dynamo, oncall: distributed, oncall: pt2",scan when time allows,2025-07-02T05:11:48Z +issue,149660,cd: Migrate binary builds off of Jinja,https://github.com/pytorch/pytorch/issues/149660,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2025-07-01T20:27:08Z +issue,157075,Improve debug message for metadata guard failure,https://github.com/pytorch/pytorch/issues/157075,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-07-01T19:18:17Z +issue,116414,Convolution NN for complex numbers and more special functions,https://github.com/pytorch/pytorch/issues/116414,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2025-07-01T17:58:35Z +issue,101385,torch.Tensor.is_sparse returns false for non-COO sparse tensors,https://github.com/pytorch/pytorch/issues/101385,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-07-01T16:57:40Z +issue,157220,[ONNX] torch.nn.functional.interpolate \w antialias=True isn't op.Resize compatible,https://github.com/pytorch/pytorch/issues/157220,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-06-30T22:16:24Z +issue,157321,Symmetric memory test failed with TORCH_SYMMMEM=NVSHMEM,https://github.com/pytorch/pytorch/issues/157321,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-06-30T22:14:46Z +issue,157272,`torch.reciprocal` and `torch.divide` for Complex `inf` Incorrectly Returns `NaN` Only for Tensors with >= 4 Elements on CPU,https://github.com/pytorch/pytorch/issues/157272,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cpu, module: edge cases, module: nans and infs",scan when time allows,2025-06-30T21:13:55Z +issue,157266,"Better typechecking of `int` only-operators `|`, `^`, `&`, `<<`, `>>`, `~` and `@`",https://github.com/pytorch/pytorch/issues/157266,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-06-30T20:04:48Z +issue,156725,CMake improperly configures pybind11. 3 different versions of pybind11 in use at the sametime.,https://github.com/pytorch/pytorch/issues/156725,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: pybind",scan when time allows,2025-06-30T17:07:13Z +issue,33890,Libtorch segfault when used with libqpOASES,https://github.com/pytorch/pytorch/issues/33890,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-06-30T08:16:37Z +issue,95882,Proposal: accept ellipsis as index in `torch.index_put()` to enable broadcasting,https://github.com/pytorch/pytorch/issues/95882,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-06-30T05:31:17Z +issue,157101,Documentation update: Contribution.md,https://github.com/pytorch/pytorch/issues/157101,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-06-30T01:57:34Z +issue,26334,"Supporting ""cdf"" for Student-T distribution",https://github.com/pytorch/pytorch/issues/26334,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-06-28T23:03:48Z +issue,101107,Make compiled models serializable,https://github.com/pytorch/pytorch/issues/101107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-06-28T04:13:42Z +issue,133221,Tensor parallel for convolutions and groupnorm,https://github.com/pytorch/pytorch/issues/133221,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2025-06-27T10:12:41Z +issue,140015,Find ninja from local environment in `torch.utils.cpp_extension.load`,https://github.com/pytorch/pytorch/issues/140015,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-06-27T09:07:58Z +issue,90775,"Extend ""torch.utils.cpp_extension.load"" for both lib64 and **lib**",https://github.com/pytorch/pytorch/issues/90775,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-06-27T04:34:02Z +issue,152964,[FSDP2] need dummy forward/backward to stay SPMD,https://github.com/pytorch/pytorch/issues/152964,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-06-26T23:27:27Z +issue,156811,Allow to set a seed for random tensors in torch.onnx.export,https://github.com/pytorch/pytorch/issues/156811,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-06-26T18:14:47Z +issue,94718,Option to bypass NOLA check in torch.istft,https://github.com/pytorch/pytorch/issues/94718,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2025-06-26T16:58:19Z +issue,156733,FX graph printing do newlines for parameters by default,https://github.com/pytorch/pytorch/issues/156733,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-06-26T15:45:45Z +issue,156820,put compile-based cudagraph re-recordings in tlparse,https://github.com/pytorch/pytorch/issues/156820,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, oncall: pt2",scan when time allows,2025-06-26T00:32:43Z +issue,56697,Avoid no-op suggest_memory_format call in SparseCsrTensorImpl::resize_as_sparse_csr_tensor_,https://github.com/pytorch/pytorch/issues/56697,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-06-25T20:51:20Z +issue,152261,Context Parallel -- unsharded output doesn't match output without CP.,https://github.com/pytorch/pytorch/issues/152261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-06-25T19:32:01Z +issue,156072,[typing][docs] `torch.amin` and `torch.amax` do not document `dim=None`,https://github.com/pytorch/pytorch/issues/156072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-06-25T09:31:57Z +issue,156238,Upgrade torch._scaled_grouped_mm to SM100+,https://github.com/pytorch/pytorch/issues/156238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-06-25T09:31:21Z +issue,156411,[compile] torch._dynamo.exc.TorchRuntimeError: Failed running call_function aten.lift_fresh_copy.default,https://github.com/pytorch/pytorch/issues/156411,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-06-24T18:17:21Z +issue,149315,How to Retain Computational Graph in torch.func.jvp() for Parameter Gradients?,https://github.com/pytorch/pytorch/issues/149315,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2025-06-24T14:30:39Z +issue,149155,torch.multiprocessing.Queue Zeroes Out Tensors on Retrieval,https://github.com/pytorch/pytorch/issues/149155,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing, module: windows",scan when time allows,2025-06-24T09:51:21Z +issue,121968,[RFC] Use CUDA graphs by default on torch.compile,https://github.com/pytorch/pytorch/issues/121968,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-06-23T21:59:38Z +issue,156520,[user empathy] compile for `transformers` model,https://github.com/pytorch/pytorch/issues/156520,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: cuda graphs, module: dynamo, oncall: pt2",scan when time allows,2025-06-23T18:52:15Z +issue,156622,Add a test to torch.library.opcheck for if a C++ meta kernel specializes on a symint,https://github.com/pytorch/pytorch/issues/156622,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: dynamic shapes, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-06-23T18:51:52Z +issue,89255,torch.lobpcg should support black-box linear operators like SciPy,https://github.com/pytorch/pytorch/issues/89255,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-06-23T17:23:37Z +issue,153241,Online softmax is disabled on the fly,https://github.com/pytorch/pytorch/issues/153241,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-06-22T10:22:26Z +issue,111374,Tensor.lerp inconsistent when using -Infinity between MPS and CPU,https://github.com/pytorch/pytorch/issues/111374,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-06-21T13:05:36Z +issue,88327,MSE documentation is weak,https://github.com/pytorch/pytorch/issues/88327,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2025-06-20T23:41:11Z +issue,155171,[AC] torch.utils.checkpoint.CheckpointError from HF qwen2,https://github.com/pytorch/pytorch/issues/155171,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2025-06-20T21:26:42Z +issue,156444,Tensor.is_pinned() raises error after renaming privateuseone backend.,https://github.com/pytorch/pytorch/issues/156444,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: privateuse1,scan when time allows,2025-06-20T21:23:14Z +issue,154810,Multi-dimensional tensors in datasets might get incorrectly flattened when fetching data from dataloader which is specified 'batch_sampler' when created,https://github.com/pytorch/pytorch/issues/154810,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-06-20T19:38:55Z +issue,148335,[CI] [anaconda] Docker files have conda environment installed,https://github.com/pytorch/pytorch/issues/148335,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-06-20T17:21:04Z +issue,148339,[Docs] [anaconda] Review and update,https://github.com/pytorch/pytorch/issues/148339,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-06-20T17:20:02Z +issue,148338,[CI] [anaconda] CI Build and Test scripts Windows,https://github.com/pytorch/pytorch/issues/148338,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-06-20T17:19:40Z +issue,148336,[CI] [anaconda] CI Build and Test scripts Linux,https://github.com/pytorch/pytorch/issues/148336,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-06-20T17:19:22Z +issue,147720,[ONNX Convert] Error when input to nn.AdaptiveAvgPool2d size is variable,https://github.com/pytorch/pytorch/issues/147720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-06-20T14:32:01Z +issue,151064,[ONNX] Use dlpack to transfer tensors when onnxruntime implements proper support,https://github.com/pytorch/pytorch/issues/151064,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-06-20T14:27:17Z +issue,113817,[ONNX] ONNX export of simple quantized model fails,https://github.com/pytorch/pytorch/issues/113817,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-06-20T00:15:19Z +issue,150277,"Sparse tensor indexing not implemented, but partially supported by using index_select",https://github.com/pytorch/pytorch/issues/150277,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-06-19T16:38:05Z +issue,152226,NotImplementedError: Could not run 'aten::index.Tensor' with arguments from the 'SparseCUDA' backend.,https://github.com/pytorch/pytorch/issues/152226,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-06-19T16:30:16Z +issue,153329,Segmentation fault when converting sparse COO tensor with complex values to dense,https://github.com/pytorch/pytorch/issues/153329,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: sparse",scan when time allows,2025-06-19T15:58:03Z +issue,150972,Escape hatch: way to dynamically add or remove tags from custom operators,https://github.com/pytorch/pytorch/issues/150972,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-06-18T18:11:05Z +issue,106801,"Timer benchmark stores only one time value, and therefore has broken mean/median/etc metrics",https://github.com/pytorch/pytorch/issues/106801,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: benchmark,scan when time allows,2025-06-18T17:07:06Z +issue,154602,Cudnn attention is very slow when sequence length changed in every step,https://github.com/pytorch/pytorch/issues/154602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: sdpa",scan when time allows,2025-06-18T14:54:59Z +issue,138329,DTensor RNG state for non CUDA backends,https://github.com/pytorch/pytorch/issues/138329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-06-18T03:46:30Z +issue,155552,Reproducibility of results without AVX512 by setting ATEN_CPU_CAPABILITY=avx2,https://github.com/pytorch/pytorch/issues/155552,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: intel, module: mkl, module: mkldnn",scan when time allows,2025-06-18T02:03:55Z +issue,123834,"RuntimeError: NVML_SUCCESS == r INTERNAL ASSERT FAILED at ""/opt/pytorch/pytorch/c10/cuda/CUDACachingAllocator.cpp"":830, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/123834,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-06-17T22:07:43Z +issue,156177,nn.Module._load_from_state_dict is always called with strict=True,https://github.com/pytorch/pytorch/issues/156177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-06-17T21:46:30Z +issue,155203,support for cuDNN 9.8+,https://github.com/pytorch/pytorch/issues/155203,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cudnn",scan when time allows,2025-06-17T20:47:16Z +issue,107214,The difference between input grad computed by channels last backward and the input grad computed by channels first backward of Hardswish on MPS is too large,https://github.com/pytorch/pytorch/issues/107214,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: memory format, module: mps, module: nn",scan when time allows,2025-06-17T15:59:42Z +issue,156024,Ошибка установки torch для CUDA 12.1 на GTX 1660 Ti,https://github.com/pytorch/pytorch/issues/156024,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: windows",scan when time allows,2025-06-17T15:30:03Z +issue,156152,Inconsistent `torch.rsqrt` results on complex128 between CPU and CUDA,https://github.com/pytorch/pytorch/issues/156152,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: edge cases, module: nans and infs",scan when time allows,2025-06-17T15:28:30Z +issue,156073,[codespell] fix typos in the codebase,https://github.com/pytorch/pytorch/issues/156073,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2025-06-17T14:58:13Z +issue,150544,[ONNX] dynamic_axes does not rename dynamic dimension in torch.onnx.export,https://github.com/pytorch/pytorch/issues/150544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-06-17T14:31:33Z +issue,10454,[feature request] Rank-Revealing QR - Adding dgeqp3 support to torch.qr,https://github.com/pytorch/pytorch/issues/10454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-06-17T08:46:14Z +issue,55289,Feature Request: Add a rounding mode to round,https://github.com/pytorch/pytorch/issues/55289,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-06-17T07:20:40Z +issue,61509,[feature request] Exact euclidean distance transform,https://github.com/pytorch/pytorch/issues/61509,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distance functions,scan when time allows,2025-06-17T02:00:48Z +issue,146566,[DCP] Allow for rank-specific tensors with duplicate keys,https://github.com/pytorch/pytorch/issues/146566,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed checkpointing,scan when time allows,2025-06-17T01:33:50Z +issue,64491,TypeError when using torch.cuda.list_gpu_processes() on Windows with the WDDM driver,https://github.com/pytorch/pytorch/issues/64491,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-06-16T13:34:45Z +issue,156020,torch.fft.ifft for complex64 produces inconsistent results between CPU and CUDA,https://github.com/pytorch/pytorch/issues/156020,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: fft, module: nans and infs",scan when time allows,2025-06-16T02:42:23Z +issue,88227,Enable `torch.topk` to support `stable` flag,https://github.com/pytorch/pytorch/issues/88227,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2025-06-15T14:16:35Z +issue,146221,"Incompatible Torch and Torchvision while building from source for 2.6.0 and CUDA 12.6, RuntimeError: operator torchvision::nms does not exist",https://github.com/pytorch/pytorch/issues/146221,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: dependency bug",scan when time allows,2025-06-15T13:17:20Z +issue,117287,Error computing the norm of MaskedTensor,https://github.com/pytorch/pytorch/issues/117287,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2025-06-14T22:11:42Z +issue,143474,[ONNX] Failed to export PyTorch-2-Export-Quantized model to onnx,https://github.com/pytorch/pytorch/issues/143474,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-06-13T22:54:27Z +issue,155838,intmm is being compiled inconsistently with errors.,https://github.com/pytorch/pytorch/issues/155838,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2, oncall: quantization",scan when time allows,2025-06-13T20:35:30Z +issue,51284,result_type doesn't take dtypes and doesn't match numpy,https://github.com/pytorch/pytorch/issues/51284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-06-13T00:11:23Z +issue,82091,"RuntimeError: [1] is setting up NCCL communicator and retreiving ncclUniqueId from [0] via c10d key-value store by key '0', but store->get('0') got error: Timeout waiting for key: default_pg/0/0 after 1800000 ms",https://github.com/pytorch/pytorch/issues/82091,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: nccl, oncall: distributed",scan when time allows,2025-06-12T18:41:11Z +issue,155218,3D Conv Slow with Large Tensors (> 2**31) Elements,https://github.com/pytorch/pytorch/issues/155218,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: performance",scan when time allows,2025-06-12T10:21:04Z +issue,155001,The profiler does not seem to be able to record cuda runtime nodes,https://github.com/pytorch/pytorch/issues/155001,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,scan when time allows,2025-06-12T07:34:57Z +issue,81985,"nn.InstanceNorm and nn.GroupNorm are affected by padding, so they need to masking",https://github.com/pytorch/pytorch/issues/81985,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: masked operators, module: nestedtensor, module: norms and normalization, module: padding",scan when time allows,2025-06-12T04:22:59Z +issue,155680,[FSDP] optimizer_state_dict fails if FSDP model is not on global rank 0,https://github.com/pytorch/pytorch/issues/155680,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-06-12T03:53:41Z +issue,155667,[libtorch] Debug and Release should be shipped together on Windows,https://github.com/pytorch/pytorch/issues/155667,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2025-06-11T18:38:23Z +issue,144344,custom_op's backward changes can't invalidate `torch.compile` cache for backward,https://github.com/pytorch/pytorch/issues/144344,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-06-11T17:20:15Z +issue,155386,OpOverloads should have annotations,https://github.com/pytorch/pytorch/issues/155386,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: dispatch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-06-11T16:32:41Z +issue,131774,Different Cholesky results between Windows & Linux,https://github.com/pytorch/pytorch/issues/131774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: windows",scan when time allows,2025-06-11T16:21:13Z +issue,152720,When scoped_libary is destroyed the fake impls are not cleared,https://github.com/pytorch/pytorch/issues/152720,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-06-11T16:20:40Z +issue,155703,[pt2] [precompile] Support cudagraphs in BundledAOTAutogradCacheArtifact,https://github.com/pytorch/pytorch/issues/155703,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-06-11T14:38:11Z +issue,102400,Unexpected Behavior when using torch.isclose(),https://github.com/pytorch/pytorch/issues/102400,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: testing",scan when time allows,2025-06-10T14:58:36Z +issue,155519,[Tracking] Triton 3.2 deprecation,https://github.com/pytorch/pytorch/issues/155519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: user triton,scan when time allows,2025-06-10T13:31:36Z +issue,152294,Unexpected overflow behavior when using `torch.addcmul`,https://github.com/pytorch/pytorch/issues/152294,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2025-06-10T08:40:31Z +issue,155490,PP activation offloading,https://github.com/pytorch/pytorch/issues/155490,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pipelining, oncall: distributed",scan when time allows,2025-06-10T07:14:49Z +issue,153967,Memory Corruption in `torch.batch_norm_update_stats`,https://github.com/pytorch/pytorch/issues/153967,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-06-10T04:00:13Z +issue,155105,"[dynamo && compiled autograd] compiled autograd recompiles when using custom autograd functions and compiling the forward with backend ""eager""",https://github.com/pytorch/pytorch/issues/155105,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, module: dynamo, module: higher order operators, module: pt2-dispatcher",scan when time allows,2025-06-09T23:49:42Z +issue,155230,[scan] scan is broken in nightly,https://github.com/pytorch/pytorch/issues/155230,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: higher order operators,scan when time allows,2025-06-09T23:39:42Z +issue,155065,Flex Attention and Nested Tensor: very high VRAM usage,https://github.com/pytorch/pytorch/issues/155065,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-06-09T20:07:50Z +issue,155193,Should be ReLU6(Module),https://github.com/pytorch/pytorch/issues/155193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-06-09T20:06:54Z +issue,152870,Docs Update `wrap_triton`,https://github.com/pytorch/pytorch/issues/152870,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: higher order operators, module: pt2-dispatcher, module: user triton, oncall: pt2",scan when time allows,2025-06-09T17:43:28Z +issue,155333,[A100][cusparselt] Non-determinstic correctness issue with some algorithms,https://github.com/pytorch/pytorch/issues/155333,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-06-09T17:27:44Z +issue,154428,Inconsistent sin(x) output between CPU and CUDA for very large arguments,https://github.com/pytorch/pytorch/issues/154428,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: python frontend",scan when time allows,2025-06-09T16:58:50Z +issue,120584,`torch.set_default_tensor_type() is deprecated as of PyTorch 2.1` appearing in logs even when not using this function,https://github.com/pytorch/pytorch/issues/120584,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: docs, module: error checking",scan when time allows,2025-06-09T08:59:50Z +issue,154332,Support jvp for flex attention,https://github.com/pytorch/pytorch/issues/154332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: flex attention, module: forward ad, module: functorch",scan when time allows,2025-06-08T00:39:49Z +issue,120986,Intel MKL DFTI ERROR,https://github.com/pytorch/pytorch/issues/120986,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: empty tensor, module: fft, module: intel",scan when time allows,2025-06-08T00:02:46Z +issue,155350,CUDA_HOME doesn't seem to work with setup script,https://github.com/pytorch/pytorch/issues/155350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-06-07T14:32:22Z +issue,75173,LowRankMultivariateNormal doesn't work with 0 diagonal,https://github.com/pytorch/pytorch/issues/75173,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-06-07T12:44:45Z +issue,155157,Multi-worker dataloader for `IterableDataset` holds open process on macOS Python 3.12,https://github.com/pytorch/pytorch/issues/155157,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-06-06T22:04:34Z +issue,133676,"Can pytorch add sparse linear solvers like scipy.sparse.linalg.gmres, scipy.sparse.linalg.bicg etc.",https://github.com/pytorch/pytorch/issues/133676,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2025-06-06T13:46:37Z +issue,151695,FP8 Support for FlexAttention,https://github.com/pytorch/pytorch/issues/151695,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flex attention, module: higher order operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-06-06T13:19:39Z +issue,129656,RuntimeError: Shared memory manager connection has timed out,https://github.com/pytorch/pytorch/issues/129656,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2025-06-06T02:45:08Z +issue,152922,Enable 12.8.1,https://github.com/pytorch/pytorch/issues/152922,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-06-06T00:00:56Z +issue,155266,torch.compile with InvokeAI: 'UserDefinedObjectVariable' object has no attribute 'proxy' / compute_ancestors KeyError op4,https://github.com/pytorch/pytorch/issues/155266,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: inductor, oncall: pt2",scan when time allows,2025-06-05T22:14:24Z +issue,135126,"""RuntimeError: CUDA error: operation not supported"" fixed by downgrading toolkit version",https://github.com/pytorch/pytorch/issues/135126,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, oncall: pt2",scan when time allows,2025-06-05T16:13:26Z +issue,144068,redundant recompilation caused by duplicated Sym(),https://github.com/pytorch/pytorch/issues/144068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: dynamo, oncall: pt2",scan when time allows,2025-06-05T08:04:30Z +issue,154847,don't require recompiles when switching between torch.Tensor vs AsyncCollectiveTensor graph inputs,https://github.com/pytorch/pytorch/issues/154847,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed, oncall: pt2",scan when time allows,2025-06-05T05:37:28Z +issue,141204,Stack trace from pytest is very far away and far too find on some tests,https://github.com/pytorch/pytorch/issues/141204,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: log classifier, oncall: pt2",scan when time allows,2025-06-05T03:15:51Z +issue,154984,[FSDP2] Slower Convergence with fully_shard() Compared to DDP during Qwen2-VL Fine-Tuning,https://github.com/pytorch/pytorch/issues/154984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-06-05T02:19:58Z +issue,154824,torch inductor cudagraph tree could incorrectly release some input nodes during replay,https://github.com/pytorch/pytorch/issues/154824,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, oncall: pt2",scan when time allows,2025-06-04T22:47:29Z +issue,154351,bfloat16 Conv2d slower than float16 on 4090,https://github.com/pytorch/pytorch/issues/154351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: half, module: performance",scan when time allows,2025-06-04T20:20:38Z +issue,119548,Inconsistent behavior when indexing a Tensor with a list of lists,https://github.com/pytorch/pytorch/issues/119548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2025-06-04T15:56:12Z +issue,154214,AssertionError: Scalars are not equal! ```test_require_stride_expanded_dynamic_shapes_cuda```,https://github.com/pytorch/pytorch/issues/154214,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-06-04T15:25:56Z +issue,122454,Triangular matrix storage + matmul,https://github.com/pytorch/pytorch/issues/122454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-06-04T03:10:21Z +issue,147643,Representation string of a meta tensor is not a valid `tensor` call,https://github.com/pytorch/pytorch/issues/147643,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: meta tensors, module: python frontend",scan when time allows,2025-06-04T02:03:42Z +issue,54816,Per channel weight observer for ConvTranspose,https://github.com/pytorch/pytorch/issues/54816,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2025-06-03T22:46:37Z +issue,139108,LibTorch build error on Windows for CUDA version (debug/release),https://github.com/pytorch/pytorch/issues/139108,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: cuda, module: windows",scan when time allows,2025-06-03T20:59:29Z +issue,144767,Unable to build with ATEN_THREADING=TBB option,https://github.com/pytorch/pytorch/issues/144767,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs, module: tbb",scan when time allows,2025-06-03T20:38:45Z +issue,154560,Enable GB200 for dynamo/torchbench tests,https://github.com/pytorch/pytorch/issues/154560,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda, oncall: pt2",scan when time allows,2025-06-03T20:24:00Z +issue,153273,Auto format lint not making suggestions,https://github.com/pytorch/pytorch/issues/153273,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-06-03T20:12:53Z +issue,152608,[triton pin update] Run Inductor CI on pin updates for Triton and the PyTorch nightly branch,https://github.com/pytorch/pytorch/issues/152608,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: user triton, oncall: releng",scan when time allows,2025-06-03T20:08:38Z +issue,154128,matmul uses excessive memory in batch cases with more than 3 dimensions,https://github.com/pytorch/pytorch/issues/154128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2025-06-03T16:55:05Z +issue,154960,[FSDP2] all_gather_copy_in for cpu offload,https://github.com/pytorch/pytorch/issues/154960,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2025-06-03T15:59:54Z +issue,154855,"test_tensorboard.py::TestTensorBoardSummary::test_histogram_auto Fails with s390x-periodic / linux-manylinux-2_28-py3-cpu-s390x / test (default, 7, 10, linux.s390x)",https://github.com/pytorch/pytorch/issues/154855,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",scan when time allows,2025-06-03T13:55:05Z +issue,154791,[discussion] Specialized frontend method for computing Gram / covariance matrix,https://github.com/pytorch/pytorch/issues/154791,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: python frontend",scan when time allows,2025-06-03T13:52:37Z +issue,115122,[dynamo] Graph breaks from copy.deepcopy,https://github.com/pytorch/pytorch/issues/115122,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: graph breaks, oncall: pt2",scan when time allows,2025-06-02T19:23:18Z +issue,84520,MPS backend appears to be limited to 32 bits,https://github.com/pytorch/pytorch/issues/84520,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-06-02T17:18:45Z +issue,150612,RAM leak during data loading with multiprocessing and Conv3d on CPU in Dataset __getitem__,https://github.com/pytorch/pytorch/issues/150612,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: data, module: dataloader, module: memory usage",scan when time allows,2025-06-02T08:46:02Z +issue,154836,[FSDP2] fix unit test test_all_gather_extension_outer_size_stride,https://github.com/pytorch/pytorch/issues/154836,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-06-02T04:08:04Z +issue,148670,MPS Backend Error: ComplexDouble (complex128) Conversion Fails When Diffusers Transformer Creates 64‐bit Complex Tensors,https://github.com/pytorch/pytorch/issues/148670,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: mps",scan when time allows,2025-06-02T03:31:20Z +issue,154489,[wishlist item] Assume data-dependent info based on tensor construction,https://github.com/pytorch/pytorch/issues/154489,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-06-02T01:34:15Z +issue,154463,Mega Cache/ Torchcompile cache Not working,https://github.com/pytorch/pytorch/issues/154463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-06-02T00:35:34Z +issue,111317,Torch 2.1 compile + FSDP (mixed precision) + LlamaForCausalLM: `RuntimeError: attempting to assign a gradient with dtype 'c10::BFloat16' to a tensor with dtype 'float'.`,https://github.com/pytorch/pytorch/issues/111317,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed, oncall: pt2",scan when time allows,2025-06-01T09:11:52Z +issue,154456,torch.compile tracing doesn't seem to be using the cache,https://github.com/pytorch/pytorch/issues/154456,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, oncall: pt2",scan when time allows,2025-06-01T07:58:57Z +issue,154474,Multiple torch.fft APIs produce inconsistent results for infinite inputs between CPU and GPU,https://github.com/pytorch/pytorch/issues/154474,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: fft, module: nans and infs",scan when time allows,2025-05-31T00:27:59Z +issue,154083,torch.compile joint trace materializes constant tensors,https://github.com/pytorch/pytorch/issues/154083,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-05-30T22:55:50Z +issue,73501,`index_copy` has different index behavior with `index_fill`,https://github.com/pytorch/pytorch/issues/73501,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: edge cases",scan when time allows,2025-05-30T22:08:33Z +issue,150017,`__setitem__` with bool mask and dtype mismatch fails,https://github.com/pytorch/pytorch/issues/150017,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: edge cases, module: type promotion",scan when time allows,2025-05-30T21:54:32Z +issue,154557,grouped_mm optional zero initialization of the output,https://github.com/pytorch/pytorch/issues/154557,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-05-30T16:23:22Z +issue,153861,[BUG] `functionalize` silent data corruption with pre-strided tensors,https://github.com/pytorch/pytorch/issues/153861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-05-30T15:05:45Z +issue,69519,Feature Request: CUDA torch.histogram (and histogramdd),https://github.com/pytorch/pytorch/issues/69519,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: cuda, module: sorting and selection",scan when time allows,2025-05-30T08:44:07Z +issue,153334,Selective Activation Checkpointing on custom autograd.Function,https://github.com/pytorch/pytorch/issues/153334,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2025-05-29T21:40:48Z +issue,152566,[Benchmark] High compilation time variance on benchmark dashboards,https://github.com/pytorch/pytorch/issues/152566,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: infra",scan when time allows,2025-05-29T20:16:30Z +issue,154603,"RuntimeError: devptr INTERNAL ASSERT FAILED at ""/build/source/c10/cuda/CUDACachingAllocator.cpp"":3696, please report a bug to PyTorch. entry in cache has missing shared_ptr",https://github.com/pytorch/pytorch/issues/154603,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2025-05-29T18:18:55Z +issue,153957,MPS backend gradient correctness issues with large shapes,https://github.com/pytorch/pytorch/issues/153957,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-05-29T17:30:42Z +issue,50192,GRU and LSTM fail for seq_len = 0,https://github.com/pytorch/pytorch/issues/50192,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2025-05-29T14:56:44Z +issue,107503,[nightly][jit] bad constant exponent (e+38.f) in default_program fused_mul_div_add,https://github.com/pytorch/pytorch/issues/107503,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2025-05-29T01:32:09Z +issue,154419,Crash in `torch.sparse.softmax`,https://github.com/pytorch/pytorch/issues/154419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-05-28T22:01:03Z +issue,154496,torch.fft.irfft(n) doesn’t handle non-Hermitian inputs in a consistent way,https://github.com/pytorch/pytorch/issues/154496,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: fft",scan when time allows,2025-05-28T21:00:44Z +issue,141133,CuDNN SDPA Issue Tracker,https://github.com/pytorch/pytorch/issues/141133,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: sdpa",scan when time allows,2025-05-28T16:52:06Z +issue,154501,`torch.add` and `torch.sub` return incorrect results when computing complex numbers containing `inf` on both CPU and GPU,https://github.com/pytorch/pytorch/issues/154501,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",scan when time allows,2025-05-28T13:31:40Z +issue,152295,"`vmap` not working on `torch.arange`, `torch.scalar_tensor`, and `torch.ones`",https://github.com/pytorch/pytorch/issues/152295,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2025-05-28T02:36:42Z +issue,154241,`RNNBase` modules break parameter sharing due to `flatten_parameters()`,https://github.com/pytorch/pytorch/issues/154241,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2025-05-27T20:51:13Z +issue,154248,[Testing][CUDA] test_cublas_addmm_reduced_precision_size_10000_backend_cublaslt_cuda_float16 failure with linux-focal-cuda12.6-py3-gcc11-slow-gradcheck,https://github.com/pytorch/pytorch/issues/154248,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: half",scan when time allows,2025-05-27T16:35:37Z +issue,136750,Poor-quality random numbers generated by torch.poisson on gpus,https://github.com/pytorch/pytorch/issues/136750,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-05-27T16:27:12Z +issue,151884,"The ""eager"" and ""aot_eager"" backends have different behavior for the expected gradient tensor of the torch.expend_as operator",https://github.com/pytorch/pytorch/issues/151884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-05-27T16:14:04Z +issue,117449,Deprecate and remove usage of from __future__ import annotations in codebase,https://github.com/pytorch/pytorch/issues/117449,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-05-27T10:56:28Z +issue,109770,Slow performance when running torch.jit traced model with Flash Attention using libtorch on Windows,https://github.com/pytorch/pytorch/issues/109770,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sdpa, module: windows",scan when time allows,2025-05-27T10:24:53Z +issue,154322,float16 → float32 conversion yields unexpected zero matrix for matrices > 43000 × 43000 in MPS,https://github.com/pytorch/pytorch/issues/154322,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: correctness (silent), module: mps",scan when time allows,2025-05-25T13:05:47Z +issue,144615,`collect_env.py` fails with `'NoneType' object has no attribute 'splitlines'` if pytorch is installed without pip,https://github.com/pytorch/pytorch/issues/144615,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: collect_env.py, module: devx",scan when time allows,2025-05-25T12:43:20Z +issue,127337,Add option for custom ops to automatically get a FakeTensor kernel (during static shapes),https://github.com/pytorch/pytorch/issues/127337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-05-25T03:31:16Z +issue,154187,pybind11 loading for c10::Scalar NYI,https://github.com/pytorch/pytorch/issues/154187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-05-24T02:47:13Z +issue,130114,[feature request] [discussion] Baseline ONNX interpreter / executor in python / PyTorch,https://github.com/pytorch/pytorch/issues/130114,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-05-24T01:53:15Z +issue,154272,"[FSDP2] for mixed precision, input casting can get blocked when cuda streams are full",https://github.com/pytorch/pytorch/issues/154272,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-05-24T01:03:40Z +issue,64208,"[feature request] ""Batched"" index_select (i.e. simplified torch.gather with not specifying full index)",https://github.com/pytorch/pytorch/issues/64208,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2025-05-23T21:55:23Z +issue,120115,`detectron2_maskrcnn` OOMs on eager with A100 40G.,https://github.com/pytorch/pytorch/issues/120115,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-05-23T20:42:49Z +issue,128971,Trying to use forward AD with _scaled_dot_product_flash_attention that does not support it because it has not been implemented yet.,https://github.com/pytorch/pytorch/issues/128971,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2025-05-23T17:19:35Z +issue,154028,Add clamped FP8 (E4M3) cast for overflow-safe inference,https://github.com/pytorch/pytorch/issues/154028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2025-05-23T16:12:33Z +issue,154096,CMake Error (CMAKE_MINIMUM_REQUIRED) when build 2.7.0,https://github.com/pytorch/pytorch/issues/154096,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2025-05-23T05:43:17Z +issue,153103,[Inductor] Investigate computing global amaxes via atomics instead of 2 stage reduction approach,https://github.com/pytorch/pytorch/issues/153103,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-05-23T04:08:29Z +issue,28777,cudnn.determinstic=True causes dilated convolution to be >10x slower,https://github.com/pytorch/pytorch/issues/28777,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2025-05-23T04:05:57Z +issue,149196,(Will PR) Multiprocessing with CUDA_VISIBLE_DEVICES seems to give the wrong device,https://github.com/pytorch/pytorch/issues/149196,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2025-05-23T03:12:57Z +issue,154017,[RFC] Generalized Per-Operator Device Capability Registry for PyTorch Operator Testing,https://github.com/pytorch/pytorch/issues/154017,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: privateuse1, module: rocm",scan when time allows,2025-05-22T17:35:09Z +issue,106602,onednn ops supported in pytorch,https://github.com/pytorch/pytorch/issues/106602,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: intel,scan when time allows,2025-05-22T17:10:24Z +issue,127915,Document the torch.fx.traceback.preserve_node_meta function,https://github.com/pytorch/pytorch/issues/127915,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:20Z +issue,127886,Document the torch.fx.experimental.proxy_tensor.make_fx function,https://github.com/pytorch/pytorch/issues/127886,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:20Z +issue,127917,Document the torch.cuda.profiler.start function,https://github.com/pytorch/pytorch/issues/127917,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:19Z +issue,112587,"Fix docstring errors in __init__.py, _tensor_docs.py, _meta_registrations.py, _tensor.py",https://github.com/pytorch/pytorch/issues/112587,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:16Z +issue,112585,"Fix docstring errors in _VF.py, _appdirs.py, hub.py, _classes.py, _storage_docs.py, _linalg_utils.py, torch_version.py, quasirandom.py, random.py, __future__.py, _lowrank.py, _vmap_internals.py, _sources.py, __config__.py, _lobpcg.py, _namedtensor_internals.py",https://github.com/pytorch/pytorch/issues/112585,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:16Z +issue,112588,"Fix docstring errors in _torch_docs.py, serialization.py, overrides.py, _utils.py",https://github.com/pytorch/pytorch/issues/112588,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:15Z +issue,112600,Fix docstring errors in loss.py,https://github.com/pytorch/pytorch/issues/112600,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:14Z +issue,112641,"Fix docstring errors in post_localSGD_optimizer.py, functional_sgd.py, _functional_collectives.py, optimizer.py, utils.py, api.py, server_process_global_profiler.py, functions.py, backend_registry.py, __init__.py, options.py, internal.py, functional_adam.py",https://github.com/pytorch/pytorch/issues/112641,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:13Z +issue,112638,"Fix docstring errors in spectral_ops_fuzz_test.py, simple_timeit.py, timer_interface.py, op_benchmark.py, _stubs.py, fuzzer.py, compare.py, compile.py, interp.py, hipify_python.py, common.py, end_to_end.py, timer.py, __init__.py, sparse_fuzzer.py, blas_compare_setup.py",https://github.com/pytorch/pytorch/issues/112638,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:13Z +issue,112644,"Fix docstring errors in default_hooks.py, post_localSGD_hook.py, debugging_hooks.py, utils.py, hierarchical_model_averager.py, optimizer_overlap_hooks.py, mixed_precision_hooks.py, quantization_hooks.py, ddp_zero_hook.py, __init__.py, powerSGD_hook.py, averagers.py",https://github.com/pytorch/pytorch/issues/112644,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:12Z +issue,112646,"Fix docstring errors in stream.py, pipe.py, blockpartition.py, microbatch.py, namespace.py, profile.py, tracker.py, portal.py, layout.py, __init__.py",https://github.com/pytorch/pytorch/issues/112646,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:11Z +issue,113186,"Fix docstring errors in shard.py, op_registry_utils.py, local_timer.py, file_based_local_timer.py, api.py, distributed.py, cycling_iterator.py, __init__.py, _utils.py, reshard.py, common_op_utils.py, elastic_distributed_sampler.py, utils.py, log_level.py, store.py, logging.py, sharder.py, metadata.py",https://github.com/pytorch/pytorch/issues/113186,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:10Z +issue,113122,Request to create a tutorial for loading a model dumped by `torch.export`,https://github.com/pytorch/pytorch/issues/113122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2025-05-22T17:10:10Z +issue,113187,"Fix docstring errors in embedding.py, _limiter_utils.py, _dynamo_utils.py, embedding_bag.py, tensor_ops.py, api.py, _internals.py, _common.py, init.py, _exec_order_utils.py, _traversal_utils.py, chunk_sharding_spec.py, _trace_utils.py",https://github.com/pytorch/pytorch/issues/113187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:09Z +issue,113188,"Fix docstring errors in _common_utils.py, _optim_utils.py, _wrap_utils.py, _unshard_param_utils.py, _fsdp_extensions.py, api.py, _debug_utils.py, _utils.py, wrap.py, sharded_grad_scaler.py",https://github.com/pytorch/pytorch/issues/113188,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:08Z +issue,113192,"Fix docstring errors in checkpoint_example.py, basic_strategy.py, op_schema.py, contract.py, redistribute.py, __init__.py, api.py, device_mesh.py, _utils.py, sharding_prop.py, random.py, checkpoint_activation.py, placement_types.py, fully_shard.py, parallel_mode.py, replicate.py",https://github.com/pytorch/pytorch/issues/113192,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:07Z +issue,113190,"Fix docstring errors in _state_dict_utils.py, _runtime_utils.py, _shard_utils.py",https://github.com/pytorch/pytorch/issues/113190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-22T17:10:07Z +issue,154086,[cuBLAS] Strange behavior with `out_dtype` argument in `torch.addmm`,https://github.com/pytorch/pytorch/issues/154086,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cublas, module: cuda, module: half",scan when time allows,2025-05-22T16:24:35Z +issue,102559,Mergebot should merge non-stacked PR,https://github.com/pytorch/pytorch/issues/102559,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-05-22T13:31:11Z +issue,153471,RFC for vector length agnostic SVE Vectorized class,https://github.com/pytorch/pytorch/issues/153471,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: vectorization",scan when time allows,2025-05-22T10:57:12Z +issue,148451,BC-linter should ignore testing/linter/adapters/,https://github.com/pytorch/pytorch/issues/148451,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2025-05-22T10:22:55Z +issue,113761,[DTensor][TP] 1-way TP raises index out of range,https://github.com/pytorch/pytorch/issues/113761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2025-05-22T10:07:12Z +issue,152150,"[Scaled MM] Update to support on B200 TN, NT, NN, TT Layouts are supported",https://github.com/pytorch/pytorch/issues/152150,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: floatx (formerly float8), module: performance",scan when time allows,2025-05-21T18:48:56Z +issue,147428,Add a config to allow print and generate all recompile reasons and not stop at first.,https://github.com/pytorch/pytorch/issues/147428,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile ux, module: dynamo, oncall: pt2",scan when time allows,2025-05-21T18:20:31Z +issue,122550,Adam optimizer：RuntimeError: t == DeviceType::CUDA INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/122550,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-05-21T17:24:49Z +issue,153638,Can't run lintrunner locally on Windows machine,https://github.com/pytorch/pytorch/issues/153638,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: windows",scan when time allows,2025-05-21T04:33:55Z +issue,153649,Use opmath_t and not double compute in fused optimizers,https://github.com/pytorch/pytorch/issues/153649,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: optimizer, module: performance",scan when time allows,2025-05-20T20:57:09Z +issue,153842,Duplicated `milestones` keys in `torch.optim.lr_scheduler.MultiStepLr` after loading a checkpoint,https://github.com/pytorch/pytorch/issues/153842,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, oncall: distributed checkpointing",scan when time allows,2025-05-20T18:50:28Z +issue,153832,export torch.cond fails at capturing identity,https://github.com/pytorch/pytorch/issues/153832,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-05-20T15:55:21Z +issue,153930,[FX] Support `torch.cond` in `torch.fx.symbolic_trace`,https://github.com/pytorch/pytorch/issues/153930,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-05-20T15:43:32Z +issue,153826,Unable to reproduce dynamo bechmark script results,https://github.com/pytorch/pytorch/issues/153826,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: benchmark, module: rocm, module: tests",scan when time allows,2025-05-20T15:28:53Z +issue,153938,Subtests not properly accounted for in test reports (e.g. on CI),https://github.com/pytorch/pytorch/issues/153938,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: testing",scan when time allows,2025-05-20T14:42:07Z +issue,153532,[RFC] [Inductor] Custom pass registration interface,https://github.com/pytorch/pytorch/issues/153532,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-05-20T13:24:38Z +issue,118173,ModuleNotFoundError: pynvml does not seem to be installed or it can't be imported.,https://github.com/pytorch/pytorch/issues/118173,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: windows",scan when time allows,2025-05-20T11:50:35Z +issue,133984,cudagraph capture p2p ops fail,https://github.com/pytorch/pytorch/issues/133984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, oncall: distributed",scan when time allows,2025-05-20T09:12:13Z +issue,111187,"torch.distributed.DistBackendError: NCCL error in: ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:1331, unhandled cuda error (run with NCCL_DEBUG=INFO for details)",https://github.com/pytorch/pytorch/issues/111187,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: wsl, oncall: distributed",scan when time allows,2025-05-20T06:22:54Z +issue,8374,[Feature request] Add torch.multiprocessing.Pipe,https://github.com/pytorch/pytorch/issues/8374,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: multiprocessing",scan when time allows,2025-05-20T03:54:36Z +issue,148065,Support alpha=inf consistently for torch.celu,https://github.com/pytorch/pytorch/issues/148065,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend",scan when time allows,2025-05-20T02:43:42Z +issue,76071,HIPFFT_EXEC_FAILED when using AMD GPU run FFT,https://github.com/pytorch/pytorch/issues/76071,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: fft",scan when time allows,2025-05-19T12:28:48Z +issue,153542,torch.set_ on a view does not sever view relation,https://github.com/pytorch/pytorch/issues/153542,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2025-05-19T08:53:10Z +issue,153594,Significant DataLoader worker startup delay and sequential-like initialization with num_workers > 0 on Windows (using 'spawn'),https://github.com/pytorch/pytorch/issues/153594,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: performance, module: windows",scan when time allows,2025-05-19T04:41:19Z +issue,105325,Padded tensor subclass,https://github.com/pytorch/pytorch/issues/105325,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-05-18T20:48:28Z +issue,150861,Allow non-Tensor subclass in `torch.Tensor._make_wrapper_subclass`,https://github.com/pytorch/pytorch/issues/150861,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-05-18T10:52:58Z +issue,105644,Tensor subclass is not preserved during backward with gradient checkpointing,https://github.com/pytorch/pytorch/issues/105644,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: activation checkpointing",scan when time allows,2025-05-17T20:07:02Z +issue,153606,native/indexingUtils.h `AdvancedIndex` has internal linkage,https://github.com/pytorch/pytorch/issues/153606,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: static linking",scan when time allows,2025-05-16T17:01:37Z +issue,73016,Unable to build and use libtorch function via pybind11: undefined symbol error upon import,https://github.com/pytorch/pytorch/issues/73016,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: mkldnn",scan when time allows,2025-05-16T10:10:19Z +issue,110056,torch.onnx.export causes floating point exception with core dump for empty slice assignment,https://github.com/pytorch/pytorch/issues/110056,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-05-16T06:30:02Z +issue,150476,Better test coverage on _inductor/scheduler.py,https://github.com/pytorch/pytorch/issues/150476,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-05-16T04:26:50Z +issue,153673,[FSDP2] revisit NCCL group coalescing vs copy-in/copy-out,https://github.com/pytorch/pytorch/issues/153673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-16T00:48:28Z +issue,152963,DTensor support for dynamic shapes is soft,https://github.com/pytorch/pytorch/issues/152963,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: pt2",scan when time allows,2025-05-15T22:54:18Z +issue,146956,[c10d] Consolidate watchdog threads,https://github.com/pytorch/pytorch/issues/146956,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-05-15T22:42:51Z +issue,151037,[release] CPU perf benchmark latency increase for 2.6->2.7 on c5.24xlarge and A100 instances,https://github.com/pytorch/pytorch/issues/151037,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: intel, module: performance, oncall: releng",scan when time allows,2025-05-15T16:08:53Z +issue,143176,Compiled Autograd + Activation Checkpointing/Offloading,https://github.com/pytorch/pytorch/issues/143176,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: compiled autograd, oncall: pt2",scan when time allows,2025-05-14T18:35:13Z +issue,90633,torch.min document not up to date,https://github.com/pytorch/pytorch/issues/90633,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-05-14T18:15:30Z +issue,151912,bfloat16 numerical errors for SDPA math backend,https://github.com/pytorch/pytorch/issues/151912,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-05-14T15:50:05Z +issue,152829,Don't hardcoded support for DTensor to_local/from_local/redistribute into dynamo,https://github.com/pytorch/pytorch/issues/152829,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2",scan when time allows,2025-05-13T22:55:47Z +issue,153302,[FSDP2] avoid GPU OOM for reshard_after_forward=int with shared post_forward_mesh,https://github.com/pytorch/pytorch/issues/153302,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-13T18:20:29Z +issue,152337,Unusually slow draft_export time,https://github.com/pytorch/pytorch/issues/152337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-05-13T18:09:22Z +issue,153242,[RFC] Add a SlimTensor representation to help AOTInductor generate standalone binaries,https://github.com/pytorch/pytorch/issues/153242,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-05-13T16:12:07Z +issue,153139,Floating Point exception in Convolution with disabled SMT,https://github.com/pytorch/pytorch/issues/153139,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: dependency bug, module: nn, module: nnpack",scan when time allows,2025-05-13T09:09:02Z +issue,153360,torch.aminmax reducing over all dims should not accept keepdim,https://github.com/pytorch/pytorch/issues/153360,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2025-05-13T02:04:04Z +issue,153361,MANIFEST.in is not being used,https://github.com/pytorch/pytorch/issues/153361,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-05-13T01:00:35Z +issue,136946,DTensor does not support `nn.init.eye_`,https://github.com/pytorch/pytorch/issues/136946,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2025-05-12T20:26:39Z +issue,121160,RFCS review request: fast Viterbi Decoding,https://github.com/pytorch/pytorch/issues/121160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-05-12T14:30:58Z +issue,91252,Open file leak when dataloader is using persistent_workers and pin_memory AND you create multiple dataloaders.,https://github.com/pytorch/pytorch/issues/91252,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2025-05-12T12:54:55Z +issue,98115,Request to cherrypick a fix into v1.13.1 (v1.8 has a CVE),https://github.com/pytorch/pytorch/issues/98115,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-05-12T08:58:33Z +issue,72643,upstream `apex.normalization.FusedRMSNorm`,https://github.com/pytorch/pytorch/issues/72643,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2025-05-12T06:34:14Z +issue,134576,Would it be possible to have a CMAKE option USE_SYSTEM_FMT,https://github.com/pytorch/pytorch/issues/134576,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-05-11T18:15:02Z +issue,73466,[feature request] torch.hub.load_state_dict_from_url to be replaced by a new good general download-a-file function and to also support local paths and google drive links / private github release links,https://github.com/pytorch/pytorch/issues/73466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2025-05-11T10:23:08Z +issue,147179,[FSDP2] OOM when use integer `reshard_after_forward` that smaller than DP size,https://github.com/pytorch/pytorch/issues/147179,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-05-10T01:27:05Z +issue,150864,[dynamo] `torch.compile` graph breaks on `setattr` of type objects,https://github.com/pytorch/pytorch/issues/150864,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-05-09T20:32:59Z +issue,153041,gen_alias_from_base ruins the result of view after inductor generated a copy for the results of the view operations.,https://github.com/pytorch/pytorch/issues/153041,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-05-09T20:06:19Z +issue,152746,"[FSDP2] fully_shard(mesh=(shard, shard)) for intra and inter node all-gathers",https://github.com/pytorch/pytorch/issues/152746,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-09T19:26:43Z +issue,153210,Inconsistent Error Handling in `torch.addcmul` with `int16` Tensors,https://github.com/pytorch/pytorch/issues/153210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: error checking, module: int overflow",scan when time allows,2025-05-09T15:28:21Z +issue,107999,`upsample_bilinear2d_backward_out_cuda` is nondeterministic,https://github.com/pytorch/pytorch/issues/107999,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2025-05-09T06:58:02Z +issue,148086,FSDP2 without sharding works slower than DDP,https://github.com/pytorch/pytorch/issues/148086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-08T23:49:13Z +issue,76082,Eliminate uses of deprecated `FindCUDA.cmake`,https://github.com/pytorch/pytorch/issues/76082,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-05-08T14:14:19Z +issue,123447,dist.barrier() hangs after calling async_save,https://github.com/pytorch/pytorch/issues/123447,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-08T07:24:07Z +issue,148306,torch.vmap incompatibility with DLPack functions,https://github.com/pytorch/pytorch/issues/148306,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dlpack, module: functorch, module: vmap",scan when time allows,2025-05-08T07:05:18Z +issue,153025,Multiple CUDA graphs utilizing multiple CUDA GPUs encounter illegal memory access during replay,https://github.com/pytorch/pytorch/issues/153025,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2025-05-07T19:09:39Z +issue,152962,TestNestedTensorOpInfoCUDA.test_compile_backward_matmul_cuda_float32 Test Failure,https://github.com/pytorch/pytorch/issues/152962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: tests",scan when time allows,2025-05-07T16:58:53Z +issue,150836,`Aborted (core dumped)` in `torch.cuda.nccl.reduce`,https://github.com/pytorch/pytorch/issues/150836,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-07T10:10:32Z +issue,151322,[CI] No workflows scheduled on PRs,https://github.com/pytorch/pytorch/issues/151322,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: flaky-tests, module: third_party",scan when time allows,2025-05-06T22:28:04Z +issue,152994,[dynamo] Actually support functools.lru_cache,https://github.com/pytorch/pytorch/issues/152994,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-05-06T21:20:16Z +issue,152884,UNSTABLE Lint / Link checks / Lint URLs / linux-job,https://github.com/pytorch/pytorch/issues/152884,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-05-06T17:14:25Z +issue,152426,"[Manylinux 2.28] Migrate Docker container to use gcc 13, CUDA 12.6 and gcc14 CUDA 12.8",https://github.com/pytorch/pytorch/issues/152426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-05-06T17:01:03Z +issue,152869,DISABLED testAssertNotRegex (__main__.CPythonTest_Assertions),https://github.com/pytorch/pytorch/issues/152869,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: flaky-tests, oncall: pt2",scan when time allows,2025-05-06T12:55:34Z +issue,152752,Checkpoint sequential doesn't raise clear error when segments is greater than number of functions,https://github.com/pytorch/pytorch/issues/152752,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2025-05-06T12:26:50Z +issue,58846,Simplification of pruned models,https://github.com/pytorch/pytorch/issues/58846,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-05-06T07:51:58Z +issue,152736,[MPS] TensorIterator and accuracy,https://github.com/pytorch/pytorch/issues/152736,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-05-05T18:13:21Z +issue,152527,can't reconstruct the communication group using PyTorch.,https://github.com/pytorch/pytorch/issues/152527,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-05-05T08:24:12Z +issue,152645,ProcessGroupGloo.allgather_into_tensor_coalesced crashes with CUDA tensors,https://github.com/pytorch/pytorch/issues/152645,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-05T07:41:30Z +issue,66200,An exported onnx model can't reduce on dim with value of 0 if 'keepdims' is false,https://github.com/pytorch/pytorch/issues/66200,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-05-04T08:50:41Z +issue,151182,A problem discovered when computing complex matrices in deep neural networks,https://github.com/pytorch/pytorch/issues/151182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",scan when time allows,2025-05-03T09:03:18Z +issue,121857,AdamW(fused=True) slower than unfused AdamW,https://github.com/pytorch/pytorch/issues/121857,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: optimizer, module: performance",scan when time allows,2025-05-02T21:41:53Z +issue,117540,Unwanted Warning in lr_scheduler.step(),https://github.com/pytorch/pytorch/issues/117540,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: lrscheduler",scan when time allows,2025-05-02T21:38:46Z +issue,151936,fully_shard() for huggingface 72B model: pytorch caches too much GPU memory,https://github.com/pytorch/pytorch/issues/151936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-05-02T19:28:29Z +issue,114357,Pytorch DDP across nodes: self._store = TCPStore( # type: ignore[call-arg] RuntimeError: Stop_waiting response is expected,https://github.com/pytorch/pytorch/issues/114357,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-05-02T18:43:21Z +issue,152699,[CI] [anaconda] Triton windows build,https://github.com/pytorch/pytorch/issues/152699,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-05-02T17:22:35Z +issue,152607,Loops impacting output when utilizing hooks,https://github.com/pytorch/pytorch/issues/152607,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-05-02T13:34:44Z +issue,152647,Check for if two tensors are overall similar instead of bitwise similar?,https://github.com/pytorch/pytorch/issues/152647,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2025-05-02T13:30:47Z +issue,9674,The state of sparse Tensors,https://github.com/pytorch/pytorch/issues/9674,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-05-01T23:53:10Z +issue,147570,`view()` + modify-in-place fails silently with DTensor,https://github.com/pytorch/pytorch/issues/147570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-05-01T20:29:20Z +issue,152097,"When using torch to convert to oxxn model, testing the inference results with actual images shows tensor mismatch",https://github.com/pytorch/pytorch/issues/152097,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-05-01T18:34:36Z +issue,152550,MPS varying seq len SDPA memory leak,https://github.com/pytorch/pytorch/issues/152550,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: mps, module: sdpa",scan when time allows,2025-05-01T17:50:30Z +issue,152437,`nn.CrossEntropyLoss` accepts negative target probabilities,https://github.com/pytorch/pytorch/issues/152437,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn, module: performance",scan when time allows,2025-05-01T14:53:50Z +issue,127928,Ability to do aot/inductor compilation from a jit model (or torch.exported model),https://github.com/pytorch/pytorch/issues/127928,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",scan when time allows,2025-05-01T12:05:20Z +issue,126518,missing docs for torch.Tag,https://github.com/pytorch/pytorch/issues/126518,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: library",scan when time allows,2025-04-30T20:47:05Z +issue,129657,Implementation of a numerically stable log(1 - softmax) function in PyTorch,https://github.com/pytorch/pytorch/issues/129657,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: python frontend",scan when time allows,2025-04-30T15:17:04Z +issue,67970,Error when using torch.fx on bert,https://github.com/pytorch/pytorch/issues/67970,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-04-30T01:38:26Z +issue,152489,UNSTABLE Lint / Lint URLs / linux-job,https://github.com/pytorch/pytorch/issues/152489,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-04-30T00:22:28Z +issue,152127,[C10D] Make collectives backwards throw an error,https://github.com/pytorch/pytorch/issues/152127,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-30T00:03:20Z +issue,151579,[BE] Consolidate PR labeling logic,https://github.com/pytorch/pytorch/issues/151579,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: devx",scan when time allows,2025-04-29T21:02:57Z +issue,152257,[FR] Support BSHM-layout scaled_dot_product_attention without transpose.,https://github.com/pytorch/pytorch/issues/152257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-04-29T19:56:32Z +issue,152381,Outdated contents in dynamo overview,https://github.com/pytorch/pytorch/issues/152381,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-04-29T00:22:57Z +issue,151776,[dynamo] Some inefficiencies around handling __torch_function__,https://github.com/pytorch/pytorch/issues/151776,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-04-28T23:45:10Z +issue,152269,Arbitrary Code Execution Risk in `torch.distributed.utils.overload` When Misused in Type Annotations,https://github.com/pytorch/pytorch/issues/152269,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-28T22:24:39Z +issue,117350,CUDA deps cannot be preloaded under Bazel,https://github.com/pytorch/pytorch/issues/117350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: build",scan when time allows,2025-04-28T22:13:31Z +issue,107573,caching keys+values in TransformerDecoderLayer for faster inference,https://github.com/pytorch/pytorch/issues/107573,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-04-26T17:51:07Z +issue,106624,"[feature request] [ux] Frontend methods for fused elementwise affine transform: mul+add+dtype convert + support `integer_tensor.mul_(float_constant)` and `float_tensor.mul(some_constant, out = integer_tensor)` maybe via new args `rounding_mode=...` and `dtype=...` + maybe support OpenCV-style saturated dtype conversions (e.g. `clamp_` before conversion)",https://github.com/pytorch/pytorch/issues/106624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: type promotion",scan when time allows,2025-04-26T10:18:31Z +issue,57947,`torch.lerp` to support argument type promotion / broadcasting - including of `input` / `end` arguments,https://github.com/pytorch/pytorch/issues/57947,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: type promotion",scan when time allows,2025-04-26T10:04:34Z +issue,146913,Add a `TORCH_LOGS_RANK=0` env var that integrates with `TORCH_LOGS`,https://github.com/pytorch/pytorch/issues/146913,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: logging, oncall: pt2",scan when time allows,2025-04-26T02:52:37Z +issue,150574,[mem profiler] mem fragmentation and pynvml view,https://github.com/pytorch/pytorch/issues/150574,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: memory usage",scan when time allows,2025-04-25T23:43:51Z +issue,152152,padding_mode `reflect` works different from others in Conv,https://github.com/pytorch/pytorch/issues/152152,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",scan when time allows,2025-04-25T19:59:15Z +issue,152188,The input for layers other than the first layer should be the hidden state from the previous layer.,https://github.com/pytorch/pytorch/issues/152188,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2025-04-25T19:57:21Z +issue,151592,Xcode 16+: duplicate LC_RPATH '@loader_path',https://github.com/pytorch/pytorch/issues/151592,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-04-25T10:11:40Z +issue,141094,plus prefix-less wheels not being published on https://download.pytorch.org/whl/torch/ for linux x86_64,https://github.com/pytorch/pytorch/issues/141094,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-04-24T22:00:57Z +issue,151960,[rfc][c10d] First Class Object-Oriented ProcessGroup and Store APIs,https://github.com/pytorch/pytorch/issues/151960,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-04-24T18:53:47Z +issue,143759,compiled autograd tests should use expecttest,https://github.com/pytorch/pytorch/issues/143759,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, module: tests, oncall: pt2",scan when time allows,2025-04-24T18:18:49Z +issue,145693,AMD MI300A Unified Memory Support,https://github.com/pytorch/pytorch/issues/145693,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-04-24T17:56:44Z +issue,151978,torch.gather returns incorrect output on cuda after unsqueezing and expanding with double precision,https://github.com/pytorch/pytorch/issues/151978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2025-04-24T16:50:32Z +issue,151858,construcing DTensor on a 2D device mesh SIGTERMs,https://github.com/pytorch/pytorch/issues/151858,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-04-24T15:50:12Z +issue,108108,[BC BREAKING] Change default behavior of scaled_dot_product_attention's causal masking alignment,https://github.com/pytorch/pytorch/issues/108108,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: nn, module: sdpa",scan when time allows,2025-04-24T09:41:13Z +issue,151878,Potential error in calling scaled_dot_product_attention in pytorch/torch/testing/_internal/distributed/_tensor /common_dtensor.py ?,https://github.com/pytorch/pytorch/issues/151878,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: context parallel, module: dtensor, oncall: distributed",scan when time allows,2025-04-24T06:37:51Z +issue,145263,[Pipelining] Problem using `torch.distributed.pipelining` on `Gemma2ForCausalLM`,https://github.com/pytorch/pytorch/issues/145263,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T21:02:31Z +issue,151759,"""_get_pg_default_device"" deprecated warning in ""Getting Started with Distributed Checkpoint (DCP)""",https://github.com/pytorch/pytorch/issues/151759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:54:49Z +issue,148257,[FSDP2] HSDP with globally sharded fp32 weights and optimizer states,https://github.com/pytorch/pytorch/issues/148257,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-04-23T20:54:49Z +issue,148269,Should DTensor support `Shard()` placement without dim requirement?,https://github.com/pytorch/pytorch/issues/148269,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-04-23T20:54:48Z +issue,148162,Specifying device_id in init_process_group causes tensor parallel + pipeline parallel to fail,https://github.com/pytorch/pytorch/issues/148162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:54:47Z +issue,148150,[pytorch elastic] [RHEL] multiple processes call to dist.destroy_process_group() cause an RuntimeError: CUDA error: CUDA-capable device(s) is/are busy or unavailable,https://github.com/pytorch/pytorch/issues/148150,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:54:47Z +issue,148048,[RFC] Customization of ElasticAgent for fault-tolerance and node-replacement in big ddp job,https://github.com/pytorch/pytorch/issues/148048,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:54:46Z +issue,147980,Torch distribute module when local dtensor is already on the correct device,https://github.com/pytorch/pytorch/issues/147980,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-04-23T20:54:45Z +issue,147793,context_parallel fails with plain sdpa kernel SDPBackend.MATH,https://github.com/pytorch/pytorch/issues/147793,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: context parallel, module: sdpa, oncall: distributed",scan when time allows,2025-04-23T20:54:44Z +issue,147732,[Distribute] len(input_specs) == len(input_args_strategy) AssertionError,https://github.com/pytorch/pytorch/issues/147732,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:54:43Z +issue,147677,distributed_c10d.broadcast causing unexpected CUDA oom error,https://github.com/pytorch/pytorch/issues/147677,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-04-23T20:54:43Z +issue,147631,torch.distributed.init_process_group with unspecified backend out of date,https://github.com/pytorch/pytorch/issues/147631,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-04-23T20:54:42Z +issue,147563,torch.distributed.elastic.multiprocessing.start_process description does not reflect API,https://github.com/pytorch/pytorch/issues/147563,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2025-04-23T20:54:42Z +issue,147502,Slow first time torch.distributed op with ray,https://github.com/pytorch/pytorch/issues/147502,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-04-23T20:54:41Z +issue,147348,[pipelining] AttributeError: 'InterpreterModule' object has no attribute,https://github.com/pytorch/pytorch/issues/147348,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pipelining, oncall: distributed",scan when time allows,2025-04-23T20:54:41Z +issue,147263,How to trigger several independent communications simultaneously?,https://github.com/pytorch/pytorch/issues/147263,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:54:39Z +issue,147141,NCCL Update 2.25.1 with CUDA 12.4 build is failing in CI,https://github.com/pytorch/pytorch/issues/147141,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2025-04-23T20:54:37Z +issue,149177,[Dist] Async op isend and irecv bug,https://github.com/pytorch/pytorch/issues/149177,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-04-23T20:42:57Z +issue,149100,Memory leak when using get_model_state_dict with FSDP-sharded models,https://github.com/pytorch/pytorch/issues/149100,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-04-23T20:42:57Z +issue,148831,FSDP2 and autocast compatibility issue,https://github.com/pytorch/pytorch/issues/148831,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-04-23T20:42:57Z +issue,148756,FSPD ValueError: expected to be in states [] but current state is TrainingState.IDLE,https://github.com/pytorch/pytorch/issues/148756,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-04-23T20:42:56Z +issue,149049,"`pytroch.distribute` should support ""meta"" device tensors",https://github.com/pytorch/pytorch/issues/149049,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:42:55Z +issue,148837,NVLS support in Pytorch,https://github.com/pytorch/pytorch/issues/148837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:42:55Z +issue,148504,[FSDP2] improve error msg for duplicate wraps,https://github.com/pytorch/pytorch/issues/148504,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-04-23T20:42:53Z +issue,148689,NUMA Binding Integration with torchrun,https://github.com/pytorch/pytorch/issues/148689,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:42:52Z +issue,148440,torch.distributed hangs between 2 Mac Devices,https://github.com/pytorch/pytorch/issues/148440,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, oncall: distributed",scan when time allows,2025-04-23T20:42:51Z +issue,148302,broadcast_object_list not release GPU,https://github.com/pytorch/pytorch/issues/148302,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:42:48Z +issue,151457,"When distributed.destroy_process_group() is executed, new memory usage will be generated on device 0, which may cause OOM under extreme conditions and thus abnormal exit.",https://github.com/pytorch/pytorch/issues/151457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:37:15Z +issue,151240,[funcol] wait() breaks the chain for backwards,https://github.com/pytorch/pytorch/issues/151240,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:37:15Z +issue,150585,[distributed] Crash when trying to use default PG after creating new PG,https://github.com/pytorch/pytorch/issues/150585,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:37:13Z +issue,150096,FSDP OOM during `sync_params_and_buffers`,https://github.com/pytorch/pytorch/issues/150096,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-04-23T20:37:12Z +issue,150336,[parallelize_module] uneven sharding + use_local_output breaks,https://github.com/pytorch/pytorch/issues/150336,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-04-23T20:37:11Z +issue,150140,"FSDP2 issue with mp_policy, checkpoint() and float input",https://github.com/pytorch/pytorch/issues/150140,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-04-23T20:37:10Z +issue,150049,2.7 test docker image has NCCL with older CUDA,https://github.com/pytorch/pytorch/issues/150049,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, oncall: distributed, oncall: releng",scan when time allows,2025-04-23T20:37:09Z +issue,149984,"ncclSystemError: System call (e.g. socket, malloc) or external library call failed or device error",https://github.com/pytorch/pytorch/issues/149984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2025-04-23T20:37:08Z +issue,149934,inconsistant behavior to pass tensor/tensor_list from torch.distributed module to cpp API,https://github.com/pytorch/pytorch/issues/149934,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:37:07Z +issue,149762,terminate called after throwing an instance of 'c10::Error',https://github.com/pytorch/pytorch/issues/149762,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2025-04-23T20:37:06Z +issue,149516,```StateDictOptions``` in combination with ```cpu_offload=True``` and ```strict=False``` not working,https://github.com/pytorch/pytorch/issues/149516,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:37:05Z +issue,151556,"Compatibility Issues Between DCP, LoRA, and Hugging Face Models",https://github.com/pytorch/pytorch/issues/151556,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:35:49Z +issue,150943,"[RFC][TCPStore] advanced store operations (queues, pub/sub)",https://github.com/pytorch/pytorch/issues/150943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-04-23T20:35:12Z +issue,151159,Failed to destroy or init process group after calling _abort_process_group,https://github.com/pytorch/pytorch/issues/151159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:34:11Z +issue,151269,[PP] Hang when num_microbatches < stages for Interleaved1F1B,https://github.com/pytorch/pytorch/issues/151269,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pipelining, oncall: distributed",scan when time allows,2025-04-23T20:27:04Z +issue,151222,[FSDP] Detail information about parameters when raising errors,https://github.com/pytorch/pytorch/issues/151222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-23T20:09:31Z +issue,151965,"[rfc][c10d] RDMA APIs (read/write, rkey)",https://github.com/pytorch/pytorch/issues/151965,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-04-23T18:46:24Z +issue,151717,[NJT] Dropout with p=0 still drops values,https://github.com/pytorch/pytorch/issues/151717,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-04-23T14:27:03Z +issue,4474,[docs] Return value type hint to be specified in docs for torch.is_grad_enabled (i.e. -> bool),https://github.com/pytorch/pytorch/issues/4474,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-04-23T09:52:15Z +issue,21176,Adding a method called `T` in native_functions causes undefined behavior on Windows,https://github.com/pytorch/pytorch/issues/21176,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2025-04-23T04:47:10Z +issue,151793,Create decomp for searchsorted,https://github.com/pytorch/pytorch/issues/151793,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-04-22T23:20:45Z +issue,147513,[RFC] Request for Feedback and Review on PRs Adding RISC-V and RVV Support,https://github.com/pytorch/pytorch/issues/147513,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpu, module: risc-v",scan when time allows,2025-04-22T22:06:27Z +issue,145949,[CUDA][Blackwell] Blackwell Tracking Issue,https://github.com/pytorch/pytorch/issues/145949,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-04-22T22:05:11Z +issue,141616,Clarify that `compile(module)` only affects the `forward` method.,https://github.com/pytorch/pytorch/issues/141616,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: pt2",scan when time allows,2025-04-22T15:32:06Z +issue,151113,Failure in test_vmap_autograd_grad_nn_functional_conv2d_cpu_float32,https://github.com/pytorch/pytorch/issues/151113,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: tests, module: vmap",scan when time allows,2025-04-22T12:07:44Z +issue,151561,2.6: vmap doesn't vectorize all args,https://github.com/pytorch/pytorch/issues/151561,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: functorch, module: vmap",scan when time allows,2025-04-22T03:15:11Z +issue,151555,[Inductor] Convolution auto-generation through Triton backend,https://github.com/pytorch/pytorch/issues/151555,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-04-21T22:28:48Z +issue,144206,Error when compiling document,https://github.com/pytorch/pytorch/issues/144206,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2025-04-21T21:54:41Z +issue,150364,AOTI doesn't error if dynamic shape gets specialized during lowering,https://github.com/pytorch/pytorch/issues/150364,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",scan when time allows,2025-04-21T18:41:13Z +issue,124025,Different results for single sample vs. batched,https://github.com/pytorch/pytorch/issues/124025,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: collect_env.py, module: docs, module: numerical-stability",scan when time allows,2025-04-21T17:20:14Z +issue,137888,Inconsistent digamma result with other frameworks (1e-6 difference),https://github.com/pytorch/pytorch/issues/137888,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2025-04-21T08:01:38Z +issue,129140,[RFC][C10D] Avoid creating new nccl communicator for each P2P pair,https://github.com/pytorch/pytorch/issues/129140,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-21T02:58:38Z +issue,137080,Discrepancy between torch.nn.functional.linear and manual computation with bias under torch.autocast,https://github.com/pytorch/pytorch/issues/137080,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: nn, module: numerical-stability",scan when time allows,2025-04-20T09:19:33Z +issue,37386,Add BufferDict container,https://github.com/pytorch/pytorch/issues/37386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-04-19T08:50:14Z +issue,148564,[Feature Request] Dynamic shapes API requires spec for all arguments.,https://github.com/pytorch/pytorch/issues/148564,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-04-18T22:59:04Z +issue,71300,"Kernel fusion for Gather, Apply, Scatter (GAS) model",https://github.com/pytorch/pytorch/issues/71300,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops, module: sparse",scan when time allows,2025-04-17T16:08:04Z +issue,150252,RuntimeError: Subtracting Reconstructed Jagged Nested Tensor Fails with Shape Mismatch,https://github.com/pytorch/pytorch/issues/150252,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nestedtensor",scan when time allows,2025-04-17T14:25:19Z +issue,151463,Inconsistent Output from nn.Conv2d with padding_mode='circular' When Using MKLDNN on a AVX2_Processor,https://github.com/pytorch/pytorch/issues/151463,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: mkldnn",scan when time allows,2025-04-17T13:50:57Z +issue,151559,SDPA doesn't document requirement for head dimension in masks,https://github.com/pytorch/pytorch/issues/151559,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-04-17T13:47:59Z +issue,149837,"Intermittent ""AssertionError: can only test a child process"" Warning with PyTorch DataLoader on Colab",https://github.com/pytorch/pytorch/issues/149837,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2025-04-17T09:34:28Z +issue,132395,`torch.multinomial` generates incorrect distribution,https://github.com/pytorch/pytorch/issues/132395,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: distributions, module: intel, module: numerical-stability",scan when time allows,2025-04-17T01:58:40Z +issue,151402,"RuntimeError: curPos <= (kUpperBound - kAppendInterval) INTERNAL ASSERT FAILED at ""../torch/csrc/jit/ir/ir.cpp"":698, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/151402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-04-16T14:53:44Z +issue,151286,vmap- out of memory,https://github.com/pytorch/pytorch/issues/151286,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vmap,scan when time allows,2025-04-16T14:49:03Z +issue,149110,[XPU] Enable Windows CI/CD test for XPU,https://github.com/pytorch/pytorch/issues/149110,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: xpu",scan when time allows,2025-04-15T20:59:27Z +issue,150571,Improve speed of pytorch docs build,https://github.com/pytorch/pytorch/issues/150571,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2025-04-15T20:38:21Z +issue,149872,Do all lazy imports for torch.compile in one place?,https://github.com/pytorch/pytorch/issues/149872,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-04-15T17:25:26Z +issue,150961,[ued] `torch.compile` cause more than 2x slow down with HF diffuser GGUF Auraflow,https://github.com/pytorch/pytorch/issues/150961,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-04-14T22:51:33Z +issue,151015,update user defined triton kernel table to include strict vs non-strict difference,https://github.com/pytorch/pytorch/issues/151015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: user triton, oncall: pt2",scan when time allows,2025-04-14T22:46:59Z +issue,150963,[ued] `torch.compile` yields lower latency when compiling transformer blocks only for ComfyUI GGUF Flux,https://github.com/pytorch/pytorch/issues/150963,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-04-14T22:46:04Z +issue,151251,Actually support LOAD_BUILD_CLASS,https://github.com/pytorch/pytorch/issues/151251,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: graph breaks, oncall: pt2",scan when time allows,2025-04-14T18:45:17Z +issue,146889,How to customize a torch.Tensor() method to access the underlying data structure of a PyTorch tensor.,https://github.com/pytorch/pytorch/issues/146889,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-04-14T17:40:25Z +issue,150928,ShardTensor gather will encounter an error when a local tensor on certain ranks has zero elements,https://github.com/pytorch/pytorch/issues/150928,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-04-14T17:26:47Z +issue,32137,torch.nn.functional.normalize epsilon too small for half precision,https://github.com/pytorch/pytorch/issues/32137,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: nn, module: norms and normalization, module: numerical-stability",scan when time allows,2025-04-14T16:57:30Z +issue,39178,Add support for rsample to MixtureSameFamily Distribution,https://github.com/pytorch/pytorch/issues/39178,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-04-14T15:25:06Z +issue,149194,LSTM slow on PackedSequence,https://github.com/pytorch/pytorch/issues/149194,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2025-04-14T10:42:35Z +issue,28440,Python/C++ API Parity: torch.optim optimizers,https://github.com/pytorch/pytorch/issues/28440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-04-14T10:33:37Z +issue,41667,[Feature Request] named tensor support for `tensordot`(tensor contract),https://github.com/pytorch/pytorch/issues/41667,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2025-04-14T03:10:05Z +issue,149453,Support launch_enter and launch_exit hooks,https://github.com/pytorch/pytorch/issues/149453,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-04-11T22:50:58Z +issue,150124,torch.unique undocumented behaviour,https://github.com/pytorch/pytorch/issues/150124,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-04-11T18:37:00Z +issue,148232,Add Structured Knowledge Accumulation (SKA) Layer to PyTorch,https://github.com/pytorch/pytorch/issues/148232,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-04-11T15:05:06Z +issue,83510,Cdist backward dependent on compute_mode,https://github.com/pytorch/pytorch/issues/83510,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2025-04-10T18:23:40Z +issue,149440,[pt2] Support statically launching triton compiled cuda kernels,https://github.com/pytorch/pytorch/issues/149440,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: user triton, oncall: pt2",scan when time allows,2025-04-10T16:59:38Z +issue,68272,Query CUDA version LibTorch has been compiled with,https://github.com/pytorch/pytorch/issues/68272,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-04-10T12:27:06Z +issue,17835,PyPy support,https://github.com/pytorch/pytorch/issues/17835,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-04-10T03:36:52Z +issue,148743,"torch.unique(t, dim=1, return_counts=True) does not work",https://github.com/pytorch/pytorch/issues/148743,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-04-09T21:44:38Z +issue,148536,Computing only the n first rows of a distance matrix with pdist,https://github.com/pytorch/pytorch/issues/148536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-04-09T20:46:09Z +issue,150922,StrictMinMaxConstraint issue in pytorch 2.4.0,https://github.com/pytorch/pytorch/issues/150922,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-04-09T20:07:11Z +issue,149304,Machine with isolated cores is not respected by torch,https://github.com/pytorch/pytorch/issues/149304,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multiprocessing",scan when time allows,2025-04-09T19:26:49Z +issue,143041,Use labels to create dedicated pools of non-scalable runners for trunk jobs,https://github.com/pytorch/pytorch/issues/143041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: infra",scan when time allows,2025-04-09T17:02:26Z +issue,132856,Enhance the compiler module to use cudagraph with PrivateUse1,https://github.com/pytorch/pytorch/issues/132856,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: privateuse1, oncall: pt2",scan when time allows,2025-04-09T16:57:19Z +issue,74236,"""histogram_cpu"" not implemented for 'Int'",https://github.com/pytorch/pytorch/issues/74236,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2025-04-09T16:47:30Z +issue,33811,"Models saved in C++ LibTorch with torch::save, cannot be loaded in python using torch.load",https://github.com/pytorch/pytorch/issues/33811,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: serialization",scan when time allows,2025-04-09T07:11:42Z +issue,150873,Add Created On | Last Updated On to the docs,https://github.com/pytorch/pytorch/issues/150873,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2025-04-08T21:24:37Z +issue,150871,register_replacement should also respect arg_kwarg_vals,https://github.com/pytorch/pytorch/issues/150871,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-04-08T19:19:54Z +issue,150663,Proposing torch.empty_cache (device_type) as generalization of torch.cuda.empty_cache() .,https://github.com/pytorch/pytorch/issues/150663,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: accelerator,scan when time allows,2025-04-08T17:10:48Z +issue,150700,CuDNN + H100 Gives Weird Gradients,https://github.com/pytorch/pytorch/issues/150700,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2025-04-08T16:44:04Z +issue,145978,What is the recommended way to use Distributed Checkpointing Save/Load with HSDP?,https://github.com/pytorch/pytorch/issues/145978,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: distributed checkpointing",scan when time allows,2025-04-08T15:58:03Z +issue,146307,Intermittent false positives in lintrunner's mypy,https://github.com/pytorch/pytorch/issues/146307,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: lint",scan when time allows,2025-04-08T09:50:50Z +issue,150718,Codegen or Lint for python-api.md,https://github.com/pytorch/pytorch/issues/150718,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2025-04-07T15:38:37Z +issue,36748,return_index option for torch.unique,https://github.com/pytorch/pytorch/issues/36748,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-04-07T12:54:34Z +issue,138696,[CI/CD] Deprecating PyTorch’s official Anaconda channel,https://github.com/pytorch/pytorch/issues/138696,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-04-05T18:44:01Z +issue,37406,3D grouped & depthwise convolution very slow on backward pass,https://github.com/pytorch/pytorch/issues/37406,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2025-04-05T11:48:50Z +issue,119076,torch.bmm fails when multiplying a batch of sparse matrices by a dense tensor,https://github.com/pytorch/pytorch/issues/119076,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-04-04T23:12:21Z +issue,119482,`pylint`: `torch.nn.functional.linear` is not callable,https://github.com/pytorch/pytorch/issues/119482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: typing",scan when time allows,2025-04-04T11:55:37Z +issue,149962,[ued][wan] Dynamic shape issue with tolist,https://github.com/pytorch/pytorch/issues/149962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-04-03T18:07:19Z +issue,150614,conv2d fp8 support,https://github.com/pytorch/pytorch/issues/150614,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: floatx (formerly float8)",scan when time allows,2025-04-03T15:30:22Z +issue,128149,torch.compile with Custom tensor subclass doesn't inline the tensor subclass methods,https://github.com/pytorch/pytorch/issues/128149,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamo,scan when time allows,2025-04-03T00:10:25Z +issue,141299,Pytorch notifications,https://github.com/pytorch/pytorch/issues/141299,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-04-02T23:42:57Z +issue,150548,Segmentation fault when using torch.tensor from a non python created thread,https://github.com/pytorch/pytorch/issues/150548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: pybind",scan when time allows,2025-04-02T22:46:33Z +issue,150531,Intermittent SSL certificate expiry warnings for `download.pytorch.org` (load balancer?),https://github.com/pytorch/pytorch/issues/150531,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-04-02T22:27:28Z +issue,150368,Support `copy` kwarg in `torch.reshape()` following Python array API standard,https://github.com/pytorch/pytorch/issues/150368,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2025-04-02T18:50:16Z +issue,150402,Feature Request: Add minLSTM and minGRU Modules,https://github.com/pytorch/pytorch/issues/150402,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend, module: rnn",scan when time allows,2025-04-02T18:48:44Z +issue,150284,torch.fill bug,https://github.com/pytorch/pytorch/issues/150284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-04-02T18:44:56Z +issue,150257,Type hint bug report relating to Sequential,https://github.com/pytorch/pytorch/issues/150257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-04-02T18:44:49Z +issue,150285,CuSparse doesn't work on sparse tensor,https://github.com/pytorch/pytorch/issues/150285,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-04-02T18:44:30Z +issue,150078,[RFC] Remove periodic/unstable jobs that has been continuously broken for more than 30 days,https://github.com/pytorch/pytorch/issues/150078,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-04-02T15:47:17Z +issue,146707,AutoAC can cause deadlocks with tensor parallelism and data-dependent flop formulas,https://github.com/pytorch/pytorch/issues/146707,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: pt2-dispatcher, oncall: pt2",scan when time allows,2025-04-01T23:23:02Z +issue,149879,Inductor Pattern Matcher's register_replacement function only works with functional `search_fn`s,https://github.com/pytorch/pytorch/issues/149879,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-04-01T19:22:24Z +issue,61474,Implement missing torch.nan* operators,https://github.com/pytorch/pytorch/issues/61474,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: numpy, module: reductions",scan when time allows,2025-04-01T17:46:04Z +issue,82430,Minor inconsistency in description of `attn_output_weights` in MultiheadAttention docs,https://github.com/pytorch/pytorch/issues/82430,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2025-04-01T12:26:37Z +issue,68352,`Delta` not defined in the docs of `HingeEmbeddingLoss`,https://github.com/pytorch/pytorch/issues/68352,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: loss, module: nn",scan when time allows,2025-04-01T03:02:55Z +issue,146898,[RFC] Test Cases Enabling for Accelerators,https://github.com/pytorch/pytorch/issues/146898,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2025-04-01T03:00:16Z +issue,146693,Partitioner moves useless and memory-heavy op from bwd to fwd,https://github.com/pytorch/pytorch/issues/146693,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, module: pt2-dispatcher, oncall: distributed, oncall: pt2",scan when time allows,2025-04-01T00:59:50Z +issue,149269,Add Optimizer: Sharpness Aware Minimization,https://github.com/pytorch/pytorch/issues/149269,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-03-31T20:07:53Z +issue,150242,Feature Request: Enhance Nested Tensor Operations for Direct RoPE Application,https://github.com/pytorch/pytorch/issues/150242,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-03-31T15:25:53Z +issue,71367,"F.cross_entropy do not have a deterministic implementation, adding deterministic support for this operation.",https://github.com/pytorch/pytorch/issues/71367,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2025-03-30T09:01:31Z +issue,149957,[ued] Prohibitive warm start latency,https://github.com/pytorch/pytorch/issues/149957,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: inductor, oncall: pt2",scan when time allows,2025-03-29T15:41:23Z +issue,149908,[ued][ChatTTS][guards] Too many recompilations,https://github.com/pytorch/pytorch/issues/149908,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-03-29T15:38:33Z +issue,149906,[ued][chatTTS][dynamo] graph break on should_compile_partial_graph=False,https://github.com/pytorch/pytorch/issues/149906,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-03-29T15:38:32Z +issue,149970,[ued][deepssek-vl2] Graph break on copy.deepcopy,https://github.com/pytorch/pytorch/issues/149970,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-03-29T15:35:24Z +issue,149956,[ued][flux][dynamo] Wrong error message with dynamo.disable,https://github.com/pytorch/pytorch/issues/149956,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-03-29T15:35:22Z +issue,149944,include cudagraph skip reasons in tlparse output,https://github.com/pytorch/pytorch/issues/149944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: dynamo, oncall: pt2",scan when time allows,2025-03-29T15:35:22Z +issue,147062,[Feature Request] Release original parameters by layer when turning on `freezing_discard_parameters`,https://github.com/pytorch/pytorch/issues/147062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-29T15:23:32Z +issue,146930,pytest test/dynamo fails,https://github.com/pytorch/pytorch/issues/146930,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-03-29T15:22:14Z +issue,79272,"PyTorch leaks a macro definition called ""CHECK"" in the C++ version",https://github.com/pytorch/pytorch/issues/79272,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-03-29T15:16:24Z +issue,132695,Backward on nested jagged tensor vector multiplication throws error on 0 dim (but not last dim),https://github.com/pytorch/pytorch/issues/132695,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-03-28T18:50:14Z +issue,149609,Batching Implementation for aten_nested_tensor_from_mask_left_aligned,https://github.com/pytorch/pytorch/issues/149609,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: nestedtensor, module: vmap",scan when time allows,2025-03-28T18:43:13Z +issue,138901,Runtime error: retains_grad_hooks not implemented for compiled autograd,https://github.com/pytorch/pytorch/issues/138901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, oncall: pt2",scan when time allows,2025-03-28T08:41:57Z +issue,149743,dynamic annotations that still allow duck sizing,https://github.com/pytorch/pytorch/issues/149743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-03-28T06:49:02Z +issue,6564,[Feature Request] Optimization with constraint (L-BFGS-B),https://github.com/pytorch/pytorch/issues/6564,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-03-27T16:42:06Z +issue,146201,L-BFGS-B support,https://github.com/pytorch/pytorch/issues/146201,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-03-27T16:40:07Z +issue,104152,Error 101: invalid device ordinal (Triggered internally at ../c10/cuda/CUDAFunctions.cpp:109.),https://github.com/pytorch/pytorch/issues/104152,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: third_party",scan when time allows,2025-03-27T16:25:21Z +issue,106529,Pytorch + ROCm+ Windows,https://github.com/pytorch/pytorch/issues/106529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-03-27T15:20:28Z +issue,75773,Handle noncontiguous inputs in distributed backend layer,https://github.com/pytorch/pytorch/issues/75773,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2025-03-26T22:56:16Z +issue,149169,[RelEng] wheel testing for new arch versions,https://github.com/pytorch/pytorch/issues/149169,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",scan when time allows,2025-03-26T22:26:19Z +issue,148984,[Sync file] the new file is not sync properly between pytorch/pytorch and pytorch/benchmark,https://github.com/pytorch/pytorch/issues/148984,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2025-03-26T22:26:11Z +issue,68978,SequentialLR cannot be used with ReduceLROnPlateau due to .step() not allowing for optional arguments,https://github.com/pytorch/pytorch/issues/68978,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2025-03-26T19:38:07Z +issue,138226,No guards for frozen dataclasses and readonly configs,https://github.com/pytorch/pytorch/issues/138226,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: guards, oncall: pt2",scan when time allows,2025-03-26T17:45:15Z +issue,146647,placeholder for shell dtype improvements in pytorch/pytorch,https://github.com/pytorch/pytorch/issues/146647,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2025-03-26T17:08:14Z +issue,149734,"[RFC] Multi-backend, multi-device test class instantiation for Inductor",https://github.com/pytorch/pytorch/issues/149734,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: testing, oncall: pt2",scan when time allows,2025-03-26T15:35:35Z +issue,139290,Using xpu module in the cuda version Pytorch,https://github.com/pytorch/pytorch/issues/139290,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: xpu",scan when time allows,2025-03-26T07:52:03Z +issue,136381,[Meta] torch.export readiness for Recommendation models,https://github.com/pytorch/pytorch/issues/136381,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-03-26T00:43:22Z +issue,133458,[CPU] jx_nest_base AMP both inductor and eager performance regression in 2024-08-10 nightly release,https://github.com/pytorch/pytorch/issues/133458,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",scan when time allows,2025-03-25T23:22:22Z +issue,148682,can we make inductor create faster fusions for tiled reductions across dim0 and dim1?,https://github.com/pytorch/pytorch/issues/148682,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-25T21:27:39Z +issue,137217,Automagically lift assertions / shape checks into torch._check,https://github.com/pytorch/pytorch/issues/137217,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-03-25T18:36:00Z +issue,136119,[torch.export] Dim discrete/finite set,https://github.com/pytorch/pytorch/issues/136119,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: export, oncall: pt2",scan when time allows,2025-03-25T17:48:58Z +issue,139268,Torch Inductor should have a way for new backend to provide build options,https://github.com/pytorch/pytorch/issues/139268,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-25T17:24:46Z +issue,139401,[export] Failure to export Mixtral-8x7B-Instruct-v0.1,https://github.com/pytorch/pytorch/issues/139401,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-03-25T17:15:32Z +issue,94083,Tracker for `scatter_reduce` additional reduction options requests,https://github.com/pytorch/pytorch/issues/94083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2025-03-25T12:34:02Z +issue,98861,Sparse Tensor not working for `torch.cat`,https://github.com/pytorch/pytorch/issues/98861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-03-25T08:58:44Z +issue,149845,CUDA error: no kernel image is available for execution on the device RTX5090D,https://github.com/pytorch/pytorch/issues/149845,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-03-25T02:58:38Z +issue,148045,Run Performance Regression Tests on new Version of Triton,https://github.com/pytorch/pytorch/issues/148045,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2025-03-24T23:55:15Z +issue,149815,comparison operators only accept scalars as the 2nd argument but not as a 1st argument,https://github.com/pytorch/pytorch/issues/149815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-03-24T23:30:06Z +issue,23859,CUDA: THTensor code complains about devices not matching when creating tensor from blob,https://github.com/pytorch/pytorch/issues/23859,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-03-24T20:31:01Z +issue,71978,from_blob / make_tensor support MemoryFormat,https://github.com/pytorch/pytorch/issues/71978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2025-03-24T20:29:28Z +issue,118421,Increase default bucket_cap_mb value to from 25MB to a more optimal value,https://github.com/pytorch/pytorch/issues/118421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-03-24T19:37:20Z +issue,149804,torch.linalg.norm RuntimeError with torch.func.grad( vmap( hessian(.) ) ),https://github.com/pytorch/pytorch/issues/149804,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: linear algebra, module: vmap",scan when time allows,2025-03-24T18:52:44Z +issue,149803,checking out NCCL when it is not used,https://github.com/pytorch/pytorch/issues/149803,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-03-24T18:52:20Z +issue,149798,Build Extension Failed with setuptools==77.0.3,https://github.com/pytorch/pytorch/issues/149798,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2025-03-24T18:48:11Z +issue,149841,`weight` parameter on functional not in nn.Module,https://github.com/pytorch/pytorch/issues/149841,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-03-24T18:43:46Z +issue,149788,`torch.vmap` does not work with tensor subclasses as expected,https://github.com/pytorch/pytorch/issues/149788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-03-24T18:34:12Z +issue,149802,(With PR) `torch.cuda.MemPool()` internal assertion failure when changing devices,https://github.com/pytorch/pytorch/issues/149802,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-03-24T18:28:30Z +issue,149796,Cannot compile SGlang with Torch 2.7 or Torch 2.8 and CUDA 12.8 (sm_120).,https://github.com/pytorch/pytorch/issues/149796,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-03-24T14:27:59Z +issue,149603,"[inductor] API for user-controlled fusion ""escape-hatch"" mechanism in inductor",https://github.com/pytorch/pytorch/issues/149603,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-24T10:32:26Z +issue,135704,[feature request] Varlen indexing function for lookup and concat of varlen BPE tokens from a tensor vocab (i.e. `detokenize(...)` and arrays of strings). NestedTensor could become a representation for arrays-of-strings,https://github.com/pytorch/pytorch/issues/135704,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-03-23T12:29:20Z +issue,148858,RuntimeError: OffsetBasedRNGTracker instantiation requires the presence of CUDA/CUDA-like device,https://github.com/pytorch/pytorch/issues/148858,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-03-21T23:52:27Z +issue,149724,"`torch.linalg.ldl_factor_ex`, `torch.linalg.ldl_factor`, and `torch.linalg.lstsq` Raise INTERNAL ASSERT FAILED",https://github.com/pytorch/pytorch/issues/149724,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: linear algebra",scan when time allows,2025-03-21T15:32:19Z +issue,149718,Constraints for distributions with mixed support,https://github.com/pytorch/pytorch/issues/149718,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-03-21T15:07:28Z +issue,149720,Implement `permute` for masked tensor,https://github.com/pytorch/pytorch/issues/149720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-03-21T15:06:52Z +issue,21462,"Slow convolution with large kernels, should be using FFT",https://github.com/pytorch/pytorch/issues/21462,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: fft, module: performance",scan when time allows,2025-03-21T00:25:30Z +issue,147475,[Export AOTI] dynamic_shapes export and compile degraded output,https://github.com/pytorch/pytorch/issues/147475,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotinductor, oncall: export, oncall: pt2",scan when time allows,2025-03-20T22:03:17Z +issue,149452,"Support save_cubin (and therefore, support cpp_wrapper use cases)",https://github.com/pytorch/pytorch/issues/149452,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-03-20T19:38:52Z +issue,149455,Support num_ctas > 1?,https://github.com/pytorch/pytorch/issues/149455,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-03-20T19:29:05Z +issue,149454,Support user defined triton kernels,https://github.com/pytorch/pytorch/issues/149454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-03-20T19:28:56Z +issue,149620,`vmap` not working on `torch.range`,https://github.com/pytorch/pytorch/issues/149620,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2025-03-20T19:26:55Z +issue,149545,"`torch.load(..., map_location=""meta"")` hangs indefinitely",https://github.com/pytorch/pytorch/issues/149545,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-03-20T19:18:59Z +issue,148144,Torch export does not preserve original edges between nodes,https://github.com/pytorch/pytorch/issues/148144,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export, oncall: pt2",scan when time allows,2025-03-20T16:43:10Z +issue,115305,torchrun to support NUMA affiinity,https://github.com/pytorch/pytorch/issues/115305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributed_tool, oncall: distributed",scan when time allows,2025-03-20T15:39:32Z +issue,149184,"[macOS] instantiating optimizer after torch.set_default_device(""mps"") throws ""RuntimeError: Placeholder storage has not been allocated on MPS device!""",https://github.com/pytorch/pytorch/issues/149184,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-03-19T20:30:19Z +issue,72759,Add softplus inverse,https://github.com/pytorch/pytorch/issues/72759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2025-03-19T09:48:35Z +issue,31772,Support Python builtins on iterators in JIT,https://github.com/pytorch/pytorch/issues/31772,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2025-03-18T19:25:12Z +issue,128150,More economical use of FX nodes when generating symbolic shapes related manipulations,https://github.com/pytorch/pytorch/issues/128150,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-03-18T18:07:33Z +issue,36437,Issue when linking C++ code with libtorch_cpu: cuda not detected,https://github.com/pytorch/pytorch/issues/36437,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2025-03-18T16:03:19Z +issue,148828,MiniCPM-o compilation issue,https://github.com/pytorch/pytorch/issues/148828,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-03-18T13:54:12Z +issue,124711,Output Discrepancy between PyTorch Model and Converted ONNX Model,https://github.com/pytorch/pytorch/issues/124711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-03-18T08:18:19Z +issue,149313,"Training error: due to torch.load(weights, map_location='cpu') - Weights only load failed.",https://github.com/pytorch/pytorch/issues/149313,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-03-18T06:57:29Z +issue,149319,Need some sort of pattern matcher recipes somewhere,https://github.com/pytorch/pytorch/issues/149319,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-17T23:59:33Z +issue,149094,How to skip backward specific steps in torch.compile,https://github.com/pytorch/pytorch/issues/149094,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-03-17T23:55:31Z +issue,148920,Rewrite cpp extension tests to not be crazy,https://github.com/pytorch/pytorch/issues/148920,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: tests",scan when time allows,2025-03-17T20:45:20Z +issue,149318,BypassFxGraphCache should have better error message for HOPs (also we should support more),https://github.com/pytorch/pytorch/issues/149318,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-03-17T20:33:05Z +issue,149324,Unguarded Usage of Facebook Internal Code?,https://github.com/pytorch/pytorch/issues/149324,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: third_party, oncall: pt2",scan when time allows,2025-03-17T20:29:04Z +issue,122485,DTensor doesn't support expand_as dispath mode,https://github.com/pytorch/pytorch/issues/122485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2025-03-17T18:49:43Z +issue,149261,MPS Error: NDArray > 2^32 bytes in scaled_dot_product_attention,https://github.com/pytorch/pytorch/issues/149261,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: mps, module: sdpa",scan when time allows,2025-03-17T16:05:32Z +issue,149284,No examples in documentation for masked_fill and masked_fill_,https://github.com/pytorch/pytorch/issues/149284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-03-17T15:33:18Z +issue,149209,Wrong argument in `CompileCounterWithBackend` when running `torch.utils.benchmark.utils.compile.bench_all`,https://github.com/pytorch/pytorch/issues/149209,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: benchmark,scan when time allows,2025-03-17T15:30:34Z +issue,149250,Implement batching for torch.isin operator,https://github.com/pytorch/pytorch/issues/149250,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2025-03-17T15:26:55Z +issue,149227,Unexpected behaviour in using torch.nn.utils.rnn.pack_padded_sequence API,https://github.com/pytorch/pytorch/issues/149227,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2025-03-17T15:25:26Z +issue,149224,Unexpected results in using torch.save API,https://github.com/pytorch/pytorch/issues/149224,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-03-17T15:20:20Z +issue,78065,Gamma and Related Functions,https://github.com/pytorch/pytorch/issues/78065,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2025-03-15T20:01:14Z +issue,149013,Unexpected out-of-boundary behavior in `grid_sample`,https://github.com/pytorch/pytorch/issues/149013,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nn, module: numpy",scan when time allows,2025-03-15T09:53:35Z +issue,141014,ROCm: libamdhip64.so is linked twice when building a torch C++ program,https://github.com/pytorch/pytorch/issues/141014,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2025-03-14T19:36:47Z +issue,149040,"Only 2D, 3D, 4D, 5D padding with non-constant padding are supported for now",https://github.com/pytorch/pytorch/issues/149040,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-03-14T19:01:28Z +issue,148976,`dist.barrier()` fails with TORCH_DISTRIBUTED_DEBUG=DETAIL and after dist.send/dist.recv calls,https://github.com/pytorch/pytorch/issues/148976,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-03-14T01:25:03Z +issue,92909,[RFC] XLA Lazy Backend Support In DistributedTensor API,https://github.com/pytorch/pytorch/issues/92909,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xla,scan when time allows,2025-03-13T18:22:18Z +issue,99304,Need TransformerEncoder to output attention map,https://github.com/pytorch/pytorch/issues/99304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-03-13T16:29:12Z +issue,149120,Add `keepdim` parameter for `torch.nn.functional.cosine_similarity`,https://github.com/pytorch/pytorch/issues/149120,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-03-13T15:24:10Z +issue,149090,Better warning once for `cuDNN/MIOpen` not enabled,https://github.com/pytorch/pytorch/issues/149090,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: tests",scan when time allows,2025-03-13T15:20:21Z +issue,148954,Aborted (core dumped) double free or corruption (out),https://github.com/pytorch/pytorch/issues/148954,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-03-13T15:13:06Z +issue,148885,test_memory_profiler_viz failed on cudamallocasync,https://github.com/pytorch/pytorch/issues/148885,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: testing",scan when time allows,2025-03-13T15:05:07Z +issue,125358,"[DeviceMesh] Add support for `group: Tuple[ProcessGroup, ...]` in `from_group()`",https://github.com/pytorch/pytorch/issues/125358,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devicemesh,scan when time allows,2025-03-13T08:46:41Z +issue,149096,How to determine which part of torch.compile undergoes recompiling after caching,https://github.com/pytorch/pytorch/issues/149096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-03-13T06:40:24Z +issue,148960,Does torch compile affect results ?,https://github.com/pytorch/pytorch/issues/148960,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-13T05:44:37Z +issue,37664,Remove everything from a GPU (including drivers),https://github.com/pytorch/pytorch/issues/37664,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-03-12T13:10:14Z +issue,147661,Add shape fn for einsum,https://github.com/pytorch/pytorch/issues/147661,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lazy, module: linear algebra, module: xla",scan when time allows,2025-03-11T23:16:50Z +issue,107130,[docs] Document dtype conversions dtype.to_complex() dtype.to_real(),https://github.com/pytorch/pytorch/issues/107130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2025-03-11T19:50:37Z +issue,130396,"Add some way to automatically opt-out of CI features, like TD.",https://github.com/pytorch/pytorch/issues/130396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-03-11T19:09:02Z +issue,67978,c10::CUDAError,https://github.com/pytorch/pytorch/issues/67978,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nccl, oncall: distributed",scan when time allows,2025-03-11T17:16:01Z +issue,68923,[cpp extension] making it possible to parallelize the building process,https://github.com/pytorch/pytorch/issues/68923,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci, module: cpp-extensions",scan when time allows,2025-03-11T07:13:52Z +issue,147628,"Optimize memory usage of torch.sort significantly, with dynamic dtype indices",https://github.com/pytorch/pytorch/issues/147628,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-03-11T02:49:29Z +issue,148927,default cudagraphable policy for custom op,https://github.com/pytorch/pytorch/issues/148927,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, oncall: pt2",scan when time allows,2025-03-10T23:57:20Z +issue,145158,Pytorch matmul for nested 4D tensors in jagged layout doesn't work,https://github.com/pytorch/pytorch/issues/145158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-03-10T23:02:22Z +issue,148775,Calling `torch.linalg.lstsq` with a wrongly-shaped `out=output` argument does not properly reshape `output`,https://github.com/pytorch/pytorch/issues/148775,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-03-10T20:02:02Z +issue,81104,Resize/reshape of sparse compressed tensors - design,https://github.com/pytorch/pytorch/issues/81104,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-03-10T19:17:19Z +issue,148247,`torch.multinomial` outputs inconsistency on ARM and x86,https://github.com/pytorch/pytorch/issues/148247,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: random",scan when time allows,2025-03-10T02:38:31Z +issue,148546,Pytorch nightly broken Flash Attention 3 compile with sycl commit - TypeError: _write_ninja_file() got an unexpected keyword argument 'sycl_cflags',https://github.com/pytorch/pytorch/issues/148546,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-03-09T22:37:30Z +issue,137268,DDP deadlock ProcessGroupNCCL's watchdog got stuck,https://github.com/pytorch/pytorch/issues/137268,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: ddp, oncall: distributed",scan when time allows,2025-03-09T01:20:02Z +issue,72581,Zero-copy on shared memory of NVIDIA Jetson devices,https://github.com/pytorch/pytorch/issues/72581,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: jetson,scan when time allows,2025-03-07T22:03:39Z +issue,97539,InfoNCE loss for contrastive learning,https://github.com/pytorch/pytorch/issues/97539,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2025-03-07T14:38:15Z +issue,148568,Suggested fixes sometimes not enough in export,https://github.com/pytorch/pytorch/issues/148568,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-03-07T01:43:49Z +issue,145511,Activation Checkpointing composability with split backward computation,https://github.com/pytorch/pytorch/issues/145511,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: pipelining, oncall: distributed",scan when time allows,2025-03-06T22:17:21Z +issue,148518,Preview (Nightly) version cuda12.8 cannot find torchaudio file,https://github.com/pytorch/pytorch/issues/148518,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2025-03-06T01:51:38Z +issue,148548,Issue with Sparse Tensor Matrix Multiplication and Broadcasting,https://github.com/pytorch/pytorch/issues/148548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-03-06T01:50:06Z +issue,148379,Improve nested jagged tensor select performance on batch dim,https://github.com/pytorch/pytorch/issues/148379,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: performance",scan when time allows,2025-03-06T00:47:29Z +issue,148044,Sweep through Potentially BC Breaking Commits in Triton,https://github.com/pytorch/pytorch/issues/148044,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-03-05T23:40:01Z +issue,148078,Video-Llama (version 1) runs much slower using Float16 than Float32 on Kunpeng CPU,https://github.com/pytorch/pytorch/issues/148078,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: cpu",scan when time allows,2025-03-05T08:01:06Z +issue,148193,[Request Help] “torch._dynamo.exc.UserError: Dynamic control flow is not supported at the moment.” “torch._dynamo.exc.UncapturedHigherOrderOpError: Cond doesn't work unless it is captured completely with torch.compile.”,https://github.com/pytorch/pytorch/issues/148193,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2025-03-05T03:17:22Z +issue,148073,torch.compile with the inductor backend slows down (exponentially?) for certain graphs,https://github.com/pytorch/pytorch/issues/148073,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-04T19:36:04Z +issue,126795,Using DTensor with device meshes that use different devices for input and output,https://github.com/pytorch/pytorch/issues/126795,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-03-04T18:49:42Z +issue,148116,[inductor][triton] Explicit kernel-arg mismatch checks,https://github.com/pytorch/pytorch/issues/148116,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-04T17:33:35Z +issue,148146,Unify OpOverload._get_dispatch and HigherOrderOperator.dispatch,https://github.com/pytorch/pytorch/issues/148146,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: aotdispatch, oncall: pt2",scan when time allows,2025-03-04T16:03:01Z +issue,148196,"[inductor][triton] Decide how to deprecate ""old triton versions""",https://github.com/pytorch/pytorch/issues/148196,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-04T15:39:46Z +issue,148103,Initial investigation for removing MOD_SKIPLIST,https://github.com/pytorch/pytorch/issues/148103,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-03-04T15:05:25Z +issue,148052,[inductor][user triton] support on-device TMA / tensor descriptor API,https://github.com/pytorch/pytorch/issues/148052,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: user triton, oncall: pt2",scan when time allows,2025-03-04T02:41:35Z +issue,147999,Adjust test_mm_triton_kernel_benchmark for unpadded tensors,https://github.com/pytorch/pytorch/issues/147999,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-03-04T02:39:26Z +issue,138813,Strange recompilations on torch 2.5 + FSDP + UNet,https://github.com/pytorch/pytorch/issues/138813,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: dynamo, module: fsdp, oncall: distributed",scan when time allows,2025-03-04T02:09:59Z +issue,136605,automatic_dynamic_shapes for mark_unbacked,https://github.com/pytorch/pytorch/issues/136605,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes,scan when time allows,2025-03-04T02:08:56Z +issue,143774,CUDA error when compiling loss function,https://github.com/pytorch/pytorch/issues/143774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, oncall: pt2",scan when time allows,2025-03-03T19:29:30Z +issue,144470,Partitioner's auto-AC misbehaves with mixed dtypes,https://github.com/pytorch/pytorch/issues/144470,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd, oncall: pt2",scan when time allows,2025-03-03T19:29:22Z +issue,119070,[DCP] Review and update examples and docstring to reflect DCP save/load API updates,https://github.com/pytorch/pytorch/issues/119070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed checkpointing,scan when time allows,2025-03-03T19:06:02Z +issue,126047,`torch_save_to_dcp`'s produced dcp checkpoint doesn't fit the load interface for FSDP,https://github.com/pytorch/pytorch/issues/126047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed checkpointing,scan when time allows,2025-03-03T19:05:47Z +issue,137327,[torch2.4][Distributed Checkpoint] new flatten logic is error-prone,https://github.com/pytorch/pytorch/issues/137327,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed checkpointing,scan when time allows,2025-03-03T19:05:06Z +issue,148098,onnx dynamo export does not support aten bucketize,https://github.com/pytorch/pytorch/issues/148098,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-03-03T17:39:09Z +issue,148113,"[inductor][triton] introduce better ""APIs"" in triton that can clean up our triton/inductor integration",https://github.com/pytorch/pytorch/issues/148113,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-03-03T17:06:32Z +issue,147874,Does CUDACachingAllocator.cpp still require deferred event creation?,https://github.com/pytorch/pytorch/issues/147874,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-03-03T17:02:58Z +issue,147884,`FxGraphDrawer` fails on `einsum` nodes,https://github.com/pytorch/pytorch/issues/147884,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-03-03T17:00:15Z +issue,129942,Torch Compile mode giving Graph Break for Higher Order Op for ProcessGroup param,https://github.com/pytorch/pytorch/issues/129942,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2",scan when time allows,2025-03-03T14:12:33Z +issue,148219,MPS vs Metal vs CPU performance comparison,https://github.com/pytorch/pytorch/issues/148219,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2025-03-01T00:49:52Z +issue,125958,Support AOT Autograd level Caching,https://github.com/pytorch/pytorch/issues/125958,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, oncall: pt2",scan when time allows,2025-02-28T18:32:19Z +issue,102374,Hooks not working in version 2.0.1+cu118,https://github.com/pytorch/pytorch/issues/102374,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-02-28T08:28:51Z +issue,134191,"AdamW is CPU-bottlenecked with FSDP2, with slow foreach kernels",https://github.com/pytorch/pytorch/issues/134191,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: optimizer",scan when time allows,2025-02-28T02:25:02Z +issue,143670,Potential rooms for fewer recompilations by introducing higher-level guards,https://github.com/pytorch/pytorch/issues/143670,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-27T21:59:13Z +issue,127225,make_graphed_callables don't work with FSDP at all even on a simple network,https://github.com/pytorch/pytorch/issues/127225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2025-02-27T18:26:02Z +issue,140437,Let torch.compiler.allow_in_graph work in more situations,https://github.com/pytorch/pytorch/issues/140437,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-27T17:31:20Z +issue,147867,Track follow ups to #147354,https://github.com/pytorch/pytorch/issues/147867,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2025-02-27T16:03:32Z +issue,64359,[feature request] `numpy.append` / `numpy.insert` / `numpy.delete` equivalents and implement dynamic arrays (reallocate storage with a surplus),https://github.com/pytorch/pytorch/issues/64359,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: viewing and reshaping",scan when time allows,2025-02-27T10:50:56Z +issue,8937,[caffe2] Is there any method to implement learning rate scheduler?,https://github.com/pytorch/pytorch/issues/8937,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-02-27T07:48:10Z +issue,148029,abs and arange Meta impls are non-standard and thus not working with other backends,https://github.com/pytorch/pytorch/issues/148029,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2025-02-27T00:28:54Z +issue,138786,Support nn.Module arguments for the function to torch.compiler.allow_in_graph,https://github.com/pytorch/pytorch/issues/138786,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-27T00:08:23Z +issue,141589,Minimizing recompiles when compiling multiple inlined `nn.Module`s,https://github.com/pytorch/pytorch/issues/141589,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: guards, oncall: pt2",scan when time allows,2025-02-26T22:41:43Z +issue,142395,[Dynamo] Do an audit on skipfiles and mark more files as inline,https://github.com/pytorch/pytorch/issues/142395,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-26T22:40:19Z +issue,133230,[dynamo] torch.compile does not support compiling a module that contains autograd within it.,https://github.com/pytorch/pytorch/issues/133230,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, module: dynamo, oncall: pt2",scan when time allows,2025-02-26T22:32:30Z +issue,133117,"[dynamo] Improve support for subclasses of popular builtin types like dict, list, set",https://github.com/pytorch/pytorch/issues/133117,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-26T22:30:26Z +issue,147597,[Dynamo] WeakRefVariable doesn't use the most updated python referent when call_function is executed,https://github.com/pytorch/pytorch/issues/147597,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-26T21:24:42Z +issue,129228,CVE-2024-5480 reported by security analyzers,https://github.com/pytorch/pytorch/issues/129228,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-02-26T11:12:57Z +issue,147859,Exporting a PyTorch Model with Dynamic ModuleList Indexing to ONNX,https://github.com/pytorch/pytorch/issues/147859,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-02-26T02:41:28Z +issue,32651,[Tensorboard] Problem with subfolders from SummaryWriter,https://github.com/pytorch/pytorch/issues/32651,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2025-02-25T17:37:13Z +issue,147682,Error in parsing arch_string in _extract_arch_version,https://github.com/pytorch/pytorch/issues/147682,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-02-25T15:29:10Z +issue,147692,Please generate a new version of pytorch into blackwell rtx5080 with cuda 12.8. Thank you very much.,https://github.com/pytorch/pytorch/issues/147692,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2025-02-25T15:24:41Z +issue,128015,Best approach for dynamically changing batch sizes on iterable datasets,https://github.com/pytorch/pytorch/issues/128015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2025-02-25T02:50:38Z +issue,129937,[tracker] Dynamo guard validation exposed issues,https://github.com/pytorch/pytorch/issues/129937,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-25T00:23:56Z +issue,130119,Compilation time for packing function is quite long,https://github.com/pytorch/pytorch/issues/130119,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, oncall: pt2",scan when time allows,2025-02-25T00:11:58Z +issue,130534,Clarify invariants established by torch.fx.passes.runtime_asserts or factor so we can have a pre-pass that reestablishes them,https://github.com/pytorch/pytorch/issues/130534,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-02-24T22:45:58Z +issue,13811,Feature request: von Mises-Fisher distribution,https://github.com/pytorch/pytorch/issues/13811,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-02-24T22:44:09Z +issue,130769,[Inductor] Add an API to register external callable candidates for inductor's Matmul/Conv tuning choices,https://github.com/pytorch/pytorch/issues/130769,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-02-24T22:32:09Z +issue,128095,[inductor] Graph breaks in CohereForAI/aya-23-8b,https://github.com/pytorch/pytorch/issues/128095,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: graph breaks, oncall: pt2",scan when time allows,2025-02-24T21:51:31Z +issue,128153,"torch.compile Jamba: long compilation time with backend=""eager""",https://github.com/pytorch/pytorch/issues/128153,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamo, module: guards, oncall: pt2",scan when time allows,2025-02-24T18:30:17Z +issue,129482,Dynamo graph breaks on boxed ProcessGroup,https://github.com/pytorch/pytorch/issues/129482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2",scan when time allows,2025-02-24T17:44:41Z +issue,147595,Non-Determinism in Faster R-CNN Despite Setting All Deterministic Flags,https://github.com/pytorch/pytorch/issues/147595,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2025-02-24T16:05:22Z +issue,147389,Fix representation of `Lazy*` modules after loading parameters,https://github.com/pytorch/pytorch/issues/147389,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-02-24T16:03:32Z +issue,147585,[SDPA backward error]Error detected in ScaledDotProductEfficientAttentionBackward0 when input seqlen is very long and with attn_mask,https://github.com/pytorch/pytorch/issues/147585,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-02-24T16:01:29Z +issue,147495,Native Node Sleep/Wake Functionality for Neural Network Layers,https://github.com/pytorch/pytorch/issues/147495,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-02-24T15:53:59Z +issue,147506,Unintuitive behavior and errors in F.pad,https://github.com/pytorch/pytorch/issues/147506,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",scan when time allows,2025-02-24T15:52:09Z +issue,61292,torch.linspace tensor support,https://github.com/pytorch/pytorch/issues/61292,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-02-24T10:59:10Z +issue,130067,Meta device not supported with autocast,https://github.com/pytorch/pytorch/issues/130067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2025-02-24T10:58:22Z +issue,147288,Calling optimizer.load_state_dict causes Tensors lost,https://github.com/pytorch/pytorch/issues/147288,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-02-23T18:15:57Z +issue,146157,AdamW refactoring broke checkpoint reloading with DCP,https://github.com/pytorch/pytorch/issues/146157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: distributed checkpointing",scan when time allows,2025-02-22T05:02:04Z +issue,147284,torch.softmax gives wierd result under specific condition on cuda double tensors with multiple rows and 781 columns.,https://github.com/pytorch/pytorch/issues/147284,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2025-02-21T23:02:47Z +issue,49844,BCEWithLogitsLoss gives out nan with -inf logits,https://github.com/pytorch/pytorch/issues/49844,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nans and infs",scan when time allows,2025-02-21T20:32:28Z +issue,147287,[inductor][fuzzer] legacy `torch.native_batch_norm` should be removed,https://github.com/pytorch/pytorch/issues/147287,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, oncall: pt2",scan when time allows,2025-02-21T13:43:44Z +issue,147581,cpp libtorch transformerimpl lack some parameter between with python pytorch,https://github.com/pytorch/pytorch/issues/147581,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2025-02-21T06:58:05Z +issue,59692,"cuda streams run sequentially, expected to run parallel",https://github.com/pytorch/pytorch/issues/59692,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2025-02-21T05:25:52Z +issue,147510,`clamp_` and `clamp` behave differently on MPS device.,https://github.com/pytorch/pytorch/issues/147510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-02-21T03:33:22Z +issue,38412,No MKL Compatible Conda installation for PyTorch 1.5,https://github.com/pytorch/pytorch/issues/38412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: mkl",scan when time allows,2025-02-21T03:21:40Z +issue,141686,[PT2] Fine-grained custom triton kernels support,https://github.com/pytorch/pytorch/issues/141686,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: user triton, oncall: pt2",scan when time allows,2025-02-20T23:11:06Z +issue,137514,Support multiple ragged dims for NJT,https://github.com/pytorch/pytorch/issues/137514,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-02-20T21:15:26Z +issue,37443,torch.clamp_max clamp_min shouldn't be there,https://github.com/pytorch/pytorch/issues/37443,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-02-20T17:01:42Z +issue,133586,No way for low-overhead total norm in native PyTorch with large number of tensors,https://github.com/pytorch/pytorch/issues/133586,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: nn, oncall: pt2",scan when time allows,2025-02-20T02:55:32Z +issue,131667,[RFE] Improve the error reporting if MIG is incorrectly used with NVML,https://github.com/pytorch/pytorch/issues/131667,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-02-19T20:40:33Z +issue,52134,Build using Py_LIMITED_API and then build wheels with the stable ABI abi3 tag,https://github.com/pytorch/pytorch/issues/52134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: performance, oncall: releng",scan when time allows,2025-02-19T19:16:05Z +issue,146910,"INTERNAL ASSERT FAILED at ""/pytorch/aten/src/ATen/native/sparse/SparseCsrTensorMath.cpp"":817, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/146910,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-02-19T13:56:12Z +issue,143112,FFT half precision only let CUDA pass the check,https://github.com/pytorch/pytorch/issues/143112,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: privateuse1",scan when time allows,2025-02-19T09:50:58Z +issue,145214,Nested tensor support for pointwise matrix multiplication of nested tensor and normal tensor,https://github.com/pytorch/pytorch/issues/145214,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2025-02-19T09:30:15Z +issue,121668,Warning static library kineto_LIBRARY-NOTFOUND not found when trying build tutorial torch.compiler_aot_inductor.html,https://github.com/pytorch/pytorch/issues/121668,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp, oncall: releng",scan when time allows,2025-02-19T04:50:58Z +issue,121485,libtorch memory leakage when inference by Multi-thread,https://github.com/pytorch/pytorch/issues/121485,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: memory usage",scan when time allows,2025-02-19T04:39:37Z +issue,146629,[inductor][idea] Defer realize/inline decisions,https://github.com/pytorch/pytorch/issues/146629,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-02-18T21:19:48Z +issue,147239,`DeviceCopy in input program` source hint,https://github.com/pytorch/pytorch/issues/147239,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-02-18T18:39:46Z +issue,147264,`Dim.DYNAMIC` inferred to be constant,https://github.com/pytorch/pytorch/issues/147264,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-02-18T17:48:48Z +issue,147387,Track test count regressions,https://github.com/pytorch/pytorch/issues/147387,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-02-18T17:46:13Z +issue,143969,[distributed] parallelize_module error with `SequenceParallel`,https://github.com/pytorch/pytorch/issues/143969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-02-17T19:26:18Z +issue,81084,[ONNX] Exporting the operator `::svd` to ONNX opset version 13 is not supported.,https://github.com/pytorch/pytorch/issues/81084,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2025-02-16T13:14:45Z +issue,146708,Periodic Activations Module,https://github.com/pytorch/pytorch/issues/146708,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-02-14T20:06:26Z +issue,101335,fsdp training with the seq2seqTranier module gets stuck during evaluation.,https://github.com/pytorch/pytorch/issues/101335,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2025-02-14T18:42:23Z +issue,107909,Provide a `reset_parameters()` method for MultiheadAttention to support FSDP meta device initializtion,https://github.com/pytorch/pytorch/issues/107909,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, module: nn",scan when time allows,2025-02-14T18:40:04Z +issue,109002,[FSDP] How can I wrap a model that has both nn.Parameter and nn.Module,https://github.com/pytorch/pytorch/issues/109002,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2025-02-14T18:39:38Z +issue,63911,Add efficient symmetric tensor representations,https://github.com/pytorch/pytorch/issues/63911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: linear algebra",scan when time allows,2025-02-14T17:10:50Z +issue,139210,[Runtime Error] Build PyTorch with cuda12.2 on Jetson AGX Orin with jetpack 5.1.4,https://github.com/pytorch/pytorch/issues/139210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: cuda, module: jetson",scan when time allows,2025-02-14T04:28:52Z +issue,147118,padding fails on view from large tensor,https://github.com/pytorch/pytorch/issues/147118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: edge cases, module: padding",scan when time allows,2025-02-13T18:16:46Z +issue,113560,t.contiguous() ~10 slower in eager mode compared to torch.compile,https://github.com/pytorch/pytorch/issues/113560,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2025-02-13T17:59:21Z +issue,146848,_scale_dot_product_efficient_attention eager/meta computation mismatch on output tensor shape,https://github.com/pytorch/pytorch/issues/146848,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sdpa, oncall: pt2",scan when time allows,2025-02-13T12:34:23Z +issue,132136,`torch.nn.transformer.forward` returns incorrect value inside `torch.no_grad()` blocks.,https://github.com/pytorch/pytorch/issues/132136,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-02-13T11:14:43Z +issue,146976,Adding sparse_coo_tensor objects together behaves differently with CPU and CUDA,https://github.com/pytorch/pytorch/issues/146976,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-02-13T09:00:09Z +issue,146964,Feature Request: Interface to Check Compilation Status Inside Compiled Module,https://github.com/pytorch/pytorch/issues/146964,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-13T08:58:59Z +issue,146119,File based multi gpu,https://github.com/pytorch/pytorch/issues/146119,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-02-13T07:14:37Z +issue,116393,Error: IndexError: map::at When using torch.distributed.all_reduce(tensor),https://github.com/pytorch/pytorch/issues/116393,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-02-13T02:04:22Z +issue,146944,win-vs2022-cpu-py3 test failures in test-default-2-3-lf.windows.4xlarge.nonephemeral_37054642004,https://github.com/pytorch/pytorch/issues/146944,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2025-02-12T22:19:09Z +issue,146901,Trying to use forward AD with _scaled_dot_product_efficient_attention,https://github.com/pytorch/pytorch/issues/146901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: forward ad, module: sdpa",scan when time allows,2025-02-12T22:14:49Z +issue,91439,Pytorch2.0 doesn't support compiling GRU and RNN model,https://github.com/pytorch/pytorch/issues/91439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-02-12T20:03:34Z +issue,143571,Issues linking to libtorch on M2 mac,https://github.com/pytorch/pytorch/issues/143571,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: arm,scan when time allows,2025-02-12T17:15:21Z +issue,146982,Porting Pytorch to AIX Operating System.,https://github.com/pytorch/pytorch/issues/146982,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2025-02-12T15:03:33Z +issue,146108,[associative_scan] Support lifted arguments for inductor,https://github.com/pytorch/pytorch/issues/146108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-02-12T10:07:04Z +issue,86274,TransformerEncoder/TransformerDecoder has same initial parameters for all layers,https://github.com/pytorch/pytorch/issues/86274,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-02-12T01:12:02Z +issue,146885,Add numpy.core.multiarray.scalar to default safe globals,https://github.com/pytorch/pytorch/issues/146885,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2025-02-11T13:15:59Z +issue,144012,[RFC] Add CPP Grouped GEMM Template for Inductor CPU,https://github.com/pytorch/pytorch/issues/144012,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-02-11T10:27:22Z +issue,146765,"Automatically resolve tensor mismatch issues, tensor conversion, and moving tensors to devices",https://github.com/pytorch/pytorch/issues/146765,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-02-11T09:33:38Z +issue,146831,Missing Typing for Loss Functions Causes Poor Code Completion in VSCode,https://github.com/pytorch/pytorch/issues/146831,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: typing",scan when time allows,2025-02-11T09:14:41Z +issue,78064,nn.CosineSimilarity returns value larger than 1,https://github.com/pytorch/pytorch/issues/78064,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: nn",scan when time allows,2025-02-10T23:54:06Z +issue,121595,Different Dropout behavior on macOS and Linux,https://github.com/pytorch/pytorch/issues/121595,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility, module: random",scan when time allows,2025-02-10T16:23:05Z +issue,146570,inconsistency in geometric_ on CPU and GPU,https://github.com/pytorch/pytorch/issues/146570,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2025-02-10T14:30:28Z +issue,145710,LBFGS-B with cuda implementation (and CPU one too),https://github.com/pytorch/pytorch/issues/145710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-02-09T13:20:55Z +issue,146699,`#type: ignore` linter,https://github.com/pytorch/pytorch/issues/146699,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: typing",scan when time allows,2025-02-08T13:43:48Z +issue,146711,Add different padding modes to torch.nn.utils.rnn.pad_sequence,https://github.com/pytorch/pytorch/issues/146711,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2025-02-08T01:31:43Z +issue,146697,There is a performance drop because we have not yet implemented the batching rule for aten::mse_loss_backward.,https://github.com/pytorch/pytorch/issues/146697,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: functorch, module: vmap",scan when time allows,2025-02-08T01:31:07Z +issue,146732,increase lwork/rwork sizes for all float->int conversions,https://github.com/pytorch/pytorch/issues/146732,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2025-02-07T23:31:33Z +issue,146328,DeepSeek: Grouped/Ragged gemm with offsets on the GPU,https://github.com/pytorch/pytorch/issues/146328,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: nestedtensor",scan when time allows,2025-02-07T22:23:01Z +issue,130948,[inductor][cpu]opacus_cifar10 performance regression in 2024-07-15 nightly release,https://github.com/pytorch/pytorch/issues/130948,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",scan when time allows,2025-02-07T21:04:35Z +issue,126211,Conflicting OMP implementations linked when building with MKLDNN,https://github.com/pytorch/pytorch/issues/126211,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn, module: openmp",scan when time allows,2025-02-07T20:57:00Z +issue,104702,"[feature request] Specialized memory layouts and wide blocked/tiled dtypes for cublasLt (`CUBLASLT_ORDER_COL32`, `CUBLASLT_ORDER_COL4_4R2_8C`, `CUBLASLT_ORDER_COL32_2R_4R4`)/onednn: e.g. torch.float16x32 / torch.int8x32 / torch.bits1x512 (akin to torch.quint2x4)",https://github.com/pytorch/pytorch/issues/104702,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2025-02-07T18:42:39Z +issue,104591,version libcudnn_ops_infer.so.8 not defined in file libcudnn_ops_infer.so.8 with link time reference,https://github.com/pytorch/pytorch/issues/104591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-02-07T16:49:18Z +issue,146701,Unable to Import PyTorch After Upgrade in Docker Environment,https://github.com/pytorch/pytorch/issues/146701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2025-02-07T15:40:48Z +issue,146086,Enhancements to dim_order API for Ambiguity Detection,https://github.com/pytorch/pytorch/issues/146086,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: memory format, module: python frontend",scan when time allows,2025-02-07T09:49:15Z +issue,144955,ModuleNotFoundError: No module named 'torch.privateuseone',https://github.com/pytorch/pytorch/issues/144955,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: privateuse1",scan when time allows,2025-02-07T06:18:40Z +issue,146118,[NCCL] Segfault in topology path selection when calculating max bandwidth on nightly,https://github.com/pytorch/pytorch/issues/146118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2025-02-07T03:11:21Z +issue,32570,[feature request] Out-variant and dtype argument for torch.argmax / torch.argmin / torch.argsort (and friends),https://github.com/pytorch/pytorch/issues/32570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: reductions, module: sorting and selection",scan when time allows,2025-02-07T01:12:23Z +issue,146021,multi node Error when dist.destroy_process_group,https://github.com/pytorch/pytorch/issues/146021,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2025-02-05T09:02:11Z +issue,146096,`torch.nn.functional.conv2d` 8 times slower in torch 2.5.1 compared to 2.3.1,https://github.com/pytorch/pytorch/issues/146096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2025-02-04T23:14:28Z +issue,145388,create DISABLED issues for specific runner labels,https://github.com/pytorch/pytorch/issues/145388,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-02-04T21:41:25Z +issue,25162,Incorrect Validation Accuracy Due to Distributed Sampler,https://github.com/pytorch/pytorch/issues/25162,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",scan when time allows,2025-02-04T20:08:03Z +issue,145356,[autograd] inconsistent jvp results,https://github.com/pytorch/pytorch/issues/145356,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2025-02-04T19:00:46Z +issue,145947,flops counter error with PyTorch2.5 and 2.6,https://github.com/pytorch/pytorch/issues/145947,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flop counter, oncall: distributed",scan when time allows,2025-02-04T08:02:05Z +issue,115339,Add support for measuring elapsed time in cuda-graphs,https://github.com/pytorch/pytorch/issues/115339,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2025-02-04T00:19:02Z +issue,142145,Fix disabled 3.13 cpp_extension test failing due to registration error,https://github.com/pytorch/pytorch/issues/142145,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: privateuse1",scan when time allows,2025-02-03T20:16:03Z +issue,145592,"_pickle.UnpicklingError: invalid load key, ''.",https://github.com/pytorch/pytorch/issues/145592,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-02-03T06:55:32Z +issue,130359,FP8 Rowwise Scaling does not work on L4 GPU,https://github.com/pytorch/pytorch/issues/130359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2025-02-03T03:03:09Z +issue,146034,This lib does not have new types of loss functions like focal Loss,https://github.com/pytorch/pytorch/issues/146034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-02-02T18:57:45Z +issue,143864,Adaptive pool MPS,https://github.com/pytorch/pytorch/issues/143864,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: pooling",scan when time allows,2025-02-01T11:13:39Z +issue,145851,FIRE Relative Positional Encodings,https://github.com/pytorch/pytorch/issues/145851,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-01-31T21:01:13Z +issue,77939,Fails to compile with GCC 12.1.0,https://github.com/pytorch/pytorch/issues/77939,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-01-31T08:46:35Z +issue,146040,Poor performance of any() on tensors even worse on cuda vs cpu,https://github.com/pytorch/pytorch/issues/146040,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2025-01-30T22:50:37Z +issue,146056,RuntimeError: CUDA error: the provided PTX was compiled with an unsupported toolchain. Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.,https://github.com/pytorch/pytorch/issues/146056,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-01-30T22:50:00Z +issue,146036,"torch/autograd/graph.py"" torch.func.{grad, vjp, jacrev, hessian} don't yet support saved tensor hooks",https://github.com/pytorch/pytorch/issues/146036,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2025-01-30T22:49:24Z +issue,145665,Pickling duplicates Storage,https://github.com/pytorch/pytorch/issues/145665,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pickle,scan when time allows,2025-01-30T19:44:43Z +issue,92912,functorch.functionalize doesn't error out with logcumsumexp.out,https://github.com/pytorch/pytorch/issues/92912,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: meta tensors, module: xla",scan when time allows,2025-01-29T20:45:02Z +issue,145944,Add align_to_window option to onnx version of torch.stft,https://github.com/pytorch/pytorch/issues/145944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: onnx",scan when time allows,2025-01-29T20:16:36Z +issue,145943,Add center and padding options to onnx version of torch.stft,https://github.com/pytorch/pytorch/issues/145943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: onnx",scan when time allows,2025-01-29T20:16:25Z +issue,134236,[CI] Binary Test Workflows are failing when updating triton hash,https://github.com/pytorch/pytorch/issues/134236,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: releng",scan when time allows,2025-01-29T17:30:45Z +issue,145907,There may be a documentation Error in torch.nn.CrossEntropyLoss Formula,https://github.com/pytorch/pytorch/issues/145907,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: loss",scan when time allows,2025-01-29T16:04:35Z +issue,75184,"PyTorch source code compile fail after ""Built target fbgemm_avx2""",https://github.com/pytorch/pytorch/issues/75184,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-01-29T14:32:44Z +issue,145729,KL div between t-distribution,https://github.com/pytorch/pytorch/issues/145729,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-01-28T23:44:56Z +issue,145624,Local config flags for torch.compile,https://github.com/pytorch/pytorch/issues/145624,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2025-01-28T19:28:01Z +issue,145233,Not using set_num_threads results in very slow .all(),https://github.com/pytorch/pytorch/issues/145233,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2025-01-27T18:49:00Z +issue,145716,support arm architectures in your Docker images on docker hub? (,https://github.com/pytorch/pytorch/issues/145716,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: docker",scan when time allows,2025-01-27T15:45:37Z +issue,104113,Documentation building fails due to torchgen,https://github.com/pytorch/pytorch/issues/104113,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2025-01-27T05:06:20Z +issue,72065,Batched sparse-sparse matrix multiplication/ sparse torch.einsum,https://github.com/pytorch/pytorch/issues/72065,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-01-25T22:13:22Z +issue,107294,"'MPS' training Issue(s) with NanoGPT: -Inf, NaN's",https://github.com/pytorch/pytorch/issues/107294,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: intel, module: mps",scan when time allows,2025-01-24T22:29:26Z +issue,91368,"PyTorch memory leak reference cycle in for loop, Mac M1",https://github.com/pytorch/pytorch/issues/91368,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: memory usage, module: mps",scan when time allows,2025-01-24T22:11:02Z +issue,119677,Unexpected Gradient Results From conv1d on MacOS,https://github.com/pytorch/pytorch/issues/119677,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: correctness (silent), module: functorch, module: mps",scan when time allows,2025-01-24T21:58:06Z +issue,145493,Add a lint rule to avoid the word `interface` in C++,https://github.com/pytorch/pytorch/issues/145493,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: lint",scan when time allows,2025-01-24T20:12:08Z +issue,145453,[CUDA] Illegal Memory Access with `AdaptiveMaxPool2d`,https://github.com/pytorch/pytorch/issues/145453,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: nn",scan when time allows,2025-01-24T20:10:08Z +issue,143955,Issue with aten::_sparse_coo_tensor_with_dims_and_tensors on Apple Silicon GPU (MPS) Backend when Using Whisper Model,https://github.com/pytorch/pytorch/issues/143955,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: sparse",scan when time allows,2025-01-24T18:41:32Z +issue,145468,[inductor][torchbench] Unsupported operator issue when running the torch_multimodal_clip model with batch size 4.,https://github.com/pytorch/pytorch/issues/145468,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sdpa, oncall: pt2",scan when time allows,2025-01-24T15:48:00Z +issue,145368,The possible error in the pytorch documentation of RNN.,https://github.com/pytorch/pytorch/issues/145368,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2025-01-24T03:27:21Z +issue,51151,-fno-omit-frame-pointer by default in our builds,https://github.com/pytorch/pytorch/issues/51151,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-01-24T03:05:20Z +issue,145262,Missing create_graph arguments in torch.func apis,https://github.com/pytorch/pytorch/issues/145262,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2025-01-24T01:32:09Z +issue,144223,Support LayerNorm2d,https://github.com/pytorch/pytorch/issues/144223,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-01-24T01:22:24Z +issue,80827,Inconsistent computation of gradient in MaxUnPooling,https://github.com/pytorch/pytorch/issues/80827,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: correctness (silent), module: determinism",scan when time allows,2025-01-23T07:45:00Z +issue,145357,FP8: E5M2: The FP8 E5M2 result is not `inf` when casting a FP32 value larger than max normal value of FP8 E5M2 (57344),https://github.com/pytorch/pytorch/issues/145357,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: floatx (formerly float8), module: nans and infs",scan when time allows,2025-01-23T02:20:15Z +issue,143215,"Runners, torchbench, & the future",https://github.com/pytorch/pytorch/issues/143215,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-01-23T02:17:12Z +issue,141079,[triton 3.2] test_convolution_as_mm failure on A100,https://github.com/pytorch/pytorch/issues/141079,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-01-22T21:24:04Z +issue,145350,[CUDA] Illegal Memory Access with `ReplicationPad2D`,https://github.com/pytorch/pytorch/issues/145350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: nn",scan when time allows,2025-01-22T18:13:41Z +issue,145349,[CUDA] Illegal Memory Access with `AdaptiveAvgPool2d`,https://github.com/pytorch/pytorch/issues/145349,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: nn",scan when time allows,2025-01-22T18:13:26Z +issue,145326,[inductor][triton] refactor ASTSource.make_ir integration,https://github.com/pytorch/pytorch/issues/145326,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: user triton, oncall: pt2",scan when time allows,2025-01-22T18:07:47Z +issue,145014,Add ATen functions in native_functions.yaml to torch_in_graph_functions list automatically,https://github.com/pytorch/pytorch/issues/145014,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamo,scan when time allows,2025-01-22T15:24:47Z +issue,145242,Expose configurable path instead of using fixed path in the inductor module for serialized pattern generation,https://github.com/pytorch/pytorch/issues/145242,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-01-22T15:23:15Z +issue,140296,asynchronous copies from accelerator to cpu: what should be the expected behaviour?,https://github.com/pytorch/pytorch/issues/140296,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: python frontend",scan when time allows,2025-01-21T22:35:53Z +issue,142884,DataLoader hangs when object fails during pickling,https://github.com/pytorch/pytorch/issues/142884,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2025-01-21T21:31:18Z +issue,145043,Bug when using reparameterized model evaluating with DDP,https://github.com/pytorch/pytorch/issues/145043,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-01-21T21:14:33Z +issue,144913,UserWarning: cuDNN SDPA backward got grad_output.strides() != output.strides(),https://github.com/pytorch/pytorch/issues/144913,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: sdpa",scan when time allows,2025-01-21T17:42:01Z +issue,145145,[RFC] Improve performance for layer_norm op for cuda with revectorized,https://github.com/pytorch/pytorch/issues/145145,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2025-01-21T17:39:26Z +issue,145040,Issue: Illegal Memory Access in Backward Pass of `scaled_dot_product_attention` with Custom Attention Mask,https://github.com/pytorch/pytorch/issues/145040,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-01-21T17:38:28Z +issue,144982,[compiled autograd] It would be nice if the compiiled autograd graph was actually runnable,https://github.com/pytorch/pytorch/issues/144982,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, oncall: pt2",scan when time allows,2025-01-21T17:38:01Z +issue,121961,Numerical error when using torch.nn.functional.pad with a large array on MPS device,https://github.com/pytorch/pytorch/issues/121961,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps, module: nn",scan when time allows,2025-01-21T15:49:59Z +issue,144928,Add API to detect if activation checkpointing is enabled in the current region or not,https://github.com/pytorch/pytorch/issues/144928,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2025-01-21T15:34:11Z +issue,143913,Set up Mac builds with clang >= 17 even though Xcode only has at most clang 16,https://github.com/pytorch/pytorch/issues/143913,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: ci",scan when time allows,2025-01-20T20:22:20Z +issue,134462,Cannot create and distribute array in torch.func.grad,https://github.com/pytorch/pytorch/issues/134462,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: functorch, oncall: distributed",scan when time allows,2025-01-17T20:02:47Z +issue,144671,torch.stack for sequences,https://github.com/pytorch/pytorch/issues/144671,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: typing",scan when time allows,2025-01-17T14:32:57Z +issue,139582,nn.LSTM documentation,https://github.com/pytorch/pytorch/issues/139582,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2025-01-17T14:21:03Z +issue,144960,Memory-efficient attention is not selected if inputs's ndim != 4,https://github.com/pytorch/pytorch/issues/144960,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2025-01-17T00:44:04Z +issue,144690,Support new CUDA conda package layout natively in cpp_extension.CUDAExtension,https://github.com/pytorch/pytorch/issues/144690,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda",scan when time allows,2025-01-16T23:19:12Z +issue,93737,Improve handling of generators,https://github.com/pytorch/pytorch/issues/93737,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-01-16T22:48:32Z +issue,131450,Make torch.compile support inter-op parallelism,https://github.com/pytorch/pytorch/issues/131450,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-01-16T21:58:39Z +issue,144890,Ways the HUD compilers dashboard could be better,https://github.com/pytorch/pytorch/issues/144890,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2025-01-16T18:34:01Z +issue,144969,Support something similar to export dynamic dims for torch.compile with fullgraph=True,https://github.com/pytorch/pytorch/issues/144969,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-01-16T15:38:16Z +issue,23425,Hanging on when one gpu node return zero as loss in the context of distributed data parallel training,https://github.com/pytorch/pytorch/issues/23425,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-01-16T10:16:27Z +issue,144375,Unable to compile models using tensorrt backend: CUDNN_STATUS_BAD_PARAM_STREAM_MISMATCH,https://github.com/pytorch/pytorch/issues/144375,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-01-16T06:03:22Z +issue,144875,[Monitoring] Display on HUD the information about runners that failed to be created (which cause jobs to queue),https://github.com/pytorch/pytorch/issues/144875,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2025-01-15T23:46:59Z +issue,144688,topK for sparse Vectors,https://github.com/pytorch/pytorch/issues/144688,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-01-15T20:34:31Z +issue,141832,Optimizers' `differentiable` flag doesn't work,https://github.com/pytorch/pytorch/issues/141832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-01-15T18:58:01Z +issue,144861,Region check for in-place read and write does not always work,https://github.com/pytorch/pytorch/issues/144861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: partial aliasing,scan when time allows,2025-01-15T18:16:12Z +issue,144611,[CUDA] Illegal Memory Access with `ConvTranspose2d`,https://github.com/pytorch/pytorch/issues/144611,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-01-15T18:04:56Z +issue,144612,[CUDA] Illegal Memory Access with `torch.bmm`,https://github.com/pytorch/pytorch/issues/144612,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-01-15T18:04:45Z +issue,144711,Tensor Stride Inconsistent (?) Behavior When One of the Dimension is 1,https://github.com/pytorch/pytorch/issues/144711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2025-01-15T16:35:57Z +issue,144633,144x less efficient CPU usage when training NN past a certain width,https://github.com/pytorch/pytorch/issues/144633,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: performance",scan when time allows,2025-01-15T15:54:41Z +issue,129627,[RFC][DTensor] DTensor Strided Sharding: A More Flexible Way To Shard Tensors,https://github.com/pytorch/pytorch/issues/129627,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-01-15T05:46:53Z +issue,141789,torch.roll runs too slow at MPS backend,https://github.com/pytorch/pytorch/issues/141789,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2025-01-15T02:05:44Z +issue,141594,Marking integer inputs to a compiled region requires `torch._dynamo.assume_static_by_default = True`,https://github.com/pytorch/pytorch/issues/141594,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2025-01-14T17:41:59Z +issue,144677,Tensorboard SummaryWriter.add_hparams doesn't log hparam metrics dictionary,https://github.com/pytorch/pytorch/issues/144677,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2025-01-14T14:20:50Z +issue,75721,Sparse tensors: indexing with `index` and/or being able run `sparse_tensor[index]`,https://github.com/pytorch/pytorch/issues/75721,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-01-14T13:31:19Z +issue,144478,[dynamo] Support using UserDefinedFunction as argument (as_proxy).,https://github.com/pytorch/pytorch/issues/144478,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-01-14T09:12:05Z +issue,121894,TORCH_USE_CUDA_DSA compiler flag not working,https://github.com/pytorch/pytorch/issues/121894,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-01-14T00:18:15Z +issue,135210,Incorrect calculation of standard deviation when using complex numbers on MPS backend,https://github.com/pytorch/pytorch/issues/135210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: correctness (silent), module: mps",scan when time allows,2025-01-13T19:09:26Z +issue,144673,Thread-safe approach on temporarily changing the `set_default_type`,https://github.com/pytorch/pytorch/issues/144673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2025-01-13T18:36:08Z +issue,142029,[RFC] Add Intel GPU Support to Torch Test Cases,https://github.com/pytorch/pytorch/issues/142029,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: testing, module: xpu",scan when time allows,2025-01-13T18:19:51Z +issue,144660,Implement the `mode` property for transformed distributions,https://github.com/pytorch/pytorch/issues/144660,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2025-01-13T17:18:07Z +issue,144643,nn.functional.interpolate doesn't work correctly with NCW tensors when output W exceeds 2^23,https://github.com/pytorch/pytorch/issues/144643,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nn",scan when time allows,2025-01-13T17:17:28Z +issue,144639,torch.vmap + autograd.Function + current_level bug,https://github.com/pytorch/pytorch/issues/144639,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch, module: vmap",scan when time allows,2025-01-13T17:13:54Z +issue,143836,Integration of AdamCPR Optimizer into PyTorch,https://github.com/pytorch/pytorch/issues/143836,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2025-01-11T22:52:49Z +issue,142467,Wrong error message,https://github.com/pytorch/pytorch/issues/142467,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: error checking",scan when time allows,2025-01-11T16:43:05Z +issue,144403,Extended functionality for torch.quantization.fuse_modules,https://github.com/pytorch/pytorch/issues/144403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2025-01-11T01:16:40Z +issue,134430,"[MPS] `F.interpolate` with `mode=""nearest-exact""` differs from expected output",https://github.com/pytorch/pytorch/issues/134430,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-10T20:09:26Z +issue,73920,"pytorch/pytorch:1.8.1-cuda10.2-cudnn7-devel docker container contains cudnn 8.2, not 7.x as the name implies",https://github.com/pytorch/pytorch/issues/73920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2025-01-10T03:31:11Z +issue,60069,Symbolic trace with *args,https://github.com/pytorch/pytorch/issues/60069,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2025-01-10T02:13:04Z +issue,137484,Better error message in `torch.linalg.vector_norm`,https://github.com/pytorch/pytorch/issues/137484,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking",scan when time allows,2025-01-10T02:11:42Z +issue,144495,c10/util/BFloat16-math.h has undefined behavior,https://github.com/pytorch/pytorch/issues/144495,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: core aten",scan when time allows,2025-01-09T22:36:01Z +issue,94691,Nan is output by GRU on mps,https://github.com/pytorch/pytorch/issues/94691,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:38:38Z +issue,78929,Add type() support for mps backend,https://github.com/pytorch/pytorch/issues/78929,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",scan when time allows,2025-01-09T19:31:26Z +issue,104106,batch size unexpectedly affects model inference on Mac M1,https://github.com/pytorch/pytorch/issues/104106,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:26:46Z +issue,104832,"MPS matmul with sliced (strided) out argument produces wrong output, may corrupt memory",https://github.com/pytorch/pytorch/issues/104832,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:26:41Z +issue,103023,MPS bug: padding_idx in nn.Embedding does not prevent gradient accumulation,https://github.com/pytorch/pytorch/issues/103023,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:24:12Z +issue,107455,Moving tensor to MPS using .to(torch.device('mps') deletes entries from tensor,https://github.com/pytorch/pytorch/issues/107455,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:23:27Z +issue,102852,The operator 'aten::poisson' is not currently implemented for the MPS device,https://github.com/pytorch/pytorch/issues/102852,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-01-09T19:23:00Z +issue,107694,masked_fill_ outputs incorrect results for 'mps' tensor after transpose,https://github.com/pytorch/pytorch/issues/107694,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:22:25Z +issue,109943,Problems when loading PT files und Linux - Duda which are created under Mac Apple Silicon MPS,https://github.com/pytorch/pytorch/issues/109943,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-01-09T19:19:54Z +issue,111173,unique(return_counts=True) fails on MPS for unsorted tensors with 1M+ elements,https://github.com/pytorch/pytorch/issues/111173,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:18:18Z +issue,111479,multi_head_attention_forward generates different values on MPS compared to CPU,https://github.com/pytorch/pytorch/issues/111479,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:16:58Z +issue,101167,round float16 calculation error in mps backend,https://github.com/pytorch/pytorch/issues/101167,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T19:15:16Z +issue,113062,MPS device: Sample from MultivariateNormal distribution,https://github.com/pytorch/pytorch/issues/113062,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-01-09T19:13:15Z +issue,116857,Operator 'aten::_weight_norm_interface' not implemented MPS device.,https://github.com/pytorch/pytorch/issues/116857,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-01-09T19:10:01Z +issue,100334,MPS device inference all same value,https://github.com/pytorch/pytorch/issues/100334,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T18:59:59Z +issue,96982,[mps] conv1d outputs zeros,https://github.com/pytorch/pytorch/issues/96982,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T18:51:27Z +issue,96225,[MPS] F.conv1d and F.conv2d produce incorrect gradients when minibatch >= 2^16,https://github.com/pytorch/pytorch/issues/96225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T18:50:51Z +issue,124029,rand_like on permuted parameter not working on MPS,https://github.com/pytorch/pytorch/issues/124029,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T18:29:04Z +issue,125098,RuntimeError: MPS: Unsupported Border padding mode,https://github.com/pytorch/pytorch/issues/125098,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-01-09T18:28:14Z +issue,87129,[MPS] sum on a size=1 dim is ~5x slower than squeeze,https://github.com/pytorch/pytorch/issues/87129,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2025-01-09T18:21:12Z +issue,134534,MiniCPM-V 2.6 transformer model returns nonsense tokens when running on MPS backend vs CPU backend.,https://github.com/pytorch/pytorch/issues/134534,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T18:16:43Z +issue,136692,NotImplementedError: The operator 'aten::linalg_matrix_exp' is not currently implemented for the MPS device.,https://github.com/pytorch/pytorch/issues/136692,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2025-01-09T18:14:04Z +issue,83015,Incorrect tensor conversion to m1 MPS.,https://github.com/pytorch/pytorch/issues/83015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-09T18:11:19Z +issue,137581,Expose underlying cudastream_t for torch.cuda.Stream,https://github.com/pytorch/pytorch/issues/137581,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: python frontend",scan when time allows,2025-01-09T17:13:16Z +issue,9407,[feature request] Make `torch.gather` broadcastable.,https://github.com/pytorch/pytorch/issues/9407,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-01-09T15:06:04Z +issue,143564,[Bug] Memory leak in C++ libtorch,https://github.com/pytorch/pytorch/issues/143564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2025-01-08T12:13:25Z +issue,142703,We don't have test coverage for aarch64 (ARM64) bfloat16 feature (__ARM_FEATURE_BF16),https://github.com/pytorch/pytorch/issues/142703,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2025-01-07T21:52:47Z +issue,140563,NCCL hangs cause timeout,https://github.com/pytorch/pytorch/issues/140563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2025-01-07T21:43:59Z +issue,144328,Debug build fails to compile on x86 with WERROR=1,https://github.com/pytorch/pytorch/issues/144328,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2025-01-07T15:26:31Z +issue,141869,Use Python 3.9 type annotations,https://github.com/pytorch/pytorch/issues/141869,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2025-01-07T12:18:42Z +issue,144116,RNN batch_first argument only works on the input not h_0 when if should work on both,https://github.com/pytorch/pytorch/issues/144116,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2025-01-07T06:14:08Z +issue,144210,Add inheritance possibilities,https://github.com/pytorch/pytorch/issues/144210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2025-01-07T06:11:32Z +issue,44255,Allow to use system zstd,https://github.com/pytorch/pytorch/issues/44255,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2025-01-07T00:57:37Z +issue,142414,[dynamo] Tracking issue for multiple inheritance support,https://github.com/pytorch/pytorch/issues/142414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2025-01-06T23:23:26Z +issue,144033,Training fails with Torch 2.1.0 on Nvidia Jetpack 5.1.2,https://github.com/pytorch/pytorch/issues/144033,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: jetson,scan when time allows,2025-01-06T15:38:30Z +issue,144217,PyTorch Enum PYI stubs are invalid. Fix typing of PYI stubs,https://github.com/pytorch/pytorch/issues/144217,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: typing",scan when time allows,2025-01-06T01:57:43Z +issue,143658,Log more contextual data when nan is detected under the anomaly mode,https://github.com/pytorch/pytorch/issues/143658,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nans and infs",scan when time allows,2025-01-06T01:53:54Z +issue,43255,Adding entropy function analogous to SciPy,https://github.com/pytorch/pytorch/issues/43255,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2025-01-05T23:12:57Z +issue,50334,Function Request: np.interp,https://github.com/pytorch/pytorch/issues/50334,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2025-01-04T12:31:41Z +issue,142250,Embedding forward performance analysis,https://github.com/pytorch/pytorch/issues/142250,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-01-03T22:02:29Z +issue,144109,Uneven Sharding in DTensor Leads to unexpected tensor resolution with `full_tensor`,https://github.com/pytorch/pytorch/issues/144109,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2025-01-03T20:03:03Z +issue,132605,MPS lgamma function changes results when using broadcasting,https://github.com/pytorch/pytorch/issues/132605,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2025-01-03T19:18:15Z +issue,143931,The model compiled with torch.compile encounters an error when run.,https://github.com/pytorch/pytorch/issues/143931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2025-01-03T17:43:47Z +issue,135050,Setting a scalar and a 0D tensor or a 0D tensor and a scalar to `min` and `max` argument of `clamp()` respectively works,https://github.com/pytorch/pytorch/issues/135050,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: python frontend",scan when time allows,2025-01-03T09:39:38Z +issue,10978,[feature request] padding for torch.cat,https://github.com/pytorch/pytorch/issues/10978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2025-01-02T18:48:55Z +issue,138460,"CUDA Binary dependency chain is wrong, leading to bad binary packaging",https://github.com/pytorch/pytorch/issues/138460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2025-01-02T18:11:57Z +issue,144025,torch.cuda.empty_cache() causes extra memory usage on 'cuda:0',https://github.com/pytorch/pytorch/issues/144025,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2025-01-02T15:36:06Z +issue,144040,Torch.sparse.mm failing gradient computation at half precision.,https://github.com/pytorch/pytorch/issues/144040,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: sparse",scan when time allows,2025-01-02T15:25:45Z +issue,144001,torch.utils.flop_counter.FlopCounterMode,https://github.com/pytorch/pytorch/issues/144001,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flop counter,scan when time allows,2025-01-02T15:21:14Z +issue,132218,[profiler] CUDA runtime op has wrong device time because of lazy init,https://github.com/pytorch/pytorch/issues/132218,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2025-01-02T03:18:17Z +issue,127615,TORCH_LIBRARY breaks when passing (unexpanded) macro as namespace argument,https://github.com/pytorch/pytorch/issues/127615,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: library,scan when time allows,2025-01-01T19:17:25Z +issue,141822,F.cross_entropy unexpectedly slower than F.log_softmax + torch.gather,https://github.com/pytorch/pytorch/issues/141822,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2025-01-01T01:30:05Z +issue,139281,`torch._numpy.ndarray.astype()` does not accept Numpy Dtypes correctly,https://github.com/pytorch/pytorch/issues/139281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-12-31T16:40:05Z +issue,85773,Conda Pytorch (Pytorch channel) in WSL2 Ubuntu can't find libcudnn shared objects,https://github.com/pytorch/pytorch/issues/85773,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2024-12-31T08:25:40Z +issue,88137,"Add eq, to, masked_select, index_select, narrow to nested tensors",https://github.com/pytorch/pytorch/issues/88137,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-12-31T07:46:08Z +issue,143947,Add differentiable flag to SGD,https://github.com/pytorch/pytorch/issues/143947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-12-30T22:52:50Z +issue,48273,torch.multinomial example is incorrect,https://github.com/pytorch/pytorch/issues/48273,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs",scan when time allows,2024-12-30T19:01:11Z +issue,143965,test case got killed file_based_local_timer_test.py test_get_timer_recursive,https://github.com/pytorch/pytorch/issues/143965,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2024-12-30T14:55:03Z +issue,126344,PyTorch: RuntimeError: device >= 0 && device < num_gpus INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/126344,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-12-30T02:46:09Z +issue,122973,mps training is slower than cpu,https://github.com/pytorch/pytorch/issues/122973,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2024-12-28T19:24:44Z +issue,92132,Torchrun seems to have problem with virtual environment,https://github.com/pytorch/pytorch/issues/92132,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: r2p",scan when time allows,2024-12-28T13:50:41Z +issue,106221,RProp improvement tracker,https://github.com/pytorch/pytorch/issues/106221,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-12-27T19:20:31Z +issue,126691,TORCH_CHECK used within torch.compile does not throw legible errors,https://github.com/pytorch/pytorch/issues/126691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor,scan when time allows,2024-12-27T02:36:54Z +issue,143860,_transform_bias_rescale_qkv cpu op get error on debug build,https://github.com/pytorch/pytorch/issues/143860,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-27T02:14:28Z +issue,139602,[RFC]: Adding Triton Backend for Aten operators,https://github.com/pytorch/pytorch/issues/139602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-12-25T19:25:00Z +issue,113750,[docs] Add reference/decomp impl snippets to functions in online docs for hackability and educational purposes / compensate for some unclear language in existing docs,https://github.com/pytorch/pytorch/issues/113750,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: norms and normalization",scan when time allows,2024-12-25T18:20:22Z +issue,143561,"""aten::any"" and ""aten::all"" behavior for unsigned tensors different than signed",https://github.com/pytorch/pytorch/issues/143561,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: reductions, module: unsigned int",scan when time allows,2024-12-25T15:06:57Z +issue,74669,String dtypes for torch Tensors,https://github.com/pytorch/pytorch/issues/74669,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-12-25T14:24:17Z +issue,120407,Dark mode please 🙏🏻,https://github.com/pytorch/pytorch/issues/120407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-12-24T09:20:58Z +issue,143754,nn.LayerNorm is slower than naive implementation when dimension is low,https://github.com/pytorch/pytorch/issues/143754,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization, module: performance",scan when time allows,2024-12-23T17:12:29Z +issue,143588,Full BFGS optimizer,https://github.com/pytorch/pytorch/issues/143588,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-12-23T17:11:02Z +issue,143669,String representation of nn.MultiheadAttention should contain arguments,https://github.com/pytorch/pytorch/issues/143669,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-23T14:28:07Z +issue,139005,Dependency management corrupted in pytorch 2.5.0,https://github.com/pytorch/pytorch/issues/139005,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-12-21T00:25:32Z +issue,143655,反射填充的反向传播没有确定性实现,https://github.com/pytorch/pytorch/issues/143655,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2024-12-20T22:31:35Z +issue,143453,Compiler Bisector Improvements,https://github.com/pytorch/pytorch/issues/143453,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: inductor,scan when time allows,2024-12-20T17:40:37Z +issue,142301,Unfold,https://github.com/pytorch/pytorch/issues/142301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-20T17:35:27Z +issue,138360,[ROCm] PyTorch Profiler Seg Fault on PyTorch Nightly,https://github.com/pytorch/pytorch/issues/138360,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rocm, oncall: profiler",scan when time allows,2024-12-20T16:50:22Z +issue,47917,LibTorch cannot load PyTorch exported model,https://github.com/pytorch/pytorch/issues/47917,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: serialization",scan when time allows,2024-12-20T08:41:41Z +issue,143440,Locale issues in colab: after tensor(1j).cuda().abs() !commands cannot be executed.,https://github.com/pytorch/pytorch/issues/143440,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: third_party",scan when time allows,2024-12-19T22:55:25Z +issue,52675,[feature request] Introduce env var `TORCH_USE_CUDA=OFF` and some `torch.cuda.enabled = False` to faithfully/completely disable cuda (torch.cuda.is_available() <- False and torch.cuda.get_device_count() <- 0) and driver loading/interaction,https://github.com/pytorch/pytorch/issues/52675,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-12-19T19:09:14Z +issue,89491,build: cmake: functorch.so not installed at expected location,https://github.com/pytorch/pytorch/issues/89491,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-12-19T07:35:15Z +issue,91524,functorch.so is installed back into the source directory,https://github.com/pytorch/pytorch/issues/91524,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: functorch",scan when time allows,2024-12-19T07:34:28Z +issue,141746,How to specify the port for processes with rank > 1 in the Gloo communication backend?,https://github.com/pytorch/pytorch/issues/141746,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-12-19T03:52:52Z +issue,135611,Indexing Numpy array with Tensor Gives Unexpected Results,https://github.com/pytorch/pytorch/issues/135611,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-12-19T00:44:29Z +issue,143424,higher rank convolution,https://github.com/pytorch/pytorch/issues/143424,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-18T22:58:26Z +issue,143497,No reproducibility after ONNX export of fully converted QAT model,https://github.com/pytorch/pytorch/issues/143497,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: quantization",scan when time allows,2024-12-18T16:27:44Z +issue,143334,Applying python 'next' directly on a tensor is painfully slow,https://github.com/pytorch/pytorch/issues/143334,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-12-18T14:48:22Z +issue,109539,Torch FX SubgraphMatcher Any / Oneof Patterns,https://github.com/pytorch/pytorch/issues/109539,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: fx.passes",scan when time allows,2024-12-18T11:50:25Z +issue,143434,Missing nightly 20241217 on x86_64,https://github.com/pytorch/pytorch/issues/143434,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-12-18T00:18:37Z +issue,142025,[export] NotImplementedError: No registered serialization name for ,https://github.com/pytorch/pytorch/issues/142025,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2024-12-17T22:09:41Z +issue,139305,torch.export torchaudio kaldi module,https://github.com/pytorch/pytorch/issues/139305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2024-12-17T22:09:40Z +issue,137274,Not decomposing CompositeImplicitAutograd ops in post-dispatch export changes autocast (and possibly more) behavior when compared with eager mode,https://github.com/pytorch/pytorch/issues/137274,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2024-12-17T22:09:37Z +issue,143079,PyTorch depends on custom actions stored in contributor personal repos,https://github.com/pytorch/pytorch/issues/143079,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-12-17T21:51:46Z +issue,26340,CPU version of PyTorch on PyPI,https://github.com/pytorch/pytorch/issues/26340,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpu, oncall: releng",scan when time allows,2024-12-17T20:19:08Z +issue,142176,torch.fx.GraphModule constructor doesn't copy all parameters/buffers from source graph module,https://github.com/pytorch/pytorch/issues/142176,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: fx, oncall: pt2",scan when time allows,2024-12-17T19:29:24Z +issue,139429,Add 'no_type_conversion' flag to register_buffer(),https://github.com/pytorch/pytorch/issues/139429,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-12-17T07:11:07Z +issue,143274,torch._logging.set_logs kind of sucks for Jupyter notebooks,https://github.com/pytorch/pytorch/issues/143274,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2024-12-17T03:01:50Z +issue,143293,`bias=False` fails in `Transformer` when `batch_first=True` and in eval mode,https://github.com/pytorch/pytorch/issues/143293,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-17T03:00:39Z +issue,143306,Add CPU scalar support in addcdiv,https://github.com/pytorch/pytorch/issues/143306,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-12-17T02:41:36Z +issue,143206,Define release branch creation readiness process,https://github.com/pytorch/pytorch/issues/143206,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-12-17T01:19:44Z +issue,141417,"Could you provide a standardized interface or documentation for mapping all versions of torch, torchvision, and torchaudio?",https://github.com/pytorch/pytorch/issues/141417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: releng",scan when time allows,2024-12-17T01:19:43Z +issue,141675,Stuck at dynamo with dynamic shape enabled in PT2.5.1 which was not observed PT2.4.0.,https://github.com/pytorch/pytorch/issues/141675,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2024-12-17T00:25:01Z +issue,126921,[ONNX] view(dtype=dtype) is not supported by both onnx.export and onnx.dynamo_export,https://github.com/pytorch/pytorch/issues/126921,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-12-16T20:53:23Z +issue,142104,[compiled autograd] Does compiled autograd under torch 2.4 support using selective checkpoint?,https://github.com/pytorch/pytorch/issues/142104,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, oncall: pt2",scan when time allows,2024-12-16T20:17:29Z +issue,142392,Stft changes number of times when with center=False and n_fft > win_length,https://github.com/pytorch/pytorch/issues/142392,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: scipy compatibility",scan when time allows,2024-12-16T18:17:50Z +issue,61309,The class weights implementation for loss functions is incorrect,https://github.com/pytorch/pytorch/issues/61309,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-12-16T17:43:50Z +issue,78809,Feature Request: Hessenberg and Schur decompositions,https://github.com/pytorch/pytorch/issues/78809,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-12-16T13:32:11Z +issue,143109,Failed to build Pytorch in Alpine Linux based image,https://github.com/pytorch/pytorch/issues/143109,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: distributed",scan when time allows,2024-12-16T08:21:24Z +issue,143189,`torch.linalg.qr_ex` and `torch.linalg.lstsq_ex`,https://github.com/pytorch/pytorch/issues/143189,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-12-15T10:48:42Z +issue,68703,Feature request: Extend `torch.eye` for creating a batch of eye matrices,https://github.com/pytorch/pytorch/issues/68703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2024-12-15T08:43:02Z +issue,89884,[RFC] PyTorch Tensor Parallel(TP) User API for Distributed Training,https://github.com/pytorch/pytorch/issues/89884,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2024-12-14T07:56:40Z +issue,142146,Inconsistent Presence of 'out' Arg in cross_entropy_loss vs Other Loss Terms in Python Bindings,https://github.com/pytorch/pytorch/issues/142146,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-12-13T20:55:13Z +issue,142852,Deprecate `torch.squeeze()` with unspecified `dim`,https://github.com/pytorch/pytorch/issues/142852,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: viewing and reshaping",scan when time allows,2024-12-13T20:31:24Z +issue,143208,We should switch Tensor object (and other perf-critical objects) to use the new managed dict feature from CPython,https://github.com/pytorch/pytorch/issues/143208,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-12-13T19:38:16Z +issue,142229,[Pipelining] Backward pass tries to detach view tensor in-place,https://github.com/pytorch/pytorch/issues/142229,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-12-13T17:25:59Z +issue,142875,[user empathy day] No speedups for SAM2 model,https://github.com/pytorch/pytorch/issues/142875,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: pt2",scan when time allows,2024-12-12T17:13:48Z +issue,142825,"Add [low, high] interval for torch.distributions.beta",https://github.com/pytorch/pytorch/issues/142825,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-12-12T15:27:10Z +issue,115024,InstanceNorm1d throws bad warning,https://github.com/pytorch/pytorch/issues/115024,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-12T09:51:11Z +issue,64043,Uncleared memory use after torch.load(),https://github.com/pytorch/pytorch/issues/64043,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: serialization",scan when time allows,2024-12-12T09:22:02Z +issue,142484,`RuntimeError: MKL FFT error: Intel oneMKL DFTI ERROR: Inconsistent configuration parameters` in `TestJitCPU.test_variant_consistency_jit_fft_fft2_cpu_complex64` against mkl 2024*,https://github.com/pytorch/pytorch/issues/142484,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: intel, module: linear algebra, module: mkl",scan when time allows,2024-12-12T02:34:02Z +issue,142815,torch.linalg.solve fails on CPU with multiple threads and batch dimension,https://github.com/pytorch/pytorch/issues/142815,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: mkl",scan when time allows,2024-12-12T00:08:03Z +issue,139131,[FlexAttention] Update the way we generate kernel options,https://github.com/pytorch/pytorch/issues/139131,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flex attention, module: higher order operators, module: inductor, module: pt2-dispatcher",scan when time allows,2024-12-11T15:14:14Z +issue,142493,Custom ops are not logged to dynamo_compile/TORCH_TRACE when an error occurs,https://github.com/pytorch/pytorch/issues/142493,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: logging, oncall: pt2",scan when time allows,2024-12-11T12:15:22Z +issue,142515,[MPS] `out_channels` <= 2**16 guard for convolution is too broad,https://github.com/pytorch/pytorch/issues/142515,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",scan when time allows,2024-12-11T02:37:47Z +issue,139908,Inductor vs. Liger Performance Track,https://github.com/pytorch/pytorch/issues/139908,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-12-10T23:44:27Z +issue,67598,unnecessary warning with autocast,https://github.com/pytorch/pytorch/issues/67598,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2024-12-10T22:04:39Z +issue,58533,Migrate wheels to manylinux2010,https://github.com/pytorch/pytorch/issues/58533,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-12-10T20:15:53Z +issue,55948,Deprecate torch.stft returning real-valued tensors and torch.istft accepting real-valued inputs,https://github.com/pytorch/pytorch/issues/55948,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: fft",scan when time allows,2024-12-10T20:15:18Z +issue,129732,Torch Distributed Pipelinging API datatype mismatch,https://github.com/pytorch/pytorch/issues/129732,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pipelining, oncall: distributed",scan when time allows,2024-12-10T19:56:02Z +issue,114153,[Tracker] Inconsistencies between CPU and GPU computation,https://github.com/pytorch/pytorch/issues/114153,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability",scan when time allows,2024-12-10T18:26:21Z +issue,138800,Batching rule not defined for `aten::_make_dual`.,https://github.com/pytorch/pytorch/issues/138800,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2024-12-10T18:17:27Z +issue,115618,"RuntimeError: handle_0 INTERNAL ASSERT FAILED at ""../c10/cuda/driver_api.cpp"":15",https://github.com/pytorch/pytorch/issues/115618,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-12-10T13:25:57Z +issue,140222,inconsistency in ```torch.nn.functional.log_softmax``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140222,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: numerical-stability",scan when time allows,2024-12-10T07:03:05Z +issue,142469,Using DeviceMesh Slicing with torch.compile(),https://github.com/pytorch/pytorch/issues/142469,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2024-12-10T06:33:19Z +issue,142331,torch.is_vulkan_available() false in windows,https://github.com/pytorch/pytorch/issues/142331,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2024-12-09T22:19:04Z +issue,141963,Do an audit on skipfiles and mark more files as inline,https://github.com/pytorch/pytorch/issues/141963,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2024-12-09T21:45:39Z +issue,142315,Better fusion on triton templates,https://github.com/pytorch/pytorch/issues/142315,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-12-09T21:31:26Z +issue,141472,[sparse] `mul`: missing support with CSC arguments,https://github.com/pytorch/pytorch/issues/141472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-12-09T17:36:30Z +issue,142353,Extend custom op fwd/bwd functions to support data types beyond the basic set,https://github.com/pytorch/pytorch/issues/142353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: custom-operators,scan when time allows,2024-12-09T06:05:46Z +issue,22049,Cannot update part of the parameters in DistributedDataParallel.,https://github.com/pytorch/pytorch/issues/22049,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-12-08T11:39:42Z +issue,141934,[Dynamo] Symbolic variables are reduced to Ints in compiled graph inputs,https://github.com/pytorch/pytorch/issues/141934,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2024-12-06T21:59:29Z +issue,142171,[3.13] Fix failing module tracker dynamo-wrapped test due to recursion error,https://github.com/pytorch/pytorch/issues/142171,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-12-06T21:12:42Z +issue,140881,Implement Cut Cross Entropy,https://github.com/pytorch/pytorch/issues/140881,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend",scan when time allows,2024-12-06T20:06:01Z +issue,137182,torch_np tests failing with Numpy 2.1.1,https://github.com/pytorch/pytorch/issues/137182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: tests",scan when time allows,2024-12-06T07:35:54Z +issue,141964,Handling custom context managers,https://github.com/pytorch/pytorch/issues/141964,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-12-06T05:26:16Z +issue,141216,Implement ADOPT Optimizer into torch.optim module,https://github.com/pytorch/pytorch/issues/141216,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-12-05T22:46:57Z +issue,142174,[3.13] Fix skipped numpy 2.0 dynamo-wrapped tests,https://github.com/pytorch/pytorch/issues/142174,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python frontend, module: tests",scan when time allows,2024-12-05T21:53:40Z +issue,128395,_foreach_atan2,https://github.com/pytorch/pytorch/issues/128395,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer",scan when time allows,2024-12-05T21:21:58Z +issue,142129,Feature Request - Backport torch.special.gammainc for Backward Compatibility,https://github.com/pytorch/pytorch/issues/142129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-12-05T19:21:25Z +issue,124216,[NestedTensor] NJT with lengths set fails torch.compile(),https://github.com/pytorch/pytorch/issues/124216,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-12-05T18:59:41Z +issue,142127,[Bug]need check for duplicate malloc calls in CUDAPluggableAllocator,https://github.com/pytorch/pytorch/issues/142127,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-12-05T16:41:01Z +issue,141018,The description of key_padding_mask in float mode is incorrect,https://github.com/pytorch/pytorch/issues/141018,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-05T12:55:00Z +issue,12449,"Could not find a package configuration file provided by ""Torch"" with any of the following names:",https://github.com/pytorch/pytorch/issues/12449,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2024-12-04T08:17:16Z +issue,134682,RuntimeError: cuDNN version incompatibility: happened when I have lstm layer,https://github.com/pytorch/pytorch/issues/134682,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cudnn",scan when time allows,2024-12-04T05:46:18Z +issue,28829,torch.cuda.close request to be able to reset communications,https://github.com/pytorch/pytorch/issues/28829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-12-04T03:49:33Z +issue,141977,frozen modules,https://github.com/pytorch/pytorch/issues/141977,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-12-03T23:39:16Z +issue,141801,[typing] add `nn.Module` overload type hint to `torch.compile`.,https://github.com/pytorch/pytorch/issues/141801,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: typing",scan when time allows,2024-12-03T23:02:23Z +issue,141828,Adding to Private Type Stubs,https://github.com/pytorch/pytorch/issues/141828,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: typing",scan when time allows,2024-12-03T23:01:28Z +issue,141758,Test in `test/test_overrides.py` failed with AssertionError,https://github.com/pytorch/pytorch/issues/141758,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2024-12-03T23:00:57Z +issue,139643,[triton 3.2] test_convolution2,https://github.com/pytorch/pytorch/issues/139643,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-12-03T22:47:35Z +issue,141764,Replace Miniconda with minforge as option,https://github.com/pytorch/pytorch/issues/141764,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-12-03T22:07:24Z +issue,141558,Incremental CI testing,https://github.com/pytorch/pytorch/issues/141558,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-12-03T21:49:15Z +issue,141690,Reworking is_big_gpu check for inductor GEMMs,https://github.com/pytorch/pytorch/issues/141690,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-12-03T19:18:46Z +issue,141560,[Feature Request] Add source Inductor Python file path to each Triton kernel,https://github.com/pytorch/pytorch/issues/141560,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-12-03T19:13:25Z +issue,141727,"inductor and eager gives slightly different results for rms_norm, with `emulate_precision_casts=True`",https://github.com/pytorch/pytorch/issues/141727,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-12-03T18:26:43Z +issue,141139,Consider noting the PyTorch version for BC-breaking error messages,https://github.com/pytorch/pytorch/issues/141139,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking,scan when time allows,2024-12-03T15:54:06Z +issue,141890,Does vmap accept None out_dims?,https://github.com/pytorch/pytorch/issues/141890,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vmap,scan when time allows,2024-12-03T15:50:16Z +issue,141802,inconsistency in ```torch.Tensor.logcumsumexp``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/141802,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: python frontend",scan when time allows,2024-12-03T15:47:08Z +issue,140640,"Identical Formulas for BatchNorm, LayerNorm, InstanceNorm, GroupNorm",https://github.com/pytorch/pytorch/issues/140640,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-12-03T12:44:53Z +issue,141915,Test failures when migrating to SM89,https://github.com/pytorch/pytorch/issues/141915,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2024-12-03T04:59:14Z +issue,139381,Improved definition for Hardsigmoid,https://github.com/pytorch/pytorch/issues/139381,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-03T00:32:14Z +issue,140345,Flaky bot might be missing some data when making a judgement,https://github.com/pytorch/pytorch/issues/140345,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-12-02T23:31:00Z +issue,133833,Unexpected behavior of `CrossEntropyLoss()`,https://github.com/pytorch/pytorch/issues/133833,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-02T23:17:09Z +issue,141845,Using torch.func with saved tensor hooks,https://github.com/pytorch/pytorch/issues/141845,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2024-12-02T16:33:39Z +issue,141749,Incorrect error message on incorrect byte boundries on _scaled_mm on fp8,https://github.com/pytorch/pytorch/issues/141749,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2024-12-02T16:02:11Z +issue,112376,Contradictory Error Message for stride Argument in torch.conv_transpose3d(),https://github.com/pytorch/pytorch/issues/112376,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: error checking",scan when time allows,2024-12-02T14:14:16Z +issue,125297,Make the `sccache` cache easily available to all pytorch contributors in readonly mode,https://github.com/pytorch/pytorch/issues/125297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-12-02T12:05:18Z +issue,128412,A Permute Layer for torch.nn.Sequential,https://github.com/pytorch/pytorch/issues/128412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-12-02T10:18:02Z +issue,139019,Stabilizing the Kumaraswamy Distribution,https://github.com/pytorch/pytorch/issues/139019,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: numerical-stability",scan when time allows,2024-11-30T18:38:17Z +issue,133304,The performance regression of `conv_transpose1d` on CPU with variable-shaped inputs,https://github.com/pytorch/pytorch/issues/133304,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: intel",scan when time allows,2024-11-29T07:59:42Z +issue,113128,NCCL watchdog thread terminated with exception,https://github.com/pytorch/pytorch/issues/113128,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2024-11-28T08:31:07Z +issue,78109,Doesn't work when register hook to torch.nn.MultiheadAttention.out_proj,https://github.com/pytorch/pytorch/issues/78109,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-27T21:52:16Z +issue,72450,Memory parity with JAX attention,https://github.com/pytorch/pytorch/issues/72450,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2024-11-27T21:52:06Z +issue,57117,MultiheadAttention.out_proj.weight is not explicitly initialized,https://github.com/pytorch/pytorch/issues/57117,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-27T21:51:57Z +issue,139191,"When I use register_full_backward_hook and register_forward_hook in nn.MultiheadAttention, The MultiheadAttention executed wrong conditional branch and get wrong backward grads.",https://github.com/pytorch/pytorch/issues/139191,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:51:48Z +issue,124931,"Missing description of Transformer argument ""memory_mask"" shape in 3D (including the batch dimension) case",https://github.com/pytorch/pytorch/issues/124931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:51:43Z +issue,91435,Batch_first attribute in quantizable multiheadattention,https://github.com/pytorch/pytorch/issues/91435,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:51:34Z +issue,80738,Output for `aten::_native_multi_head_attention` appears inconsistent with entry in `native_functions.yaml`,https://github.com/pytorch/pytorch/issues/80738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:51:16Z +issue,96633,Request for adding Warning/Error feature when dropout set to 1.0 in Transformer layer,https://github.com/pytorch/pytorch/issues/96633,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:51:04Z +issue,94428,nn.TransformerEncoderLayer fastpath (BetterTransformer) is much slower with src_key_padding_mask,https://github.com/pytorch/pytorch/issues/94428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:50:55Z +issue,95210,Add parallel attention layers and Multi-Query Attention (MQA) from PaLM to the fast path for transformers,https://github.com/pytorch/pytorch/issues/95210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:50:51Z +issue,90842,nn.MultiheadAttention softmax inconsistent in training mode,https://github.com/pytorch/pytorch/issues/90842,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:50:48Z +issue,92990,Feature request: access to variable,https://github.com/pytorch/pytorch/issues/92990,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:50:47Z +issue,115362,torch._fused_sdp_choice for fake and real tensors returns different results,https://github.com/pytorch/pytorch/issues/115362,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: sdpa",scan when time allows,2024-11-27T21:50:29Z +issue,120793,"libtorch MultiheadAttention doesn't accept different dimensions for Q, K, V",https://github.com/pytorch/pytorch/issues/120793,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:50:26Z +issue,113713,[cuDNN][cuDNN V8 API] cuDNN Flash-Attention Upstreaming RFC/tracking issue,https://github.com/pytorch/pytorch/issues/113713,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: sdpa",scan when time allows,2024-11-27T21:50:21Z +issue,119188,scaled_dot_product_attention: different output between the reference implementation and the one from torch.nn.functional,https://github.com/pytorch/pytorch/issues/119188,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T21:50:17Z +issue,57721,nn.TransformerEncoder cannot deal with large negative value even when this value is masked by src_key_padding_mask,https://github.com/pytorch/pytorch/issues/57721,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: nn",scan when time allows,2024-11-27T21:49:37Z +issue,64006,Missing device and dtype description in torch.nn.MultiheadAttention,https://github.com/pytorch/pytorch/issues/64006,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2024-11-27T21:49:14Z +issue,67999,Improved Transformer and MultiHeadAttention design,https://github.com/pytorch/pytorch/issues/67999,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:49:04Z +issue,66678,Adding modern linear transformer variants to `transformer` module,https://github.com/pytorch/pytorch/issues/66678,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:48:47Z +issue,64509,Return attention weights in nn.Transformer,https://github.com/pytorch/pytorch/issues/64509,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:43:25Z +issue,64525,nn.TransformerEncoder - all nan values issues when src_key_padding_mask provided,https://github.com/pytorch/pytorch/issues/64525,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: nn",scan when time allows,2024-11-27T21:43:14Z +issue,60939,Functional multi_head_attention_forward softmax get nan for fp16 mode,https://github.com/pytorch/pytorch/issues/60939,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:43:06Z +issue,112015,op scaled_dot_product_attention case different results,https://github.com/pytorch/pytorch/issues/112015,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T21:36:53Z +issue,120790,MultiheadAttention returns incorrect results for windowed long causal mask for long input,https://github.com/pytorch/pytorch/issues/120790,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:36:48Z +issue,118972,Inconsistent usage of `attn_mask` and `is_causal` arguments,https://github.com/pytorch/pytorch/issues/118972,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:36:39Z +issue,118628,TransformerEncoderLayer fast path predicts NaN when provided attention bias,https://github.com/pytorch/pytorch/issues/118628,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:36:34Z +issue,116077,TransformerEncoderLayer and TransformerDecoderLayer do not work,https://github.com/pytorch/pytorch/issues/116077,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:36:31Z +issue,100347,nn.MultiheadAttention doesn't use efficient scaled_dot_product_attention,https://github.com/pytorch/pytorch/issues/100347,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2024-11-27T21:36:25Z +issue,100087,TransformerEncoderLayer behavior inconsistent between training and evaluation mode,https://github.com/pytorch/pytorch/issues/100087,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:36:21Z +issue,98291,Make BetterTransformer implementation non-blocking,https://github.com/pytorch/pytorch/issues/98291,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:35:31Z +issue,96883,Expected scalar type Half but found Float when running nn.MultiheadAttention with AMP,https://github.com/pytorch/pytorch/issues/96883,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:35:27Z +issue,98203,"AssertionError: was expecting embedding dimension of 22, but got 1320",https://github.com/pytorch/pytorch/issues/98203,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:35:23Z +issue,116472,Using a 4D attention mask in MultiheadAttention forward,https://github.com/pytorch/pytorch/issues/116472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:34:53Z +issue,105925,Possible speed up of nn.MultiheadAttention,https://github.com/pytorch/pytorch/issues/105925,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:34:42Z +issue,106630,no_grad() changes output of TransformerDecoder module during evaluation,https://github.com/pytorch/pytorch/issues/106630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:34:37Z +issue,110702,sliding_window attention in scaled_dot_product,https://github.com/pytorch/pytorch/issues/110702,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T21:34:24Z +issue,124937,Conflict between bias=False and why_not_sparsity_fast_path in Transformer Module,https://github.com/pytorch/pytorch/issues/124937,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T21:33:43Z +issue,138552,The default setting of PyTorch should show warning always.,https://github.com/pytorch/pytorch/issues/138552,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: nn",scan when time allows,2024-11-27T21:33:40Z +issue,138993,Support `attn_mask` in `jagged_scaled_dot_product_attention`,https://github.com/pytorch/pytorch/issues/138993,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: sdpa",scan when time allows,2024-11-27T21:33:19Z +issue,140930,[cudnn_frontend] Error: No execution plans support the graph when using CUDNN_ATTENTION in SDPA,https://github.com/pytorch/pytorch/issues/140930,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: sdpa",scan when time allows,2024-11-27T21:32:49Z +issue,69321,Poor Transformer inference scalability on CPU,https://github.com/pytorch/pytorch/issues/69321,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2024-11-27T20:13:22Z +issue,135150,Undocumented fast pass behavior in nn.TransformerEncoderLayer causes failures in test_transformerencoderlayer_cuda_float32 (__main__.TestNNDeviceTypeCUDA),https://github.com/pytorch/pytorch/issues/135150,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:12:24Z +issue,121193,Detailed documentation with an example for the causal mask in nn.TransformerEncoder,https://github.com/pytorch/pytorch/issues/121193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:12:24Z +issue,118853,MultiheadAttention returns different results for binary mask and float mask,https://github.com/pytorch/pytorch/issues/118853,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:12:23Z +issue,113590,Implement FlashFFTConv algorithm,https://github.com/pytorch/pytorch/issues/113590,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: nn, module: sdpa",scan when time allows,2024-11-27T20:12:23Z +issue,115472,Potential issue with custom transformer masks when using fast path and batch_first=True,https://github.com/pytorch/pytorch/issues/115472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:12:22Z +issue,132700,Forward Hooks in Transformer Not Working When `model.eval()` and `torch.no_grad()`,https://github.com/pytorch/pytorch/issues/132700,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:12:11Z +issue,138907,Select SDPA backend smartly by `sdp_params` and benchmark results,https://github.com/pytorch/pytorch/issues/138907,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:11:44Z +issue,138649,SDPA 2.5 Issue tracking,https://github.com/pytorch/pytorch/issues/138649,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:11:43Z +issue,124718,"Add scaled_dot_product_attention ""scale"" argument to nn.MultiHeadAttention",https://github.com/pytorch/pytorch/issues/124718,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:11:41Z +issue,113524,_fused_sdp_choice returns wrong backend in pt2,https://github.com/pytorch/pytorch/issues/113524,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:11:41Z +issue,110154,"ValueError: args contained 2 None's after flattening. When exporting a ScriptModule or ScriptFunction, no args may be None because that breaks type propagation.",https://github.com/pytorch/pytorch/issues/110154,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: onnx",scan when time allows,2024-11-27T20:10:40Z +issue,135590,Discussion about the testing methods in test_transformers.py after math backend always uses fp32,https://github.com/pytorch/pytorch/issues/135590,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:10:03Z +issue,133780,2.4.0: sdpa disallows batch-of-0 in Flash (default backend) and cuDNN.,https://github.com/pytorch/pytorch/issues/133780,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: sdpa",scan when time allows,2024-11-27T20:10:03Z +issue,129523,"`test_dummy_mha_with_nt_cuda` fails on `sm70`, `sm75`",https://github.com/pytorch/pytorch/issues/129523,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nestedtensor, module: sdpa",scan when time allows,2024-11-27T20:10:01Z +issue,123911,"`torch.softmax(inp, dtype=torch.float32).to(torch.float16)` is not equivalent to `torch.softmax(inp)` for fp16 input",https://github.com/pytorch/pytorch/issues/123911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: nn, module: numerical-stability, module: sdpa",scan when time allows,2024-11-27T20:09:58Z +issue,121465,[RFC] PagedAttention Support,https://github.com/pytorch/pytorch/issues/121465,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:09:55Z +issue,119131,"`scaled_dot_product_attention` produces NaNs for math backend, but not memory efficient backend when used in mixed precision.",https://github.com/pytorch/pytorch/issues/119131,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability, module: sdpa",scan when time allows,2024-11-27T20:09:55Z +issue,116237,support Tensor type for scale in scaled_dot_product_attention,https://github.com/pytorch/pytorch/issues/116237,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:09:53Z +issue,110681,[RFC] Scaled Dot Product Attention API Changes,https://github.com/pytorch/pytorch/issues/110681,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend, module: sdpa",scan when time allows,2024-11-27T20:09:53Z +issue,103484,No backward implementation for `torch._native_multi_head_attention`,https://github.com/pytorch/pytorch/issues/103484,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:08:28Z +issue,104739,Error reporting uses formal parameter names of downstream C++ function,https://github.com/pytorch/pytorch/issues/104739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:08:16Z +issue,138420,batch_first support for multi_head_attention_forward,https://github.com/pytorch/pytorch/issues/138420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sdpa",scan when time allows,2024-11-27T20:05:15Z +issue,137486,"[feature request] [ux] Introduce special value `SDPBackend.ALL` (or similar functionality) to mean ""all available backends"" for `torch.nn.attention.sdpa_kernel(...)` (some special value `SDPBackend.ERROR` already exists)",https://github.com/pytorch/pytorch/issues/137486,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T20:05:13Z +issue,137192,Setting `nn.TransformerEncoder()` and `nn.TransformerDecoder()` to `nn.Transformer()` doesn't get any errors with `nn.Transformer()`'s arguments having wrong type values,https://github.com/pytorch/pytorch/issues/137192,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:04:53Z +issue,137173,`batch_first` argument of `nn.Transformer()` should be `True` by default because it's not convenient to get a warning with the default settings,https://github.com/pytorch/pytorch/issues/137173,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:04:52Z +issue,131254,Masked Attention has no effect in ``TransformerEncoderLayer``,https://github.com/pytorch/pytorch/issues/131254,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:02:10Z +issue,126973,Performance Degradation in F.linear with Batch Size > 1 in Multi-Head Attention,https://github.com/pytorch/pytorch/issues/126973,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2024-11-27T20:02:10Z +issue,123642,TransformerDecoderLayer fails when on cuda and batch dimension is zero (on cpu is fine),https://github.com/pytorch/pytorch/issues/123642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:02:05Z +issue,122660,Deprecate `torch.nn.MultiHeadAttention` and `torch.nn.Transformer`-related modules,https://github.com/pytorch/pytorch/issues/122660,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T20:02:05Z +issue,121632,"[BUG] Given boolean tgt_mask, TransformerDecoder produces wrong results with MPS backend",https://github.com/pytorch/pytorch/issues/121632,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: nn",scan when time allows,2024-11-27T20:02:03Z +issue,119653,I hope PyTorch will open the sliding window local attention feature as soon as possible.,https://github.com/pytorch/pytorch/issues/119653,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sdpa,scan when time allows,2024-11-27T19:56:30Z +issue,119371,Nan output after masked TransforrmerDecoder,https://github.com/pytorch/pytorch/issues/119371,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:55:44Z +issue,116175,Performance for `TransformerEncoderLayer` and `TransformerDecoderLayer` drops severely if pytorch version changed from `pt1.10` to `pt2.0`,https://github.com/pytorch/pytorch/issues/116175,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2024-11-27T19:55:40Z +issue,114796,Add head_mask for transformers,https://github.com/pytorch/pytorch/issues/114796,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:55:36Z +issue,103836,Please consider the SCFA/dynamic flash attention for your implementation of scaled dot product attention,https://github.com/pytorch/pytorch/issues/103836,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sdpa, module: sparse",scan when time allows,2024-11-27T19:53:57Z +issue,109987,Static quantization for Transformer block : AttributeError 'function' object has no attribute 'is_cuda',https://github.com/pytorch/pytorch/issues/109987,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:52:31Z +issue,108522,nn.Transformer has dropout layers that BERT / GPT-2 do not have,https://github.com/pytorch/pytorch/issues/108522,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2024-11-27T19:52:30Z +issue,107663,`RuntimeError: expected scalar type BFloat16 but found Float` with `torch.nn.TransformerEncoder`,https://github.com/pytorch/pytorch/issues/107663,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: nn",scan when time allows,2024-11-27T19:52:28Z +issue,103668,MultiheadAttention should split embed_dim into four parameters,https://github.com/pytorch/pytorch/issues/103668,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:52:25Z +issue,102333,"transformer encoder-layer, the sample-Independent attn_mask(dim=3) has different behaviors when training and validating",https://github.com/pytorch/pytorch/issues/102333,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:52:25Z +issue,99012,"[BUG]Float32 attention mask not working with torch.autocast(""cpu"")",https://github.com/pytorch/pytorch/issues/99012,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:50:43Z +issue,97128,"MultiHeadAttention, fast path broken with `bias=False` or uneven number of heads",https://github.com/pytorch/pytorch/issues/97128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:50:41Z +issue,97111,TransformerEncoder truncates output when some token positions are masked by `src_key_padding_mask` across batch,https://github.com/pytorch/pytorch/issues/97111,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: nn",scan when time allows,2024-11-27T19:50:40Z +issue,90866,Support arbitrary masks for _nested_tensor_from_mask in nn.TransformerEncoder,https://github.com/pytorch/pytorch/issues/90866,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: nn",scan when time allows,2024-11-27T19:50:35Z +issue,89394,torch.nn.TransformerEncoderLayer missing exception description information.,https://github.com/pytorch/pytorch/issues/89394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn",scan when time allows,2024-11-27T19:50:34Z +issue,82919,Adding a warning of non-compatibility with forward hooks for the fast path of TransformerEncoderLayer,https://github.com/pytorch/pytorch/issues/82919,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:50:31Z +issue,80494,[bug] libtorch bug in nn::MultiheadAttention and nn::Transformer,https://github.com/pytorch/pytorch/issues/80494,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cpp, module: nn",scan when time allows,2024-11-27T19:48:44Z +issue,78954,Can we have Additive Attention?,https://github.com/pytorch/pytorch/issues/78954,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-27T19:48:43Z +issue,78754,Getting NotImplementedError when trying to implement E2E support for `prim::is_nested` Op in torch-mlir.,https://github.com/pytorch/pytorch/issues/78754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-11-27T19:48:42Z +issue,74092,Two consecutive nn.LayerNorm are used in transformer model when norm_first is False,https://github.com/pytorch/pytorch/issues/74092,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:48:41Z +issue,69353,Pruning `torch.nn.MultiheadAttention` causes RuntimeError,https://github.com/pytorch/pytorch/issues/69353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pruning,scan when time allows,2024-11-27T19:48:39Z +issue,67651,MultiHeadAttention in quantizable seems incorrect with batch_first=True,https://github.com/pytorch/pytorch/issues/67651,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-11-27T19:48:38Z +issue,57230,Implementation of Self Attention vs Encoder Decoder Attention Causing Early Evaluation for PyTorch/XLA,https://github.com/pytorch/pytorch/issues/57230,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-27T19:48:34Z +issue,55228,Requesting a feature to calculate attention weights.,https://github.com/pytorch/pytorch/issues/55228,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-27T19:48:32Z +issue,50258,More general MultiHeadAttention and Transformer modules,https://github.com/pytorch/pytorch/issues/50258,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-27T19:48:31Z +issue,49555,Gunicorn preload flag not working with PyTorch library,https://github.com/pytorch/pytorch/issues/49555,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:17Z +issue,46031,Custom attention,https://github.com/pytorch/pytorch/issues/46031,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:17Z +issue,49791,Caching for autoregressive decoding of Transformer,https://github.com/pytorch/pytorch/issues/49791,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2024-11-27T19:47:16Z +issue,43865,"Allow custom kwargs for forward method of nn.TransformerEncoderLayer, nn.TransformerDecoderLayer.",https://github.com/pytorch/pytorch/issues/43865,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:15Z +issue,40932,MultiheadAttention set(-inf) cause 'Nan' in loss computing,https://github.com/pytorch/pytorch/issues/40932,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: nn",scan when time allows,2024-11-27T19:47:13Z +issue,41070,Cannot define v_dim in MultiheadAttention,https://github.com/pytorch/pytorch/issues/41070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:12Z +issue,34760,incomplete implementation doubt,https://github.com/pytorch/pytorch/issues/34760,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:11Z +issue,34573,Restructure `multi_head_attention_forward`,https://github.com/pytorch/pytorch/issues/34573,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:11Z +issue,34453,[Feature request] Query padding mask for nn.MultiheadAttention,https://github.com/pytorch/pytorch/issues/34453,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:08Z +issue,32590,[FYI] MultiheadAttention / Transformer,https://github.com/pytorch/pytorch/issues/32590,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:08Z +issue,32529,addition of attention based techniques to pytorch,https://github.com/pytorch/pytorch/issues/32529,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",scan when time allows,2024-11-27T19:47:08Z +issue,31277,nn.MultiHeadAttention with different similarity measures,https://github.com/pytorch/pytorch/issues/31277,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:08Z +issue,28884,Why attn_mask is not 3D tensor in nn.MultiheadAttention?,https://github.com/pytorch/pytorch/issues/28884,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:05Z +issue,28657,[docs] Unclear input/output format for TransformerEncoderLayer,https://github.com/pytorch/pytorch/issues/28657,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:05Z +issue,25132,Feature request: Fix dimension convention for masks in transformer,https://github.com/pytorch/pytorch/issues/25132,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:04Z +issue,24930,Successive Layer Normalization in nn.Transformer,https://github.com/pytorch/pytorch/issues/24930,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:47:04Z +issue,21518,`attn_mask` in nn.MultiheadAttention is additive,https://github.com/pytorch/pytorch/issues/21518,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2024-11-27T19:45:19Z +issue,21876,nn.TransformerLayer,https://github.com/pytorch/pytorch/issues/21876,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-27T19:45:09Z +issue,138399,`dtype` promotion of `out=` functions on meta inputs not consistent.,https://github.com/pytorch/pytorch/issues/138399,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: meta tensors, module: type promotion",scan when time allows,2024-11-27T12:30:01Z +issue,141467,import torch failed after installing nightly wheel,https://github.com/pytorch/pytorch/issues/141467,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-11-27T12:05:09Z +issue,130470,nested_tensor jagged_from_list doesn't correctly handle corner case of empty list,https://github.com/pytorch/pytorch/issues/130470,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-11-27T09:12:53Z +issue,118991,Memory Leak in v2.0.1,https://github.com/pytorch/pytorch/issues/118991,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2024-11-27T05:46:21Z +issue,130137,How to get stream operators in custom backend compiler ?,https://github.com/pytorch/pytorch/issues/130137,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: pt2",scan when time allows,2024-11-27T05:20:33Z +issue,98668,Cannot use `checkpoint_sequential` with `torch.compile`,https://github.com/pytorch/pytorch/issues/98668,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, oncall: distributed, oncall: pt2",scan when time allows,2024-11-27T05:11:45Z +issue,133092,Umbrella issue for Dynamo tracing through ProcessGroup related torch.distributed APIs,https://github.com/pytorch/pytorch/issues/133092,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: pt2",scan when time allows,2024-11-27T05:02:01Z +issue,140416,[DDP + Dynamo] Failed to get compiled graphs,https://github.com/pytorch/pytorch/issues/140416,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: distributed, oncall: pt2",scan when time allows,2024-11-26T23:54:49Z +issue,141422,"What is ""recompilation profiler"" in doc? (Seems to have a dangling link)",https://github.com/pytorch/pytorch/issues/141422,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-11-26T23:22:21Z +issue,132940,Enabling GPU mixed precision algorithms,https://github.com/pytorch/pytorch/issues/132940,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-11-26T21:23:39Z +issue,140712,No fake impl or Meta kernel for Communication Operator,https://github.com/pytorch/pytorch/issues/140712,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-11-26T13:16:19Z +issue,49291,add inverse cdf for Chi-square Distribution,https://github.com/pytorch/pytorch/issues/49291,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-11-26T10:59:15Z +issue,141335,Rounding issue of tensor sizes when sharding,https://github.com/pytorch/pytorch/issues/141335,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-11-26T09:33:08Z +issue,112443,torch.export emits node outside of Core ATen IR,https://github.com/pytorch/pytorch/issues/112443,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2024-11-26T02:02:47Z +issue,141507,[AudioLM] Inductor `LoweringException: NotImplementedError: View`,https://github.com/pytorch/pytorch/issues/141507,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-11-26T00:39:33Z +issue,138217,[PT2] Don't graph break on loss.backward(),https://github.com/pytorch/pytorch/issues/138217,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compiled autograd, module: dynamo, oncall: pt2",scan when time allows,2024-11-25T22:57:16Z +issue,141343,numpy's division compiled but returning Different results,https://github.com/pytorch/pytorch/issues/141343,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, oncall: pt2",scan when time allows,2024-11-25T21:26:20Z +issue,141350,Support saving pytorch models in OCI registries,https://github.com/pytorch/pytorch/issues/141350,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-11-25T16:55:11Z +issue,141295,Memory efficient reshape for stride-0 tensors,https://github.com/pytorch/pytorch/issues/141295,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-11-25T16:53:13Z +issue,62475,Learning rate scheduler list index out of range,https://github.com/pytorch/pytorch/issues/62475,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2024-11-25T15:36:52Z +issue,141177,`CrossEntropyLoss` much slower with class probabilities vs. class indices target,https://github.com/pytorch/pytorch/issues/141177,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-11-25T15:06:49Z +issue,141449,Segmentation fault (core dumped) in `_fft_r2c`,https://github.com/pytorch/pytorch/issues/141449,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking",scan when time allows,2024-11-25T01:40:48Z +issue,141450,Segmentation fault (core dumped) in `_fft_c2c`,https://github.com/pytorch/pytorch/issues/141450,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking",scan when time allows,2024-11-25T01:40:33Z +issue,120003,[FSDP2] Eager-Mode Execution Tracker,https://github.com/pytorch/pytorch/issues/120003,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2024-11-24T19:33:37Z +issue,141203,SEGV in `torch.native_batch_norm`,https://github.com/pytorch/pytorch/issues/141203,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: sanitizers",scan when time allows,2024-11-22T23:37:36Z +issue,140879,fr_trace usability issues,https://github.com/pytorch/pytorch/issues/140879,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-11-22T21:08:34Z +issue,141337,`torch.det` with python 3.10 and torch 2.5.0+cu121 raises RuntimeError,https://github.com/pytorch/pytorch/issues/141337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2024-11-22T21:04:01Z +issue,140301,[RFC] Distributed manual_seed API for N-D parallelism,https://github.com/pytorch/pytorch/issues/140301,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2024-11-22T21:03:43Z +issue,88320,`torch.load()` cannot load data saved at non-zero position in a file (`failed finding central directory`),https://github.com/pytorch/pytorch/issues/88320,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-11-22T08:48:57Z +issue,141125,[Dynamo] Dynamo is not aware of the function users override,https://github.com/pytorch/pytorch/issues/141125,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-11-22T03:13:20Z +issue,141277,Multiple torch.{*}_copy variants are not in generated docs/docs skiplist,https://github.com/pytorch/pytorch/issues/141277,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-11-21T20:21:26Z +issue,93743,Support data-dependent control flow operators,https://github.com/pytorch/pytorch/issues/93743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: onnx, oncall: pt2",scan when time allows,2024-11-21T19:22:04Z +issue,74235,RuntimeError: CUDA error: unspecified launch failure,https://github.com/pytorch/pytorch/issues/74235,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2024-11-21T02:09:20Z +issue,140237,Cannot export FeaStConv-based network to onnx via torch.jit.script,https://github.com/pytorch/pytorch/issues/140237,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-11-21T01:56:00Z +issue,140787,Apple Silicon Unified Memory usage,https://github.com/pytorch/pytorch/issues/140787,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2024-11-21T01:22:46Z +issue,140706,"MPS random number generation is slow, if not hanging forever",https://github.com/pytorch/pytorch/issues/140706,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2024-11-21T01:20:44Z +issue,98925,Request for deterministic support for reflection_pad2d_backward_cuda,https://github.com/pytorch/pytorch/issues/98925,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism, module: padding",scan when time allows,2024-11-20T21:21:32Z +issue,141144,"Spsolve for sparse, complex matrices",https://github.com/pytorch/pytorch/issues/141144,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-11-20T19:06:58Z +issue,140703,[Feature Request] The `.forward()`/etc. shape API,https://github.com/pytorch/pytorch/issues/140703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-19T21:49:44Z +issue,140892,inconsistency in ```torch.special.xlogy``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140892,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: numerical-stability",scan when time allows,2024-11-19T14:43:15Z +issue,68009,Properly document the `to.dtype_layout` overload of `Tensor.to`,https://github.com/pytorch/pytorch/issues/68009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-11-19T06:22:18Z +issue,140596,[export][non-strict] Passing in kwargs torch.export fails at non_strict_utils.fakify(),https://github.com/pytorch/pytorch/issues/140596,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: export, oncall: pt2",scan when time allows,2024-11-18T21:37:55Z +issue,140592,Inductor pattern matcher - missing val,https://github.com/pytorch/pytorch/issues/140592,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-11-18T20:22:26Z +issue,41485,GLOO infiniband with PyTorch,https://github.com/pytorch/pytorch/issues/41485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-11-18T18:31:56Z +issue,140560,`skipIfXYZ` decorators will skip CPU tests as well,https://github.com/pytorch/pytorch/issues/140560,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2024-11-18T18:11:57Z +issue,138957,`DataLoader` probably shouldn't use `fork` by default,https://github.com/pytorch/pytorch/issues/138957,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2024-11-18T18:08:21Z +issue,50346,how to save weights when using RPC framework,https://github.com/pytorch/pytorch/issues/50346,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2024-11-18T17:04:45Z +issue,139324,[FR] Support sub-group partition ProcessGroup,https://github.com/pytorch/pytorch/issues/139324,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-11-17T02:17:44Z +issue,124016,Expand stale management to Issues,https://github.com/pytorch/pytorch/issues/124016,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-11-16T17:16:34Z +issue,138248,"Adam optimizer without first moment estimate, for less vram",https://github.com/pytorch/pytorch/issues/138248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-11-16T13:36:16Z +issue,140711,"after around 13 iteration, tensor img_syn contains many ""NaN"" value when trained on cifari10, but on minist dataset, it works fine",https://github.com/pytorch/pytorch/issues/140711,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-16T05:54:47Z +issue,140701,inconsistency in ```torch.special.polygamma``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2024-11-16T00:33:49Z +issue,140796,inconsistency in ```torch.special.xlog1py``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140796,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2024-11-16T00:32:32Z +issue,139520,inductor/test_move_constructors_to_cuda.py::TestMoveConstructorsToCuda::test_multi_gpu unit test failure,https://github.com/pytorch/pytorch/issues/139520,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-11-15T21:17:45Z +issue,140570,"On the Apple M1 GPU, torch.where() finds tensor entries in the wrong place",https://github.com/pytorch/pytorch/issues/140570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: correctness (silent), module: mps",scan when time allows,2024-11-15T18:55:57Z +issue,140513,"Using @torch.compile, there are slight differences in output when run on different GPUs on the same machine.",https://github.com/pytorch/pytorch/issues/140513,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, oncall: pt2",scan when time allows,2024-11-15T17:36:52Z +issue,140704,vmap with mps tensor fails to recognized batched shape,https://github.com/pytorch/pytorch/issues/140704,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: mps, module: vmap",scan when time allows,2024-11-15T16:29:42Z +issue,140687,“Can't swap an already initialized allocator” when calling torch.cuda.memory.change_current_allocator,https://github.com/pytorch/pytorch/issues/140687,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-11-15T15:33:54Z +issue,140778,Floating point exception (core dumped) in `torch.sparse.sampled_addmm`,https://github.com/pytorch/pytorch/issues/140778,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-11-15T15:32:51Z +issue,92134,[Bug][Dataloader] unable to mmap 2048 bytes from file : Cannot allocate memory (12),https://github.com/pytorch/pytorch/issues/92134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2024-11-15T05:09:14Z +issue,115500,Batchnorm NAN in amp autocast mode.,https://github.com/pytorch/pytorch/issues/115500,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: norms and normalization",scan when time allows,2024-11-14T10:22:41Z +issue,103150,Build fails at linking torch_shm_manager on aarch64,https://github.com/pytorch/pytorch/issues/103150,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-11-14T09:09:28Z +issue,127388,"torch=2.1, import torch; fails AttributeError: '_OpNamespace' 'aten' object has no attribute 'sym_constrain_range_for_size'",https://github.com/pytorch/pytorch/issues/127388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: quantization",scan when time allows,2024-11-14T08:27:10Z +issue,125134,binaries/dump_operator_names.cc missing iostream include,https://github.com/pytorch/pytorch/issues/125134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: wsl",scan when time allows,2024-11-14T05:45:36Z +issue,140516,[autograd engine && compiled autograd] support for partial backward compilation,https://github.com/pytorch/pytorch/issues/140516,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: compiled autograd, oncall: pt2",scan when time allows,2024-11-14T03:09:27Z +issue,140602,"torch.from_numpy() returns a cpu tensor, even under `with torch.device(""meta"")`",https://github.com/pytorch/pytorch/issues/140602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: meta tensors, module: numpy",scan when time allows,2024-11-14T03:04:09Z +issue,112727,[dynamo] Implement iter fallback (and possibly all iters/generators) as polyfill,https://github.com/pytorch/pytorch/issues/112727,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-11-13T22:42:36Z +issue,109309,Support the `ExitStack` context manager (or a simplified version),https://github.com/pytorch/pytorch/issues/109309,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-11-13T22:17:35Z +issue,134067,subprocess tensor creation gets stuck,https://github.com/pytorch/pytorch/issues/134067,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: multiprocessing",scan when time allows,2024-11-13T02:56:55Z +issue,140396,Optional ```clamp_k``` Argument for Handling Out-of-Range Values in ```torch.topk```,https://github.com/pytorch/pytorch/issues/140396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2024-11-13T02:36:49Z +issue,140348,inconsistency in ```torch.special.i0e``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140348,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: numerical-stability, module: special",scan when time allows,2024-11-13T01:19:05Z +issue,140017,ghstack-mergebility-check often gives false positives,https://github.com/pytorch/pytorch/issues/140017,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-11-12T21:49:31Z +issue,140018,TargetDeterminator skips op_info tests for operator directly changed there,https://github.com/pytorch/pytorch/issues/140018,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-11-12T21:47:15Z +issue,73764,Group convolution slower than manually running separate convolutions in CUDA streams,https://github.com/pytorch/pytorch/issues/73764,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn, module: performance",scan when time allows,2024-11-12T18:27:54Z +issue,140252,inconsistency in ```torch.special.logit``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140252,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: special",scan when time allows,2024-11-11T19:54:47Z +issue,140214,torch.svd_lowrank - non reproducible,https://github.com/pytorch/pytorch/issues/140214,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: sparse",scan when time allows,2024-11-11T19:47:56Z +issue,116102,torch.func.grad with respect to intermediate outputs,https://github.com/pytorch/pytorch/issues/116102,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-11-11T16:12:03Z +issue,140012,Default value for get_num_threads,https://github.com/pytorch/pytorch/issues/140012,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2024-11-11T01:51:32Z +issue,139962,fbgemm build errors on warning that VLAs are a clang extension when building with clang18,https://github.com/pytorch/pytorch/issues/139962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2024-11-09T00:01:41Z +issue,140112,inconsistency in ```torch.nn.BatchNorm1d``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/140112,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: numerical-stability",scan when time allows,2024-11-08T17:12:56Z +issue,100741,Reading `state_dict()` in C++,https://github.com/pytorch/pytorch/issues/100741,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn, module: serialization",scan when time allows,2024-11-08T16:49:26Z +issue,140090,Add Android Support (Python Wheel),https://github.com/pytorch/pytorch/issues/140090,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: binaries",scan when time allows,2024-11-08T14:58:44Z +issue,140088,Building PyTorch from source fails when magma.h is in /usr/include,https://github.com/pytorch/pytorch/issues/140088,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: magma",scan when time allows,2024-11-08T13:46:24Z +issue,139958,Can't seem to run tip of tree on Linux,https://github.com/pytorch/pytorch/issues/139958,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-11-08T13:38:26Z +issue,139910,Setting a `str` value to `weight` and `pos_weight` argument of `BCEWithLogitsLoss()` gets an indirect error message,https://github.com/pytorch/pytorch/issues/139910,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-08T13:32:50Z +issue,38419,[RFC] Add tar-based IterableDataset implementation to PyTorch,https://github.com/pytorch/pytorch/issues/38419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-11-08T04:24:27Z +issue,75577,Cannot use socks5h proxy because of urllib: `urlopen error Remote end closed connection without response`,https://github.com/pytorch/pytorch/issues/75577,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: hub, module: vision",scan when time allows,2024-11-08T02:18:29Z +issue,139876,Allow using context managers in torch.fx,https://github.com/pytorch/pytorch/issues/139876,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: python frontend",scan when time allows,2024-11-07T15:57:34Z +issue,139916,Both `input` and `target` argument or at least `target` argument of `BCELoss()` shoud accept the values out of `0 <= x <= 1` so that `BCELoss()` with `torch.sigmoid()` and `nn.Sigmoid()` gets the same results as `BCEWithLogitsLoss()`.,https://github.com/pytorch/pytorch/issues/139916,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-11-07T15:57:16Z +issue,139609,There is a bug in torch.cat for non-standard boolean,https://github.com/pytorch/pytorch/issues/139609,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: edge cases, module: python frontend",scan when time allows,2024-11-07T15:55:15Z +issue,32421,Segmentation fault in C++ API torch::from_blob(...).clone(),https://github.com/pytorch/pytorch/issues/32421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-11-07T12:32:57Z +issue,46027,Padding mode for ConvTransposeNd,https://github.com/pytorch/pytorch/issues/46027,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2024-11-07T03:51:39Z +issue,139874,Modify CMake files to support Clang for Windows when building Libtorch-dependent apps,https://github.com/pytorch/pytorch/issues/139874,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: windows",scan when time allows,2024-11-06T23:47:57Z +issue,139521,boxing-unboxing overhead seems significant,https://github.com/pytorch/pytorch/issues/139521,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2024-11-06T17:03:20Z +issue,139701,`softmax` fails to commute with `permute` / `transpose`,https://github.com/pytorch/pytorch/issues/139701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-11-06T16:28:46Z +issue,139472,_refs.div.floor_rounding returns NaN instead of +- inf when a divide by 0 occurs,https://github.com/pytorch/pytorch/issues/139472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2024-11-06T16:28:24Z +issue,65936,[POLL][RFC] DataParallel Deprecation,https://github.com/pytorch/pytorch/issues/65936,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deprecation,scan when time allows,2024-11-06T12:56:48Z +issue,59786,Support negative step sizes for slicing,https://github.com/pytorch/pytorch/issues/59786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2024-11-06T02:09:55Z +issue,132386,torch cpu float16 range is not aligned with scipy on polygamma.,https://github.com/pytorch/pytorch/issues/132386,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: edge cases, module: half, module: special",scan when time allows,2024-11-06T02:08:22Z +issue,113456,Expose FakeTensor and FakeTensorMode as public APIs,https://github.com/pytorch/pytorch/issues/113456,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: faketensor, module: onnx",scan when time allows,2024-11-06T00:16:01Z +issue,139795,Support scalar tensor in `value` arg of foreach op,https://github.com/pytorch/pytorch/issues/139795,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer",scan when time allows,2024-11-05T22:34:00Z +issue,122094,Investigate torch.compile Windows support.,https://github.com/pytorch/pytorch/issues/122094,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: windows, oncall: pt2",scan when time allows,2024-11-05T18:35:24Z +issue,139648,[Tracker] complex128 tests are disabled for all foreach internally,https://github.com/pytorch/pytorch/issues/139648,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: mta, module: tests",scan when time allows,2024-11-04T18:47:38Z +issue,117348,The bitwise_XX.Scalar_Tensor should be dispatched to core IR but is currently not,https://github.com/pytorch/pytorch/issues/117348,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: decompositions,scan when time allows,2024-11-04T06:47:38Z +issue,138773,[funcol] functional collectives are 67% slower than torch.distributed collectives,https://github.com/pytorch/pytorch/issues/138773,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: distributed",scan when time allows,2024-11-03T02:06:37Z +issue,139304,"[RFC][Pipelining] RNG state communication, avoid seed checkpoint",https://github.com/pytorch/pytorch/issues/139304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-11-01T21:44:49Z +issue,139333,torch.nn.ReflectionPad2d throws CUDA error on large tensors,https://github.com/pytorch/pytorch/issues/139333,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2024-11-01T18:07:52Z +issue,139342,torch.logcumsumexp runtime error inconsistency on CPU and CUDA (also documentation missing constraint),https://github.com/pytorch/pytorch/issues/139342,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: python frontend",scan when time allows,2024-11-01T18:07:29Z +issue,139245,ncclInternalError: Internal check failed,https://github.com/pytorch/pytorch/issues/139245,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2024-11-01T18:05:53Z +issue,139241,Please implement batching rule for torch.nn.functional.multi_margin_loss,https://github.com/pytorch/pytorch/issues/139241,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2024-11-01T13:24:25Z +issue,139178,"the mpi backend not support to transfer float16/bfloat16 tensor and would post ""IndexError: map::at"" when transferring float16/bfloat16 tensor",https://github.com/pytorch/pytorch/issues/139178,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mpi,scan when time allows,2024-11-01T09:25:58Z +issue,139382,"RuntimeError: free_upper_bound + pytorch_used_bytes[device] <= device_total INTERNAL ASSERT FAILED at ""../c10/cuda/CUDAMallocAsyncAllocator.cpp"":527",https://github.com/pytorch/pytorch/issues/139382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-10-31T21:47:41Z +issue,139200,find_spec does something weird to Python Path when loading modules,https://github.com/pytorch/pytorch/issues/139200,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-10-31T21:46:00Z +issue,54137,Add additive angular margin loss,https://github.com/pytorch/pytorch/issues/54137,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2024-10-31T19:15:39Z +issue,68291,"INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1634272068185/work/aten/src/ATen/native/LinearAlgebraUtils.h"":244",https://github.com/pytorch/pytorch/issues/68291,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-10-31T19:07:17Z +issue,120735,"`torch.optim.lr_scheduler.{LRScheduler,*}`: lack of documentation",https://github.com/pytorch/pytorch/issues/120735,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: lrscheduler",scan when time allows,2024-10-31T16:02:21Z +issue,137368,Fused AdamW causing Illegal memory access on H100 for large models,https://github.com/pytorch/pytorch/issues/137368,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: optimizer",scan when time allows,2024-10-31T15:48:33Z +issue,127736,torch.nn.BatchNorm1d missing check on values of parameters eps and momentum,https://github.com/pytorch/pytorch/issues/127736,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-10-31T01:50:07Z +issue,139252,Ban relative imports in test/,https://github.com/pytorch/pytorch/issues/139252,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: testing, module: tests",scan when time allows,2024-10-30T17:16:40Z +issue,139269,TensorBoard images loading error,https://github.com/pytorch/pytorch/issues/139269,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, module: windows",scan when time allows,2024-10-30T16:46:29Z +issue,61173,Increased memory usage with AMP,https://github.com/pytorch/pytorch/issues/61173,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda",scan when time allows,2024-10-30T06:37:34Z +issue,139185,Operators being traced as method calls in torch.fx,https://github.com/pytorch/pytorch/issues/139185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-10-29T21:31:49Z +issue,139170,`pkg_resources` module is deprecated by setuptools,https://github.com/pytorch/pytorch/issues/139170,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-10-29T21:30:51Z +issue,139163,SubgraphMatcher may fail to match when the matching pattern having call_module IR,https://github.com/pytorch/pytorch/issues/139163,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-10-29T21:29:49Z +issue,138042,`Calculate Docker Image` step times out if run on LF runners,https://github.com/pytorch/pytorch/issues/138042,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-10-29T20:48:02Z +issue,139188,PCH build fail with sccache-v0.8.2,https://github.com/pytorch/pytorch/issues/139188,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: third_party",scan when time allows,2024-10-29T19:49:38Z +issue,106455,"vmap, jacrev, jacfwd, hessian, etc., in libTorch",https://github.com/pytorch/pytorch/issues/106455,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: functorch, module: vmap",scan when time allows,2024-10-29T17:19:41Z +issue,123628,`torch.onnx.export` doesn't correctly constfold constants and DequantLinear,https://github.com/pytorch/pytorch/issues/123628,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-10-29T16:47:49Z +issue,139201,First run lint on just the changes in the PR before running it over the entire PR,https://github.com/pytorch/pytorch/issues/139201,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-10-29T16:45:44Z +issue,139115,torch.nn.AvgPool2d works with long on cpu but not gpu,https://github.com/pytorch/pytorch/issues/139115,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: pooling",scan when time allows,2024-10-29T12:09:46Z +issue,139066,"Covert Conv3d to Conv2d, But output different",https://github.com/pytorch/pytorch/issues/139066,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2024-10-29T06:49:56Z +issue,138829,torch.dot has inconsistent support for int64 (long),https://github.com/pytorch/pytorch/issues/138829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-10-29T06:38:41Z +issue,138981,The error compiling problem of Qt and libtorch(my solution to eliminate the bug),https://github.com/pytorch/pytorch/issues/138981,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: dependency bug",scan when time allows,2024-10-29T02:25:20Z +issue,138811,"same data all reduce on H20, but results are different",https://github.com/pytorch/pytorch/issues/138811,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-29T02:07:39Z +issue,138743,Cleanup the scaling logic in runtime.triton_heuristics.triton_config,https://github.com/pytorch/pytorch/issues/138743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-10-28T18:01:24Z +issue,134399,Build errors while building PyTorch with BLIS,https://github.com/pytorch/pytorch/issues/134399,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-10-28T05:03:51Z +issue,122089,Infinite cycle on BSC tensor: base -> values -> base,https://github.com/pytorch/pytorch/issues/122089,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-10-28T02:21:09Z +issue,136216,[torch.export] Detect internal constrains,https://github.com/pytorch/pytorch/issues/136216,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2024-10-27T16:09:45Z +issue,8104,load_state_dict unexpectedly does not load Tensor to buffers that currently have None value,https://github.com/pytorch/pytorch/issues/8104,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: docs, module: nn",scan when time allows,2024-10-27T10:55:30Z +issue,138940,Libtorch torch::Tensor default constructor does not initialize on CPU like in python,https://github.com/pytorch/pytorch/issues/138940,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-10-25T23:47:13Z +issue,138898,cpu faster than mps for both GRUCell and LSTMCell on apple silicon M3 max: MPS Metal,https://github.com/pytorch/pytorch/issues/138898,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2024-10-25T13:56:08Z +issue,131312,"Pytorch Error: ImportError: /home/dp/anaconda3/envs/ComfyUI/lib/python3.11/site-packages/torch/lib/../../nvidia/cusparse/lib/libcusparse.so.12: undefined symbol: __nvJitLinkComplete_12_4, version libnvJitLink.so.12",https://github.com/pytorch/pytorch/issues/131312,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2024-10-25T07:01:49Z +issue,73131,Support writing tensorboard traces to AWS S3 (and other cloud storage services) in profiler,https://github.com/pytorch/pytorch/issues/73131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2024-10-25T06:15:11Z +issue,116242,"Building project using libtorch results in ""Failed to find nvToolsExt""",https://github.com/pytorch/pytorch/issues/116242,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-10-24T22:31:25Z +issue,45910,with torch.cuda.amp.autocast() get out of memory error when using with torch.no_grad() during validation,https://github.com/pytorch/pytorch/issues/45910,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: autograd, module: docs",scan when time allows,2024-10-24T13:22:44Z +issue,35527,`torch.bernoulli` and `torch.rand` to support `dtype=torch.bool` kwarg (and sypport operation of `tensor.uniform_` on bool tensors),https://github.com/pytorch/pytorch/issues/35527,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-10-24T12:59:01Z +issue,138564,[CI] Continuous spam of PyTorch CI build status notifications in fork,https://github.com/pytorch/pytorch/issues/138564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-10-24T09:34:07Z +issue,138673,dead code in test that can be removed,https://github.com/pytorch/pytorch/issues/138673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2024-10-23T20:22:05Z +issue,138726,[PP] Add unit tests to check for memory regressions,https://github.com/pytorch/pytorch/issues/138726,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pipelining,scan when time allows,2024-10-23T18:46:17Z +issue,137387,Simple explanation should be added to `capturable` parameter of `optim.RMSprop()`,https://github.com/pytorch/pytorch/issues/137387,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-10-23T18:00:49Z +issue,138468,os.environ modification is not thread safe,https://github.com/pytorch/pytorch/issues/138468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2024-10-23T08:36:34Z +issue,13130,"arm64 port for PyTorch, libtorch",https://github.com/pytorch/pytorch/issues/13130,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: ci",scan when time allows,2024-10-23T06:50:46Z +issue,138570,`torch.std_mean` returns `NaN` as mean of an `inf` array.,https://github.com/pytorch/pytorch/issues/138570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: python frontend",scan when time allows,2024-10-23T05:43:35Z +issue,125862,Pytorch nightly docker image invalidated layers,https://github.com/pytorch/pytorch/issues/125862,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docker",scan when time allows,2024-10-22T17:33:24Z +issue,138192,"Inconsistent behavior of `torch.tensor` in converting `NaN` to int32 on normal list input and numpy array, and different architectures.",https://github.com/pytorch/pytorch/issues/138192,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2024-10-22T15:55:19Z +issue,78262,Request for adding the possibility for training on sparse tensors,https://github.com/pytorch/pytorch/issues/78262,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-10-22T15:45:02Z +issue,137974,Can FlightRecorder set the stream key of collective and P2P OP into the entry.,https://github.com/pytorch/pytorch/issues/137974,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-22T04:18:08Z +issue,133877,Torch Distributed Elastic has no way to address non-retriable errors,https://github.com/pytorch/pytorch/issues/133877,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-21T20:39:04Z +issue,138396,`out=` meta device support.,https://github.com/pytorch/pytorch/issues/138396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2024-10-21T20:24:34Z +issue,138197,Compile error with -WERROR on clang-12,https://github.com/pytorch/pytorch/issues/138197,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-10-21T18:05:54Z +issue,138320,TeLU activation function,https://github.com/pytorch/pytorch/issues/138320,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-10-21T18:02:29Z +issue,138453,`torch.__config__.show()` silently initialises CUDA (?); forked processes fail with uninitialised CUDA,https://github.com/pytorch/pytorch/issues/138453,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2024-10-21T15:24:44Z +issue,138418,"Internal assert triggered at ivalue_inl.h:1966, Expected IntList but got Int",https://github.com/pytorch/pytorch/issues/138418,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: onnx, oncall: jit",scan when time allows,2024-10-21T15:20:05Z +issue,138387,torch::distributions module for C++ frontend API,https://github.com/pytorch/pytorch/issues/138387,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-10-21T15:09:38Z +issue,69858,Feature Request: Bayesian Personalized Ranking Loss,https://github.com/pytorch/pytorch/issues/69858,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-10-21T07:21:03Z +issue,138299,"Cache enablement should be tested inside of cache load() function, not at call site",https://github.com/pytorch/pytorch/issues/138299,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-10-19T00:24:47Z +issue,138353,Empty Dimensions returned in `aten::_trilinear`,https://github.com/pytorch/pytorch/issues/138353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-10-18T20:09:26Z +issue,138337,quantile does not work with float16: throws RuntimeError: quantile() input tensor must be either float or double dtype,https://github.com/pytorch/pytorch/issues/138337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: half",scan when time allows,2024-10-18T20:02:44Z +issue,122701,Unable to call len on nested tensor,https://github.com/pytorch/pytorch/issues/122701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-10-18T18:20:44Z +issue,138180,NestedTensor doesn't support elementwise operations?,https://github.com/pytorch/pytorch/issues/138180,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-10-18T16:43:38Z +issue,95108,`torch.nn.LazyLinear` crash when using torch.bfloat16 dtype in pytorch 1.12.0 and 1.13.0,https://github.com/pytorch/pytorch/issues/95108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-10-18T08:06:58Z +issue,138199,Faster dropout (aten.bernoulli) on cpu,https://github.com/pytorch/pytorch/issues/138199,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2024-10-18T02:23:11Z +issue,117546,interpolate::trilinear has different result between 1.8.0 and the newer version,https://github.com/pytorch/pytorch/issues/117546,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-10-18T02:01:18Z +issue,138168,nan in sparse mv result when I use cpu,https://github.com/pytorch/pytorch/issues/138168,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-10-18T00:51:21Z +issue,111003,Dynamo inlining should compile partial subgraphs / improve graph break with recursive calls,https://github.com/pytorch/pytorch/issues/111003,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-10-17T23:50:44Z +issue,78484,torch.lerp: discrepancy between CUDA and CPU (with extremal inputs),https://github.com/pytorch/pytorch/issues/78484,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,scan when time allows,2024-10-17T10:42:27Z +issue,137969,Flight recorder cannot mark the state of the batch P2P OP entry as completed.,https://github.com/pytorch/pytorch/issues/137969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-17T03:16:32Z +issue,137482,"`torch.ops.aten.divide_` leads to ""RuntimeError: result type Float can't be cast to the desired output type Int""",https://github.com/pytorch/pytorch/issues/137482,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: core aten,scan when time allows,2024-10-17T02:13:10Z +issue,136918,"ONNX export: w=tensor.shape() in torch.arange(end=w...) causes ""tensor does not have a device"" error",https://github.com/pytorch/pytorch/issues/136918,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-10-16T21:55:45Z +issue,136681,[triton_op] Automatically `tl.constexpr` user-written kernel params when they are static integers,https://github.com/pytorch/pytorch/issues/136681,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: inductor, module: user triton, oncall: pt2",scan when time allows,2024-10-16T21:41:42Z +issue,137951,Better document that non-blocking GPU -> CPU memory requires device sync,https://github.com/pytorch/pytorch/issues/137951,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda, module: docs",scan when time allows,2024-10-16T20:43:25Z +issue,136736,torch.unravel_index does not check out-of-bounds,https://github.com/pytorch/pytorch/issues/136736,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: numpy",scan when time allows,2024-10-16T17:57:54Z +issue,137167,`empty_like` with NJT on a different device fails because the offsets are not cast appropriately,https://github.com/pytorch/pytorch/issues/137167,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: tensor creation",scan when time allows,2024-10-16T17:32:16Z +issue,138017,`_disable_byteorder_record` parameter should be added to the doc of `save()`,https://github.com/pytorch/pytorch/issues/138017,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: serialization",scan when time allows,2024-10-16T13:27:14Z +issue,137872,[Perf]The delay between inference request impact the performance,https://github.com/pytorch/pytorch/issues/137872,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2024-10-16T13:26:30Z +issue,137968,inconsistency in ```torch.special.entr``` on CPU and GPU,https://github.com/pytorch/pytorch/issues/137968,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: special",scan when time allows,2024-10-16T07:41:58Z +issue,130644,Add aten::_weight_norm_interface as an ONNX supported TorchScript operator,https://github.com/pytorch/pytorch/issues/130644,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-10-15T23:43:57Z +issue,130539,"RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Reshape node. Name:'/Reshape_5' Status Message: /onnxruntime_src/onnxruntime/core/providers/cpu/tensor/reshape_helper.h:45 onnxruntime::ReshapeHelper::ReshapeHelper(const onnxruntime::TensorShape&, onnxruntime::TensorShapeVector&, bool) input_shape_size == size was false. The input tensor cannot be reshaped to the requested shape. Input shape:{1,32,512}, requested shape:{1,1,8,64}",https://github.com/pytorch/pytorch/issues/130539,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-10-15T23:39:37Z +issue,130366,Deadcode emitted when using `jit.script` with `onnx.export`.,https://github.com/pytorch/pytorch/issues/130366,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-10-15T23:34:59Z +issue,137931,TD skips relevant tests,https://github.com/pytorch/pytorch/issues/137931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: correctness (silent)",scan when time allows,2024-10-15T20:47:53Z +issue,86128,[TorchDispatch] Scalar Only Inputs Gets Matched To Tensor Schema,https://github.com/pytorch/pytorch/issues/86128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2024-10-15T14:21:34Z +issue,137882,Take device of input tensor into consideration for torch::data::transforms::Normalize and possibly other transformations,https://github.com/pytorch/pytorch/issues/137882,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-10-15T13:51:51Z +issue,133665,[PixelShuffle] PixelShuffle doesn't support channels_last feature for 5D inputs,https://github.com/pytorch/pytorch/issues/133665,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: nn",scan when time allows,2024-10-15T05:45:34Z +issue,137941,Bazel builds intermittently fail while trying to download mkl from anaconda %),https://github.com/pytorch/pytorch/issues/137941,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-10-14T23:48:16Z +issue,137811,Mechanism to attempt to create a VariableTracker but graph break and turn Dynamo off if we fail,https://github.com/pytorch/pytorch/issues/137811,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-10-14T18:13:58Z +issue,132579,Enable F16C instructions in CUTLASS? At least for the inductor backend?,https://github.com/pytorch/pytorch/issues/132579,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, oncall: pt2",scan when time allows,2024-10-14T18:01:26Z +issue,68114,CPU Memory Deallocation,https://github.com/pytorch/pytorch/issues/68114,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2024-10-14T14:22:00Z +issue,124454,torch.utils.cpp_extension.load recompiling every time,https://github.com/pytorch/pytorch/issues/124454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2024-10-14T14:03:32Z +issue,117357,[RFC] CI test stats,https://github.com/pytorch/pytorch/issues/117357,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-10-13T01:58:23Z +issue,135409,DataLoader + IterableDataset held up by slowest worker!?,https://github.com/pytorch/pytorch/issues/135409,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-10-12T12:53:37Z +issue,51075,RFC: Private CUDA memory pools,https://github.com/pytorch/pytorch/issues/51075,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs",scan when time allows,2024-10-12T09:58:19Z +issue,101798,Fix absolute links in pytorch repository and allow it to be proxied,https://github.com/pytorch/pytorch/issues/101798,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-10-12T06:38:12Z +issue,137777,Using torch from multiple processes is slow,https://github.com/pytorch/pytorch/issues/137777,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-10-11T19:47:33Z +issue,46642,Loss functions for complex tensors,https://github.com/pytorch/pytorch/issues/46642,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nn",scan when time allows,2024-10-11T18:51:03Z +issue,137373,"`dampening`, `maximize`, `foreach`, `differentiable` and `fused` parameter of `optim.SGD()` work with `str` values",https://github.com/pytorch/pytorch/issues/137373,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-10-11T18:23:47Z +issue,137367,"`lr`, `momentum`, `weight_decay` and `dampening` parameter of `optim.SGD()` work with `bool` values",https://github.com/pytorch/pytorch/issues/137367,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-10-11T18:18:05Z +issue,39395,Morphological operations,https://github.com/pytorch/pytorch/issues/39395,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-10-11T18:00:13Z +issue,68861,[In Gunicorn & multiprocessing environment] Cannot re-initialize CUDA in forked subprocess,https://github.com/pytorch/pytorch/issues/68861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-10-11T10:44:05Z +issue,137542,Sparse BSR matrix returns `False` is `.is_sparse()`,https://github.com/pytorch/pytorch/issues/137542,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-10-11T09:35:15Z +issue,134978,[RFC] A device-agnostic Python device memory related API design for stream-based accelerators,https://github.com/pytorch/pytorch/issues/134978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-10-11T05:32:36Z +issue,89688,Enable NCCL for PyTorch on Windows,https://github.com/pytorch/pytorch/issues/89688,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: distributed",scan when time allows,2024-10-11T04:18:02Z +issue,39757,max_unpool2d and max_unpool3d cpp signature should be similar,https://github.com/pytorch/pytorch/issues/39757,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2024-10-10T17:17:14Z +issue,126394,Building libtorch for android fails,https://github.com/pytorch/pytorch/issues/126394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vulkan, oncall: mobile",scan when time allows,2024-10-10T17:08:06Z +issue,137370,[CD] Implement a fix for: add +PTX to CUDA 12.4 nightly binaries,https://github.com/pytorch/pytorch/issues/137370,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-10-10T14:37:16Z +issue,136691,"Segfaulting/aborting unit tests do not show up in ""Show Additional Test Info"" section",https://github.com/pytorch/pytorch/issues/136691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-10-10T14:37:15Z +issue,129176,extending forward-mode AD docs should really point to an example,https://github.com/pytorch/pytorch/issues/129176,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2024-10-09T20:52:47Z +issue,137610,[RFC] Make Flight Recorder device generic,https://github.com/pytorch/pytorch/issues/137610,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-09T18:31:29Z +issue,136994,Add Mcdropout layer,https://github.com/pytorch/pytorch/issues/136994,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-10-09T16:15:38Z +issue,137012,[RFC] Use NCCL topology detection for Inductor communication time estimation,https://github.com/pytorch/pytorch/issues/137012,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: nccl, oncall: distributed, oncall: pt2",scan when time allows,2024-10-09T15:52:55Z +issue,114565,Create a reproducible build for LibTorch x64 on VS2022,https://github.com/pytorch/pytorch/issues/114565,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2024-10-09T15:48:42Z +issue,137564,Implement Incremental PCA with GPU support,https://github.com/pytorch/pytorch/issues/137564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-10-09T13:33:36Z +issue,120897,[memory profiling] about supporting the C++ Call Stack on the ARM Architecture,https://github.com/pytorch/pytorch/issues/120897,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: cuda, oncall: profiler",scan when time allows,2024-10-09T09:58:00Z +issue,137464,Add support for using NestedTensors with linalg operations,https://github.com/pytorch/pytorch/issues/137464,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: nestedtensor",scan when time allows,2024-10-09T08:53:07Z +issue,134388,Remove redundant copy in functional collectives,https://github.com/pytorch/pytorch/issues/134388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: distributed",scan when time allows,2024-10-09T06:57:23Z +issue,133823,Issues compiling `torch` with `mkl`,https://github.com/pytorch/pytorch/issues/133823,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: intel, module: mkl",scan when time allows,2024-10-09T02:00:22Z +issue,137086,conv2d boundaries,https://github.com/pytorch/pytorch/issues/137086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-10-08T19:57:35Z +issue,137419,Convert LSTM quantized by QAT to onnx,https://github.com/pytorch/pytorch/issues/137419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-10-08T12:14:10Z +issue,126032,all_gather with gloo backend does not work in inference mode,https://github.com/pytorch/pytorch/issues/126032,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2024-10-08T10:16:10Z +issue,136578,Bool convolutions (and other integral types),https://github.com/pytorch/pytorch/issues/136578,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: nn",scan when time allows,2024-10-07T20:59:02Z +issue,87514,invalid_arguments.cpp is busted,https://github.com/pytorch/pytorch/issues/87514,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: python frontend",scan when time allows,2024-10-07T19:33:08Z +issue,137412,Public API comparison between v2.4.1 and v2.5 release branch,https://github.com/pytorch/pytorch/issues/137412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: python frontend",scan when time allows,2024-10-07T17:10:37Z +issue,137409,Unable to save model after removing weight normalization due to remaining hooks,https://github.com/pytorch/pytorch/issues/137409,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-10-07T16:37:28Z +issue,116416,"error: cast from 'PyObject *(*)(THPCppFunction *, void *)' (aka '_object *(*)(torch::autograd::THPCppFunction *, void *)') to 'getter' (aka '_object *(*)(_object *, void *)') converts to incompatible function type [-Werror,-Wcast-function-type-strict]",https://github.com/pytorch/pytorch/issues/116416,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-10-07T13:54:42Z +issue,110285,TypeError: Got unsupported ScalarType BFloat16,https://github.com/pytorch/pytorch/issues/110285,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-10-07T09:23:02Z +issue,55879,Vulkan backend on desktop platforms,https://github.com/pytorch/pytorch/issues/55879,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: vulkan",scan when time allows,2024-10-05T07:17:34Z +issue,137172,`torch._batch_norm_impl_index` performs inconsistently on CPU and CUDA,https://github.com/pytorch/pytorch/issues/137172,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2024-10-04T21:37:22Z +issue,137336,NJTs with length doesn't support H2D transfer,https://github.com/pytorch/pytorch/issues/137336,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-10-04T21:23:38Z +issue,137261,list index out of range in cpp_extension.py,https://github.com/pytorch/pytorch/issues/137261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-10-04T21:20:10Z +issue,137174,`torch.vdot` performs inconsistently on CPU and CUDA,https://github.com/pytorch/pytorch/issues/137174,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-10-04T21:11:46Z +issue,137245,[RFC] Retiring ProcessGroupGloo's CUDA support,https://github.com/pytorch/pytorch/issues/137245,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-04T18:43:13Z +issue,137006,[RFC] Flight Precheck: ahead-of-hang collective mismatch detection,https://github.com/pytorch/pytorch/issues/137006,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-04T18:28:02Z +issue,134396,torch.sparse.softmax missing support for sparse CSR,https://github.com/pytorch/pytorch/issues/134396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-10-04T15:10:24Z +issue,116182,[ONNX] GFPGANv1.pth to onnx conversion error,https://github.com/pytorch/pytorch/issues/116182,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-10-04T01:25:14Z +issue,135048,Setting a `bool` tensor both to `min` and `max` argument of `clamp()` gets error while setting a `bool` tensor only to `min` or `max` argument of `clamp()` works,https://github.com/pytorch/pytorch/issues/135048,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: python frontend",scan when time allows,2024-10-03T16:50:43Z +issue,137089,Select parameters and buffers by pattern (using fnmatch or re),https://github.com/pytorch/pytorch/issues/137089,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer",scan when time allows,2024-10-03T03:13:34Z +issue,135927,[Discuss] Enable Windows inductor UTs and fix its timeout issue.,https://github.com/pytorch/pytorch/issues/135927,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: windows, oncall: pt2",scan when time allows,2024-10-02T16:42:33Z +issue,112872,libtorch exports miniz symbols,https://github.com/pytorch/pytorch/issues/112872,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",scan when time allows,2024-10-02T11:02:27Z +issue,136647,torch.compile HUD dashboard should have repro commands,https://github.com/pytorch/pytorch/issues/136647,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-10-01T20:36:22Z +issue,137045,Namespace error when using torch::unravel_index to implement a custom function in cuda.,https://github.com/pytorch/pytorch/issues/137045,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-10-01T17:26:15Z +issue,107298,dist.destroy_process_group did not destroy the process group well,https://github.com/pytorch/pytorch/issues/107298,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-10-01T12:48:50Z +issue,136733,"Be smart about autograd formulas saving either the input or output, depending on context",https://github.com/pytorch/pytorch/issues/136733,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: performance, oncall: pt2",scan when time allows,2024-09-30T20:01:12Z +issue,136913,[BUG] torch.tril return ZERO Tensor when tensor is large and in CUDA,https://github.com/pytorch/pytorch/issues/136913,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-09-30T19:56:09Z +issue,136926,Python implementation of RNN in docstring is broken,https://github.com/pytorch/pytorch/issues/136926,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2024-09-30T19:54:50Z +issue,136937,Setting the boolean value `True` to `hidden_size` argument of `nn.RNN()` gets a indirect error message,https://github.com/pytorch/pytorch/issues/136937,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2024-09-30T19:54:29Z +issue,136938,Setting wrong values to `num_layers` of `nn.RNN()` gets indirect error messages,https://github.com/pytorch/pytorch/issues/136938,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2024-09-30T19:54:07Z +issue,136955,Setting a wrong value to `dropout` argument of `nn.RNN()` gets an indirect error message,https://github.com/pytorch/pytorch/issues/136955,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2024-09-30T19:53:29Z +issue,109401,Interleaved isend and irecv causes hang,https://github.com/pytorch/pytorch/issues/109401,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-30T12:41:09Z +issue,135542,torch.nn.InstanceNorm2d and torch.nn.InstanceNorm3d returns nan with tensors of float16 dtype on cpu,https://github.com/pytorch/pytorch/issues/135542,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: nn",scan when time allows,2024-09-30T06:25:07Z +issue,136118,[RFC] Default torch.compile backend customization,https://github.com/pytorch/pytorch/issues/136118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-09-28T21:39:58Z +issue,91630,Periodic ROCM distribtued jobs are broken,https://github.com/pytorch/pytorch/issues/91630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2024-09-27T15:59:47Z +issue,47996,GPU Vendor-Agnosticism via Vulkan,https://github.com/pytorch/pytorch/issues/47996,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: vulkan",scan when time allows,2024-09-27T15:39:04Z +issue,136747,Provide `gather_mm` functionality and/or expand nested tensor support,https://github.com/pytorch/pytorch/issues/136747,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-09-27T08:25:26Z +issue,136680,torch.nn.InstanceNorm3d producing inconsistent output for float16 tensors on CPU and GPU,https://github.com/pytorch/pytorch/issues/136680,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: nn",scan when time allows,2024-09-27T02:29:53Z +issue,136815,Aborted (core dumped) in `torch.cuda.caching_allocator_delete`,https://github.com/pytorch/pytorch/issues/136815,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-09-27T02:16:57Z +issue,136699,false INTERNAL ASSERT FAILED in `torch._add_batch_dim`,https://github.com/pytorch/pytorch/issues/136699,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2024-09-27T02:16:17Z +issue,136812,"`CUDNN_BACKEND_OPERATION: cudnnFinalize Failed cudnn_status: CUDNN_STATUS_BAD_PARAM` when using `float16`, works for `bfloat16` / `float32`",https://github.com/pytorch/pytorch/issues/136812,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2024-09-27T02:15:38Z +issue,136717,Floating point exception (core dumped) in `torch.ao.nn.intrinsic.quantized.ConvReLU1d/ConvReLU2d/ConvReLU3d` when stride=0,https://github.com/pytorch/pytorch/issues/136717,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2024-09-27T01:57:40Z +issue,136083,ONNX Export Fails with Dynamic Slicing on Data-Dependent Value,https://github.com/pytorch/pytorch/issues/136083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-27T01:39:39Z +issue,136592,[ONNX] Single model export for HF models prompt and token phase,https://github.com/pytorch/pytorch/issues/136592,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-27T01:18:05Z +issue,131883,Iterating dataloader fails on sliced dataset,https://github.com/pytorch/pytorch/issues/131883,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-09-26T19:28:21Z +issue,132492,Wonder why _MultiProcessingDataLoaderIter.__next__ too slow？,https://github.com/pytorch/pytorch/issues/132492,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2024-09-26T19:09:46Z +issue,136588,DistributedSampler shuffle option doesn't work as expected,https://github.com/pytorch/pytorch/issues/136588,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-09-26T18:57:55Z +issue,136664,[BUG] torch/extension.h: undefined symbol,https://github.com/pytorch/pytorch/issues/136664,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: pybind",scan when time allows,2024-09-26T17:00:57Z +issue,106634,Confusing error message for DataLoader with num_workers=0 and non-zero timeout,https://github.com/pytorch/pytorch/issues/106634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-09-26T10:06:46Z +issue,136621,A `complex` tensor with `linalg.matrix_norm()` returns a `float` tensor is returned even if I set `dtype=torch.complex64`,https://github.com/pytorch/pytorch/issues/136621,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: linear algebra",scan when time allows,2024-09-26T02:56:49Z +issue,136656,[NJT] Dropout(0.0) with NJT increments cuda rng_state (only for no-compile),https://github.com/pytorch/pytorch/issues/136656,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-09-26T02:33:01Z +issue,98164,forward AD implimentation : _scaled_dot_product_efficient_attention,https://github.com/pytorch/pytorch/issues/98164,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2024-09-25T06:32:49Z +issue,133040,Add Test for Inductor and Dynamo Config BC Breakages,https://github.com/pytorch/pytorch/issues/133040,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: inductor, oncall: pt2",scan when time allows,2024-09-25T00:30:26Z +issue,136568,Setting a `complex` tensor to `linalg.vector_norm()` returns a `float` tensor,https://github.com/pytorch/pytorch/issues/136568,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: norms and normalization",scan when time allows,2024-09-24T21:31:58Z +issue,136479,The small shape change of input tensor leads to a significant increase in GPU memory usage in Conv3D,https://github.com/pytorch/pytorch/issues/136479,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: memory usage, module: nn",scan when time allows,2024-09-24T15:17:40Z +issue,136409,Linear Sum Assignment,https://github.com/pytorch/pytorch/issues/136409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer",scan when time allows,2024-09-24T13:31:45Z +issue,73646,Pure C binding/wrapper with libtorch for inference applications,https://github.com/pytorch/pytorch/issues/73646,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-09-24T12:59:46Z +issue,136452,Setting a `complex` tensor to `linalg.norm()` returns a `float` tensor,https://github.com/pytorch/pytorch/issues/136452,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-09-24T04:48:13Z +issue,135724,[Inductor] Device-specific padding settings,https://github.com/pytorch/pytorch/issues/135724,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-09-24T04:44:29Z +issue,136410,Implement kaiming_uniform_ for sparse tensors.,https://github.com/pytorch/pytorch/issues/136410,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: initialization, module: nn, module: sparse",scan when time allows,2024-09-23T23:09:34Z +issue,135305,Add restrictions so that Inductor never uses more memory than eager,https://github.com/pytorch/pytorch/issues/135305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-09-23T18:24:42Z +issue,22788,Pytorch deadlock from distributed multiprocessing,https://github.com/pytorch/pytorch/issues/22788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-23T05:53:31Z +issue,132388,`amp.custom_fwd` interacts poorly with custom `autograd.Function` with `setup_context` separated from `forward`,https://github.com/pytorch/pytorch/issues/132388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: autograd",scan when time allows,2024-09-22T17:45:27Z +issue,86494,Automatic broadcasting for batch addition for sparse tensors,https://github.com/pytorch/pytorch/issues/86494,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-09-22T12:31:28Z +issue,134459,linalg.lu_factor: LU without pivoting is not implemented on the CPU,https://github.com/pytorch/pytorch/issues/134459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: linear algebra",scan when time allows,2024-09-21T15:03:11Z +issue,136357,"torch.func.vjp and sparse tensors: ""Sparse CSR tensors do not have strides""",https://github.com/pytorch/pytorch/issues/136357,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: sparse",scan when time allows,2024-09-20T16:07:25Z +issue,50341,Interpolation tracking issue,https://github.com/pytorch/pytorch/issues/50341,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2024-09-20T15:43:26Z +issue,136305,Numpy related failure of gradient and sobolengine distribution testcases on POWER and x86,https://github.com/pytorch/pytorch/issues/136305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: power",scan when time allows,2024-09-19T16:55:45Z +issue,136250,A question about `getDLDevice`,https://github.com/pytorch/pytorch/issues/136250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dlpack,scan when time allows,2024-09-19T16:49:35Z +issue,97106,slow torch import on macos,https://github.com/pytorch/pytorch/issues/97106,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, module: performance",scan when time allows,2024-09-19T08:55:40Z +issue,134700,"[c10d][MPI] Attempting to create a new group after MPI causes ""RuntimeError: Underlying Non-PrefixStore shouldn't be null""",https://github.com/pytorch/pytorch/issues/134700,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mpi, oncall: distributed",scan when time allows,2024-09-17T21:57:45Z +issue,21553,Feature Request: beta cdf,https://github.com/pytorch/pytorch/issues/21553,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-09-17T20:06:14Z +issue,135930,Integer overflow while creating nested tensors,https://github.com/pytorch/pytorch/issues/135930,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-09-17T18:55:12Z +issue,136186,"""bmm_sparse_cuda"" not implemented for 'BFloat16'",https://github.com/pytorch/pytorch/issues/136186,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: sparse",scan when time allows,2024-09-17T18:32:55Z +issue,136201,Compile with USE_CPP_CODE_COVERAGE=ON throw erros,https://github.com/pytorch/pytorch/issues/136201,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-09-17T18:19:17Z +issue,94397,jacfwd and jacrev are fundamentally broken for complex inputs,https://github.com/pytorch/pytorch/issues/94397,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex, module: functorch",scan when time allows,2024-09-17T17:51:16Z +issue,69471,Include Declarations.yaml in Libtorch distributions,https://github.com/pytorch/pytorch/issues/69471,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: cpp",scan when time allows,2024-09-17T13:04:24Z +issue,124056,MPS slows down after sleep,https://github.com/pytorch/pytorch/issues/124056,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2024-09-17T06:39:01Z +issue,98675,torch.matmul with batched CSR matrix,https://github.com/pytorch/pytorch/issues/98675,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-09-17T05:16:39Z +issue,127517,"With pytest8.2.0 or later, test cases under test/distributed/ execute will meet issue ""object has no attribute 'runTest'. Did you mean: 'run_test'""",https://github.com/pytorch/pytorch/issues/127517,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-17T05:06:52Z +issue,114097,[RFC] Support for Redundant Hosts in TorchElastic,https://github.com/pytorch/pytorch/issues/114097,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2024-09-16T20:03:55Z +issue,101293,Operations to shared tensors in the forked process could lead to silent crash,https://github.com/pytorch/pytorch/issues/101293,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-09-16T19:08:41Z +issue,135487,No error raised when trying in-place operation on tensor of shape[0] > 1 (single memory location),https://github.com/pytorch/pytorch/issues/135487,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: cpu, module: edge cases, module: error checking",scan when time allows,2024-09-16T18:21:24Z +issue,136031,Conv vs. Scipy tests fail on MacOS,https://github.com/pytorch/pytorch/issues/136031,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2024-09-16T16:03:19Z +issue,136138,`num_features` of `nn.BatchNorm1d()` with a tuple of `int` works,https://github.com/pytorch/pytorch/issues/136138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-16T15:58:14Z +issue,136072,"`padding_mode` parameter of `nn.ConvTranspose1d()`, `nn.ConvTranspose2d()` and `nn.ConvTranspose3d()` is not explained in the docs",https://github.com/pytorch/pytorch/issues/136072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-16T15:57:48Z +issue,136073,The real AttributeError information,https://github.com/pytorch/pytorch/issues/136073,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-16T15:57:22Z +issue,136086,Same token different output from `Conv1d`,https://github.com/pytorch/pytorch/issues/136086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2024-09-16T15:54:44Z +issue,136071,"Setting wrong type values to `stride`, `padding`, `output_padding` and `dilation` argument of `nn.ConvTranspose1d()` gets the wrong error messages saying only `tuple of ints`",https://github.com/pytorch/pytorch/issues/136071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-16T15:52:45Z +issue,136096,"Each parameter of `nn.MaxPool1d()`, `nn.MaxPool2d()` and `nn.MaxPool3d()` should have `required` or `optional`",https://github.com/pytorch/pytorch/issues/136096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-16T15:52:19Z +issue,136098,"`int`, `float` and `complex` type with `return_indices` of `nn.MaxPool1d()` also work",https://github.com/pytorch/pytorch/issues/136098,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-16T15:52:00Z +issue,136110,Setting `True` to `return_indices` argument of `nn.MaxPool1d()` with the tensor of `int` works,https://github.com/pytorch/pytorch/issues/136110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-16T15:49:52Z +issue,136145,MultiLabelMarginLoss with customized margin.,https://github.com/pytorch/pytorch/issues/136145,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-09-16T15:33:31Z +issue,135095,Add MSCCL++ as a communication backend for PyTorch,https://github.com/pytorch/pytorch/issues/135095,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2024-09-16T07:22:22Z +issue,126381,All processes running torch.distributed.destroy_process_group() create CUDA context on device 0,https://github.com/pytorch/pytorch/issues/126381,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2024-09-15T22:10:03Z +issue,121333,Automatic graph break bug reporter,https://github.com/pytorch/pytorch/issues/121333,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: logging, oncall: pt2",scan when time allows,2024-09-14T19:31:17Z +issue,55884,[Bug] RuntimeError: could not create a primitive on Xeon,https://github.com/pytorch/pytorch/issues/55884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: intel",scan when time allows,2024-09-14T06:45:21Z +issue,135705,Multiprocessing with workers returning `torch.Tensor` and limiting of the number of tasks per worker (`maxtasksperchild=1`) hangs,https://github.com/pytorch/pytorch/issues/135705,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-09-13T23:13:10Z +issue,135992,"When calculating the loss, the input data does not contain NaN, but the output contains NaN.",https://github.com/pytorch/pytorch/issues/135992,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,scan when time allows,2024-09-13T19:36:59Z +issue,135969,Add AT_DISPATCH for layernorm gamma and beta,https://github.com/pytorch/pytorch/issues/135969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2024-09-13T19:36:35Z +issue,135837,cudaFree will be called when the reserved memory > 70G for A100 80G,https://github.com/pytorch/pytorch/issues/135837,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2024-09-13T19:35:57Z +issue,135826,(L)-BFGS with Hager-Zhang line-search,https://github.com/pytorch/pytorch/issues/135826,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-09-13T19:35:28Z +issue,135640,Issue UserWarning for missing ModuleList/ModuleDict in Module classes,https://github.com/pytorch/pytorch/issues/135640,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: molly-guard, module: nn",scan when time allows,2024-09-13T19:30:21Z +issue,110866,Gradients (Jacobian) in inference,https://github.com/pytorch/pytorch/issues/110866,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-13T13:49:19Z +issue,123047,Alternate algorithm for computing MaxPool2D under specific condition.,https://github.com/pytorch/pytorch/issues/123047,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: nn, module: performance",scan when time allows,2024-09-13T09:41:25Z +issue,119549,"Way of mocking collectives, especially data-dependent ones like all2all_single",https://github.com/pytorch/pytorch/issues/119549,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, module: minifier, oncall: distributed, oncall: pt2",scan when time allows,2024-09-13T03:21:35Z +issue,118798,Automatic dynamic shapes should give up if a symbolic variable has too many guards on it,https://github.com/pytorch/pytorch/issues/118798,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2024-09-13T03:04:56Z +issue,132617,conda pytorch does not ship libnccl.so,https://github.com/pytorch/pytorch/issues/132617,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: nccl",scan when time allows,2024-09-13T02:27:21Z +issue,130968,float8 types cannot be used under deterministic mode,https://github.com/pytorch/pytorch/issues/130968,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: floatx (formerly float8)",scan when time allows,2024-09-12T18:36:33Z +issue,135660,torch.cat and torch.stack does not invoke `__torch_function__` API,https://github.com/pytorch/pytorch/issues/135660,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: nestedtensor",scan when time allows,2024-09-12T17:58:19Z +issue,66806,Missing doc for `torch.segment_reduce`,https://github.com/pytorch/pytorch/issues/66806,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy",scan when time allows,2024-09-12T14:45:54Z +issue,33122,[feature request] make torch.multinomial behaviour compliant with rnn output dimension,https://github.com/pytorch/pytorch/issues/33122,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: distributions",scan when time allows,2024-09-12T04:31:12Z +issue,135648,"Setting a `float` value to `in_channels`, `out_channels` and `kernel_size` argument of `nn.Conv1d()` gets indirect errors and in_channels and kernel_size with with non-bools",https://github.com/pytorch/pytorch/issues/135648,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn",scan when time allows,2024-09-11T16:54:57Z +issue,96265,RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation,https://github.com/pytorch/pytorch/issues/96265,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-09-11T08:52:12Z +issue,135135,TORCH_CPP_LOG_LEVEL cannot enable VLOG,https://github.com/pytorch/pytorch/issues/135135,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2024-09-11T00:13:31Z +issue,135604,"Generic BLAS provider incorrectly considered ""found""",https://github.com/pytorch/pytorch/issues/135604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-09-10T23:07:06Z +issue,135514,add torch.linalg stubs,https://github.com/pytorch/pytorch/issues/135514,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-09-10T22:07:57Z +issue,135484,"Boolean indexing not working correctly for a numpy array indexed with a boolean pytorch tensor in the case of 2d singletons (shape of (1, 1))",https://github.com/pytorch/pytorch/issues/135484,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python frontend",scan when time allows,2024-09-10T15:47:41Z +issue,134851,GPU Precision drops significantly when kernel size is increased from 3 to 5 in Conv3D,https://github.com/pytorch/pytorch/issues/134851,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: performance",scan when time allows,2024-09-10T14:11:38Z +issue,135359,"binary_cross_entropy_with_logits outputs NaN on -inf, which is inconsistent with documentation",https://github.com/pytorch/pytorch/issues/135359,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: edge cases, module: loss, module: nans and infs",scan when time allows,2024-09-10T11:19:48Z +issue,134827,ONNX NMS Operator Output Different From Torchvision Implementation,https://github.com/pytorch/pytorch/issues/134827,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-10T09:22:32Z +issue,122626,Significant performance degradation with multiprocessing in PyTorch 2.x compared to 1.13.1,https://github.com/pytorch/pytorch/issues/122626,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-09-10T08:36:47Z +issue,135267,First all_to_all_single / barrier takes a large amount of time.,https://github.com/pytorch/pytorch/issues/135267,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-10T01:24:02Z +issue,52718,Discrepancy between CPU->GPU and GPU->CPU data transfer speeds,https://github.com/pytorch/pytorch/issues/52718,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2024-09-09T23:14:40Z +issue,135489,Sympy bottleneck in eval_relation/assumptions getit,https://github.com/pytorch/pytorch/issues/135489,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: compile-time, module: dynamic shapes, oncall: pt2",scan when time allows,2024-09-09T19:13:12Z +issue,135153,TestSDPAPrivateUse1Only tests causes irreversible changes to PyTorch internal state,https://github.com/pytorch/pytorch/issues/135153,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: privateuse1, module: tests",scan when time allows,2024-09-09T19:05:49Z +issue,134664,"Questions about CVE-2022-3171, CVE-2022-3509 and CVE-2022-3510",https://github.com/pytorch/pytorch/issues/134664,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: onnx, module: protobuf, module: third_party",scan when time allows,2024-09-09T17:24:56Z +issue,135229,Disable `retain_grad` on non-leaf tensors?,https://github.com/pytorch/pytorch/issues/135229,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2024-09-09T14:39:05Z +issue,134948,Add Validity Checks for Parameter Combinations and Input Shapes in torch.nn.MaxPool2d to Prevent Unexpected Outputs,https://github.com/pytorch/pytorch/issues/134948,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-09-09T13:36:04Z +issue,134945,Broadcasting Parameter destroys hooks in PyTorch 2.4,https://github.com/pytorch/pytorch/issues/134945,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-09T10:00:59Z +issue,135382,Stop tracking FunctionalTensor bases in Python,https://github.com/pytorch/pytorch/issues/135382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2024-09-09T01:43:57Z +issue,135437,"Cannot build libtorch from source with NVHPC compilers - ""Could NOT find Threads""",https://github.com/pytorch/pytorch/issues/135437,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2024-09-09T01:33:00Z +issue,135447,[MPS] Correctness issue in backward pass of `nn.ReplicationPad1d` and `nn.ReplicationPad2d`,https://github.com/pytorch/pytorch/issues/135447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: correctness (silent), module: mps",scan when time allows,2024-09-09T01:29:18Z +issue,65198,Multi-processing leaking file descriptors,https://github.com/pytorch/pytorch/issues/65198,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2024-09-08T15:08:29Z +issue,99781,2.0.0+cu118 package missing proper libnvrtc-builtins.so.11.8,https://github.com/pytorch/pytorch/issues/99781,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",scan when time allows,2024-09-07T21:22:38Z +issue,135398,"Models that have non-tensor elements in state_dict, are not ONNX-exportable (and not JIT-traceable)",https://github.com/pytorch/pytorch/issues/135398,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-07T01:02:04Z +issue,134885,SyntaxError when Running repro.py,https://github.com/pytorch/pytorch/issues/134885,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: minifier, oncall: pt2",scan when time allows,2024-09-06T22:36:58Z +issue,135255,Log missed optimization opportunities due to dynamic shapes in Inductor,https://github.com/pytorch/pytorch/issues/135255,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2024-09-06T22:26:10Z +issue,133281,"Make CUDAPluggableAllocator support getDeviceStats, resetPeakStats",https://github.com/pytorch/pytorch/issues/133281,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-09-06T21:06:59Z +issue,133280,Pluggable Allocator Does not raise torch.cuda.OutOfMemoryError,https://github.com/pytorch/pytorch/issues/133280,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking, module: memory usage",scan when time allows,2024-09-06T21:03:41Z +issue,120585,Functionalization of `as_strided` disregards its base tensor storage.,https://github.com/pytorch/pytorch/issues/120585,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2024-09-06T21:03:40Z +issue,135330,torch.floor_divide not triggering ZeroDivisionError on GPU,https://github.com/pytorch/pytorch/issues/135330,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking",scan when time allows,2024-09-06T19:07:22Z +issue,119971,Non core ATen ops are getting dispatched as it is instead of mapping/docmposing to other ATen ops.,https://github.com/pytorch/pytorch/issues/119971,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2024-09-06T19:04:39Z +issue,128910,[pipelining] Free memory in stage after use,https://github.com/pytorch/pytorch/issues/128910,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pipelining,scan when time allows,2024-09-06T18:42:06Z +issue,134960,Issue with Weight Synchronization When Using Consecutive DDP Instances in the Same torch.distributed Environment,https://github.com/pytorch/pytorch/issues/134960,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-06T16:50:23Z +issue,135362,MacOS runner fails at `Complete runner step`,https://github.com/pytorch/pytorch/issues/135362,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2024-09-06T16:26:31Z +issue,135198,lstm quantize issues,https://github.com/pytorch/pytorch/issues/135198,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rnn, oncall: quantization",scan when time allows,2024-09-06T15:29:01Z +issue,134860,JIT tracing a quantized model with hooks is broken,https://github.com/pytorch/pytorch/issues/134860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: quantization",scan when time allows,2024-09-06T15:26:09Z +issue,43369,"torch.load(.., map_location='cpu') fails when unserializing cuda tensors on a cpu-only device serialized with pickle",https://github.com/pytorch/pytorch/issues/43369,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pickle, module: serialization",scan when time allows,2024-09-06T13:55:23Z +issue,129634,Type Error when using from_numpy,https://github.com/pytorch/pytorch/issues/129634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-09-06T12:49:46Z +issue,123973,Dynamo-based ONNX Export: Failed to produce a graph during tracing as no tensor operations were found.,https://github.com/pytorch/pytorch/issues/123973,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2024-09-05T23:55:37Z +issue,44710,AMP support for libtorch/c++,https://github.com/pytorch/pytorch/issues/44710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2024-09-05T19:23:52Z +issue,50092,batch_isend_irecv: the receiving end cannot receive large tensors from the sending end correctly,https://github.com/pytorch/pytorch/issues/50092,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-05T04:04:34Z +issue,99807,AOTAutograd/Inductor file system cache,https://github.com/pytorch/pytorch/issues/99807,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-09-05T02:17:51Z +issue,135091,Any support for sparse linear layer in the future?,https://github.com/pytorch/pytorch/issues/135091,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-09-04T23:15:30Z +issue,123971,Dynamo Export: Support for mutating module attributes,https://github.com/pytorch/pytorch/issues/123971,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2024-09-04T23:14:19Z +issue,128325,Cannot export onnx model with autograd.functional.jacobian,https://github.com/pytorch/pytorch/issues/128325,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-04T21:46:03Z +issue,134926,register_after_fork fails on python built without multiprocessing support,https://github.com/pytorch/pytorch/issues/134926,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: python frontend",scan when time allows,2024-09-04T20:57:30Z +issue,57118,How to view VLOG information,https://github.com/pytorch/pytorch/issues/57118,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2024-09-04T19:25:04Z +issue,36380,Population Count Op,https://github.com/pytorch/pytorch/issues/36380,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-09-04T17:09:07Z +issue,123353,Type parameter (T) of Optype (Where) bound to different types (tensor(int64) and tensor(float) in node (/Where),https://github.com/pytorch/pytorch/issues/123353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-04T14:40:59Z +issue,127875,[ShapeInferenceError] Attribute dilations has incorrect size,https://github.com/pytorch/pytorch/issues/127875,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-04T14:17:48Z +issue,129200,"Export model using onnx.dynamo_export has bug of ""torch._dynamo.exc.Unsupported: call_method TupleVariable() size [ConstantVariable(int)] {}""",https://github.com/pytorch/pytorch/issues/129200,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-09-04T14:09:09Z +issue,128414,How to enable XNNPACK instead of NNPACK/MKLDNN in Windows?,https://github.com/pytorch/pytorch/issues/128414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: windows, module: xnnpack",scan when time allows,2024-09-04T10:33:25Z +issue,134326,Very large memory allocation by torch.linalg.qr,https://github.com/pytorch/pytorch/issues/134326,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-09-04T08:14:31Z +issue,134901,How to calculate second derivative using PyTorch with GPU (cuda),https://github.com/pytorch/pytorch/issues/134901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cudnn, module: double backwards, module: functorch",scan when time allows,2024-09-04T01:48:21Z +issue,134977,[BUG] found a weird problem of nn.conv2d in pytorch 2.2.0,https://github.com/pytorch/pytorch/issues/134977,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: numerical-stability",scan when time allows,2024-09-04T01:45:13Z +issue,121482,torch.linalg.eigh is very slow compared to tf.linalg.eigh,https://github.com/pytorch/pytorch/issues/121482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: mkl, module: performance",scan when time allows,2024-09-03T21:29:18Z +issue,134803,enhanced debugging prints in torch.distributed,https://github.com/pytorch/pytorch/issues/134803,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-09-03T21:22:58Z +issue,108968,Unnecessary cuda synchronizations that we should remove in PyTorch,https://github.com/pytorch/pytorch/issues/108968,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2024-09-03T20:23:02Z +issue,134947,torch.hub.load will raise RuntimeError: operator torchvision::nms does not exist,https://github.com/pytorch/pytorch/issues/134947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2024-09-03T19:13:31Z +issue,134953,_rebuild_sparse_tensor,https://github.com/pytorch/pytorch/issues/134953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-09-03T19:11:14Z +issue,134996,Setting CPU affinity doesn't work for torch.softmax when using a thread,https://github.com/pytorch/pytorch/issues/134996,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2024-09-03T19:03:45Z +issue,122641,[vulkan] Vulkan backend fails creating tensor on aarch64 (Raspberry Pi 5),https://github.com/pytorch/pytorch/issues/122641,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2024-09-01T12:03:10Z +issue,122694,elastic agent kill hang process when other processes timeout,https://github.com/pytorch/pytorch/issues/122694,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2024-08-31T02:35:36Z +issue,117736,Illegal memory access for nn.Conv2d when intermediate output has size larger than 2**31,https://github.com/pytorch/pytorch/issues/117736,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda, module: edge cases",scan when time allows,2024-08-30T23:04:27Z +issue,134641,nn.Module.to(memory_format= channels_last format) failed if containing 5D parameters,https://github.com/pytorch/pytorch/issues/134641,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",scan when time allows,2024-08-30T21:23:25Z +issue,134668,"Whether tensor parallelism supports the overlap of communication calculations for gradient computation, and how to implement it",https://github.com/pytorch/pytorch/issues/134668,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-30T17:54:43Z +issue,101878,torch.quantile on MPS doesn't sort values when dim is not None,https://github.com/pytorch/pytorch/issues/101878,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2024-08-30T17:33:04Z +issue,134637,LSTM inference c++ threads block on DropoutState,https://github.com/pytorch/pytorch/issues/134637,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: multithreading, module: rnn",scan when time allows,2024-08-30T15:06:04Z +issue,134607,torch.angle cuda version not implemented for bfloat16 and half (float16),https://github.com/pytorch/pytorch/issues/134607,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-08-30T15:04:58Z +issue,134821,Add decompositions for nanmedian and median,https://github.com/pytorch/pytorch/issues/134821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: decompositions,scan when time allows,2024-08-30T14:42:31Z +issue,15738,Beta Distribution values wrong for a=b---> 0,https://github.com/pytorch/pytorch/issues/15738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-08-30T13:56:34Z +issue,120996,"Unnecessary CUDA synchronization, or missing warnings",https://github.com/pytorch/pytorch/issues/120996,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-08-30T08:22:50Z +issue,10684,make install error： [third_party/gloo/gloo/CMakeFiles/gloo.dir/all] Error 2,https://github.com/pytorch/pytorch/issues/10684,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2024-08-30T08:00:29Z +issue,119447,Cannot install or import PyTorch Nightly,https://github.com/pytorch/pytorch/issues/119447,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-08-30T06:54:43Z +issue,134755,"""Attempted to send CUDA tensor received from another process"" even though that was not the case",https://github.com/pytorch/pytorch/issues/134755,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-08-30T05:04:53Z +issue,134612,torch.func.grad uses more memory than expected,https://github.com/pytorch/pytorch/issues/134612,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch, module: memory usage",scan when time allows,2024-08-30T04:57:30Z +issue,134691,[docs] We should add a lint for bullet points and multiple lines,https://github.com/pytorch/pytorch/issues/134691,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: lint",scan when time allows,2024-08-29T19:03:24Z +issue,133761,Incorrect GPU management and deadlocks without torch.cuda.set_device,https://github.com/pytorch/pytorch/issues/133761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-29T18:42:16Z +issue,132231,torch.nn.Bilinear weights initialization reason,https://github.com/pytorch/pytorch/issues/132231,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-28T23:32:48Z +issue,117209,"torch.onnx.export the torch,nn.TransformerEncoderLayer, the dynamic_axes work not right",https://github.com/pytorch/pytorch/issues/117209,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-08-28T22:33:33Z +issue,125173,Flight Recorder Sequence IDs are insufficient,https://github.com/pytorch/pytorch/issues/125173,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-28T22:25:01Z +issue,121519,Avoid shard 0 download on all ranks,https://github.com/pytorch/pytorch/issues/121519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-28T22:24:19Z +issue,121311,In-place native functional collectives right after a custom torch op fail to codegen in inductor,https://github.com/pytorch/pytorch/issues/121311,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: distributed, oncall: pt2",scan when time allows,2024-08-28T19:07:03Z +issue,118472,[c10d] Avoid busy looping in C10D_NCCL_CHECK_TIMEOUT,https://github.com/pytorch/pytorch/issues/118472,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2024-08-28T18:14:10Z +issue,134680,Use `torch._C._stash_obj_in_tls` for global state in serialization,https://github.com/pytorch/pytorch/issues/134680,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-08-28T14:52:32Z +issue,30968,Categorical.sample too slow,https://github.com/pytorch/pytorch/issues/30968,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-08-28T14:04:56Z +issue,126523,Summary of tests in machine readable format,https://github.com/pytorch/pytorch/issues/126523,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2024-08-28T12:06:45Z +issue,116322,[RFC] Intel GPU Runtime Upstreaming for Allocator,https://github.com/pytorch/pytorch/issues/116322,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: intel,scan when time allows,2024-08-28T10:47:30Z +issue,134570,histogramdd 'Cuda' backend,https://github.com/pytorch/pytorch/issues/134570,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-08-28T09:39:35Z +issue,125182,QR Decomposition for Sparse Matrix,https://github.com/pytorch/pytorch/issues/125182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2024-08-28T09:22:41Z +issue,33296,Training got stuck due to timeout from dataloader,https://github.com/pytorch/pytorch/issues/33296,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: performance",scan when time allows,2024-08-28T07:46:46Z +issue,133849,CatArrayBatchedCopy and AllGather don't overlap during FSDP backward,https://github.com/pytorch/pytorch/issues/133849,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2024-08-28T05:05:49Z +issue,56136,"Please add a tab of ""all"" packages in `Docs` option",https://github.com/pytorch/pytorch/issues/56136,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-08-28T02:44:46Z +issue,134190,ProcessGroupNCCL::WorkNCCL only records seq_id which is not enough to be matched with comm op in Execution Trace replay,https://github.com/pytorch/pytorch/issues/134190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-28T00:41:17Z +issue,74537,ComplexHalf Coverage Tracker,https://github.com/pytorch/pytorch/issues/74537,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: half",scan when time allows,2024-08-27T23:47:04Z +issue,123532,Request: Restrict use of tags to strictly formal release versions,https://github.com/pytorch/pytorch/issues/123532,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",scan when time allows,2024-08-27T21:55:26Z +issue,112459,GRUCell batching rule for vmap,https://github.com/pytorch/pytorch/issues/112459,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-08-27T19:51:27Z +issue,131650,"Pytorch 2.4, Cuda 12.4: RuntimeError: device >= 0 && device < num_gpus INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1720538439675/work/aten/src/ATen/cuda/CUDAContext.cpp"":49, please report a bug to PyTorch. device=^A, num_gpus=^A",https://github.com/pytorch/pytorch/issues/131650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-08-27T11:30:19Z +issue,37324,Multiprocessing: model shared between processes hangs during copy.deepcopy,https://github.com/pytorch/pytorch/issues/37324,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2024-08-27T09:58:53Z +issue,134180,[export] Incorrect behavior with torch.LongTensor and torch.Tensor,https://github.com/pytorch/pytorch/issues/134180,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2024-08-27T05:17:35Z +issue,133941,"Why core aten IR doesn’t contain scalar_tensor overload version of some ops, like bitwise-like ops, remainder and so on?",https://github.com/pytorch/pytorch/issues/133941,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: core aten,scan when time allows,2024-08-27T03:28:53Z +issue,134403,RuntimeError: cuFFT error: CUFFT_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/134403,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: fft",scan when time allows,2024-08-26T18:30:35Z +issue,133985,Feat: output dtype for count_nonzero,https://github.com/pytorch/pytorch/issues/133985,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-08-26T17:57:31Z +issue,134183,storage_module && PyModule_Check(storage_module) INTERNAL ASSERT FAILED at DynamicTypes.cpp,https://github.com/pytorch/pytorch/issues/134183,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: python frontend",scan when time allows,2024-08-26T17:29:01Z +issue,134441,Cuda memory error when testing GroupedGemm on a multi-Gpu environment,https://github.com/pytorch/pytorch/issues/134441,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-08-26T17:01:17Z +issue,134439,configure ignores USE_QNNPACK=0 and still looks for third_party/QNNPACK,https://github.com/pytorch/pytorch/issues/134439,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-08-26T17:00:29Z +issue,134069,Kron with gradient for sparse tensors,https://github.com/pytorch/pytorch/issues/134069,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2024-08-26T15:51:04Z +issue,32575,Interrupting a DDP worker while using the CUDA MPS server causes CUDA to hang until reboot,https://github.com/pytorch/pytorch/issues/32575,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2024-08-26T13:37:21Z +issue,126045,A scalar with `elements` or `test_elements` argument of `isin()` doesn't work,https://github.com/pytorch/pytorch/issues/126045,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-08-26T09:39:37Z +issue,134293,[PrivateUse1 Backend] Unable to register backend to PrivateUse1 dispatch key: ModuleNotFoundError: No module named 'torch.privateuseone',https://github.com/pytorch/pytorch/issues/134293,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: privateuse1,scan when time allows,2024-08-26T03:25:02Z +issue,126972,ONNX Exporter Fails with Handling Complex Tensors,https://github.com/pytorch/pytorch/issues/126972,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-08-25T22:03:02Z +issue,42223,Implementing packbits,https://github.com/pytorch/pytorch/issues/42223,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-08-25T16:48:25Z +issue,125683,[RFC] Add Cpp Template for GEMM related ops via max-autotune for Inductor CPU,https://github.com/pytorch/pytorch/issues/125683,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",scan when time allows,2024-08-25T10:33:02Z +issue,70513,Conda Repodata.json file not found in Pytorch channel,https://github.com/pytorch/pytorch/issues/70513,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-08-25T01:58:30Z +issue,134249,Footgun: Dynamo x tensors created inside or outside of inference_mode,https://github.com/pytorch/pytorch/issues/134249,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamo,scan when time allows,2024-08-24T20:45:23Z +issue,84994,Cuda tensor is zero when passed through multiprocessing queue,https://github.com/pytorch/pytorch/issues/84994,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-08-23T20:23:28Z +issue,134051,FSDP activation CPU offload memory not cleaned up,https://github.com/pytorch/pytorch/issues/134051,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, oncall: distributed",scan when time allows,2024-08-23T19:00:37Z +issue,20248,Sparse tensors can't be used in DataLoader running many workers,https://github.com/pytorch/pytorch/issues/20248,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: sparse",scan when time allows,2024-08-23T18:27:45Z +issue,134201,Python310 trying to run GFPGAN comes: ImportError: cannot import name 'sym_float' from 'torch' (unknown location),https://github.com/pytorch/pytorch/issues/134201,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-08-23T14:55:05Z +issue,82443,Quantization issue in transformers,https://github.com/pytorch/pytorch/issues/82443,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-08-23T02:23:55Z +issue,55910,Allow variable intermediate hidden dimensions for stacked RNN/LSTM layers,https://github.com/pytorch/pytorch/issues/55910,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2024-08-22T23:44:07Z +issue,134117,Add aarch64 and aarch64 cuda version to conda,https://github.com/pytorch/pytorch/issues/134117,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries",scan when time allows,2024-08-22T22:41:01Z +issue,134244,Setting a 1D tensor as `input` and a 0D tensor as `index` and vice versa for `gather()` works against the error message,https://github.com/pytorch/pytorch/issues/134244,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: scatter & gather ops",scan when time allows,2024-08-22T19:30:58Z +issue,132412,[CI] PyTorch eager is slow in pytorch-linux-jammy-aarch64-py3.10-gcc11-inductor-benchmarks docker,https://github.com/pytorch/pytorch/issues/132412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: ci",scan when time allows,2024-08-22T04:16:48Z +issue,79715,[FX] TypeError when tracing cat taking split's output as input,https://github.com/pytorch/pytorch/issues/79715,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-08-22T03:58:00Z +issue,100656,"On UMA systems, pytorch fails to reserve memory exceeding the initial memory size",https://github.com/pytorch/pytorch/issues/100656,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: rocm",scan when time allows,2024-08-22T02:32:52Z +issue,134079,operator benchmark diag test cuda problem,https://github.com/pytorch/pytorch/issues/134079,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: benchmark,scan when time allows,2024-08-21T23:03:26Z +issue,134098,libtorch reasoned that the crnn model was out of speed the first few times,https://github.com/pytorch/pytorch/issues/134098,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: jit",scan when time allows,2024-08-21T22:57:40Z +issue,133380,AttributeError: 'NoneType' object has no attribute 'is_failed',https://github.com/pytorch/pytorch/issues/133380,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-21T16:21:41Z +issue,134038,vmap tests aren't erroring on fallback usage,https://github.com/pytorch/pytorch/issues/134038,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2024-08-20T22:28:10Z +issue,133998,`torch.quantile`: `nan` instead of `inf`,https://github.com/pytorch/pytorch/issues/133998,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-08-20T21:39:30Z +issue,133754,sample of Binomial should return int instead of float,https://github.com/pytorch/pytorch/issues/133754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-08-20T18:17:31Z +issue,133818,"Warning when initialize `torch.nn.Linear(0, d)`",https://github.com/pytorch/pytorch/issues/133818,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-20T18:15:16Z +issue,128046,Future LRScheduler implementation with Tensor support,https://github.com/pytorch/pytorch/issues/128046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2024-08-20T15:07:43Z +issue,86989,Error: unknown architecture `armv7-a;' and Error: selected processor does not support `command' in ARM mode,https://github.com/pytorch/pytorch/issues/86989,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, oncall: quantization",scan when time allows,2024-08-20T11:38:24Z +issue,131691,TCP socket keeps trying IPv6 address and doesn't try IPv4,https://github.com/pytorch/pytorch/issues/131691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-20T01:18:56Z +issue,133383,[Inductor] Remove dead code from triton kernels,https://github.com/pytorch/pytorch/issues/133383,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-08-19T19:18:27Z +issue,133210,IPC feature supports third-party devices,https://github.com/pytorch/pytorch/issues/133210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: privateuse1, module: third_party",scan when time allows,2024-08-19T17:24:38Z +issue,89051,caffe2_interface_library CMake macro prevents linking to LibTorch as a transitive dependency,https://github.com/pytorch/pytorch/issues/89051,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-08-19T08:09:08Z +issue,51836,TypeError: can't pickle _thread.lock objects when using spawn / forkserver,https://github.com/pytorch/pytorch/issues/51836,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pickle,scan when time allows,2024-08-18T09:06:45Z +issue,45414,Module __call__ typing,https://github.com/pytorch/pytorch/issues/45414,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2024-08-17T21:09:17Z +issue,133522,"`init` of `PReLU()` works with `int`, `complex` and `bool` against what the doc says.",https://github.com/pytorch/pytorch/issues/133522,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-17T18:28:49Z +issue,108977,[feature request] Provide some sparse eigen solver(s) for PyTorch (maybe via `ARPACK` as in scipy) + SPD sparse / laplace linear system solver - maybe NVidia AMGx library?,https://github.com/pytorch/pytorch/issues/108977,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2024-08-17T11:54:54Z +issue,132428,"MultiProcTest Exception ""None is not -6""",https://github.com/pytorch/pytorch/issues/132428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-16T21:16:29Z +issue,133453,[randperm] Add argument to disallow identiy mappings,https://github.com/pytorch/pytorch/issues/133453,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2024-08-16T21:01:56Z +issue,133564,torch.cond example fails in PyTorch2.4,https://github.com/pytorch/pytorch/issues/133564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-08-16T21:01:27Z +issue,130238,how to simplify torch.fx like using onnxsim?,https://github.com/pytorch/pytorch/issues/130238,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-08-16T13:40:42Z +issue,133527,Using `PReLU()` with a `complex` tensor and without `dtype=torch.complex64` should return a simple error message,https://github.com/pytorch/pytorch/issues/133527,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-16T13:31:30Z +issue,133535,"When using a wrong value for `PReLU()`, a PReLU's error message should be returned directly instead of `empty()` error.",https://github.com/pytorch/pytorch/issues/133535,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-16T13:31:13Z +issue,53441,"Conjugate gradient Descent, and Linear operator are not present in pytorch.",https://github.com/pytorch/pytorch/issues/53441,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: lazy, module: linear algebra, module: sparse",scan when time allows,2024-08-15T23:32:28Z +issue,26288,Inplace and out arguments for BatchNorm (and other norm layers: InstanceNorm / LayerNorm / GroupNorm ...),https://github.com/pytorch/pytorch/issues/26288,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization, module: performance",scan when time allows,2024-08-15T21:12:42Z +issue,132596,MPS backend leaks memory when input sizes vary,https://github.com/pytorch/pytorch/issues/132596,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: mps",scan when time allows,2024-08-15T20:24:46Z +issue,123761,Support for no-frills FP8 matmuls,https://github.com/pytorch/pytorch/issues/123761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2024-08-15T13:33:17Z +issue,129803,3D matrix support in _scaled_mm,https://github.com/pytorch/pytorch/issues/129803,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2024-08-15T13:30:36Z +issue,133224,"When using parallelize_module to do tp parallel and pp parallel mixing, the SendRecv time is too long",https://github.com/pytorch/pytorch/issues/133224,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-14T12:54:14Z +issue,123072,Saving & Loading Model Across [In-tree and out-of-tree] Devices,https://github.com/pytorch/pytorch/issues/123072,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: privateuse1, module: serialization",scan when time allows,2024-08-14T06:00:03Z +issue,120814,torch.compile() and optimizer UX issues,https://github.com/pytorch/pytorch/issues/120814,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: pt2 optimizer, oncall: pt2",scan when time allows,2024-08-13T19:37:29Z +issue,133100,MPS clamp_min on older MacOS,https://github.com/pytorch/pytorch/issues/133100,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2024-08-13T17:50:41Z +issue,133262,export error when tracing tensor subclass tensors: `'FakeTensor' object has no attribute 'get_plain'`,https://github.com/pytorch/pytorch/issues/133262,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2024-08-13T17:13:38Z +issue,119474,Extend functionality of torch.nn.functional.fold to apply to sequences (1D tensors + batch & channels) not just images (2D tensors + batch & channels),https://github.com/pytorch/pytorch/issues/119474,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-13T14:11:08Z +issue,92701,set_default_device/torch.device has performance impact for non-factory functions,https://github.com/pytorch/pytorch/issues/92701,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: performance",scan when time allows,2024-08-13T13:15:03Z +issue,106971,Found two conflicting CUDA installs,https://github.com/pytorch/pytorch/issues/106971,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-08-13T09:39:08Z +issue,133006,Precision Differences in Using `dropout` and `linear`,https://github.com/pytorch/pytorch/issues/133006,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2024-08-13T08:30:30Z +issue,133183,convolution_overrideable not implemented when executing VAEEncode in ComfyUI,https://github.com/pytorch/pytorch/issues/133183,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: privateuse1,scan when time allows,2024-08-12T20:51:31Z +issue,133138,Layer normalization for Nested Tensors does not return the correct tensor types,https://github.com/pytorch/pytorch/issues/133138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-08-12T20:50:20Z +issue,133114,Conv1D: `CUDNN_BACKEND_OPERATION: cudnnFinalize Failed cudnn_status: CUDNN_STATUS_BAD_PARAM`,https://github.com/pytorch/pytorch/issues/133114,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn",scan when time allows,2024-08-12T20:49:02Z +issue,132696,Hanging at waitForFutureOrTimeout when actively destroying ProcessGroups,https://github.com/pytorch/pytorch/issues/132696,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-12T18:52:44Z +issue,133119,MultiProcContinuousTest doesn't work with pytest/buck,https://github.com/pytorch/pytorch/issues/133119,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-12T17:52:33Z +issue,132747,Why is kvcache transfer slow using peer-to-peer communication (two GPUs on one machine).,https://github.com/pytorch/pytorch/issues/132747,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2024-08-12T17:40:37Z +issue,9873,"Pytorch is slow when only using CPU, and cannot utilize multicore of CPU",https://github.com/pytorch/pytorch/issues/9873,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multithreading, module: performance",scan when time allows,2024-08-12T09:11:30Z +issue,40581,Libtorch C++ multiple GPU performance slower than single GPU,https://github.com/pytorch/pytorch/issues/40581,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: data parallel, module: performance",scan when time allows,2024-08-12T06:33:30Z +issue,132693,INTERNAL ASSERT FAILED in BatchLinearAlgebra.cpp,https://github.com/pytorch/pytorch/issues/132693,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: mkl",scan when time allows,2024-08-10T17:20:53Z +issue,63558,universal binaries for libtorch on mac (x86_64+arm),https://github.com/pytorch/pytorch/issues/63558,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: macos",scan when time allows,2024-08-10T13:37:11Z +issue,3898,Sparse matrices in dataloader error,https://github.com/pytorch/pytorch/issues/3898,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-08-09T23:32:29Z +issue,60294,Add a NumPy-like `pad` function,https://github.com/pytorch/pytorch/issues/60294,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: padding",scan when time allows,2024-08-09T22:00:15Z +issue,57911,Circular padding in Convolution layers should not only be wrap for once.,https://github.com/pytorch/pytorch/issues/57911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn, module: padding",scan when time allows,2024-08-09T21:59:23Z +issue,132991,"_pickle.UnpicklingError: invalid load key, '\x00'. during torch load",https://github.com/pytorch/pytorch/issues/132991,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-08-09T17:47:22Z +issue,133084,Addition of Visual Information Fidelity loss,https://github.com/pytorch/pytorch/issues/133084,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2024-08-09T12:33:19Z +issue,68616,"`torch.flip`, `torch.roll`, `torch.tile` has arguments `dims=` for both dim/multidim while other functions have argument `dim=` for the same usage",https://github.com/pytorch/pytorch/issues/68616,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: python frontend",scan when time allows,2024-08-08T20:46:38Z +issue,132212,NotImplementedError in some APIs with Nested Jagged Tensors,https://github.com/pytorch/pytorch/issues/132212,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-08-08T20:23:26Z +issue,119072,Importing PyTorch 2.2 fails with undefined symbol error: ncclCommRegister,https://github.com/pytorch/pytorch/issues/119072,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: nccl, module: third_party",scan when time allows,2024-08-08T15:39:42Z +issue,132483,Non-equivalent behaviour of deprecated set_default_tensor_type and set_default_dtype/set_default_device,https://github.com/pytorch/pytorch/issues/132483,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-08-08T12:18:23Z +issue,61244,error: invalid conversion in vec256_bfloat16.h,https://github.com/pytorch/pytorch/issues/61244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-08-08T10:38:04Z +issue,132951,torchelastic multiprocessing start_processes documentation out of date,https://github.com/pytorch/pytorch/issues/132951,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: elastic,scan when time allows,2024-08-08T01:06:20Z +issue,132919,Add a warning for `drop_last` argument in `DataLoader`,https://github.com/pytorch/pytorch/issues/132919,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-08-08T01:04:12Z +issue,128779,The name of the function `nn.L1Loss()` should be `nn.MAE()` or the name of the function `MSELoss()` should be `nn.L2Loss()`,https://github.com/pytorch/pytorch/issues/128779,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-07T20:55:00Z +issue,107896,[feature request] [ux proposal] Min-max linear normalization to be supported in F.normalize (or in a new function),https://github.com/pytorch/pytorch/issues/107896,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-07T20:52:47Z +issue,128064,"Improve debugability of warnings/errors ""Triggered internally at""",https://github.com/pytorch/pytorch/issues/128064,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: error checking",scan when time allows,2024-08-07T20:37:30Z +issue,129118,interpolate nearest get values zero when outputs over 4G elements,https://github.com/pytorch/pytorch/issues/129118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: correctness (silent), module: cuda, module: interpolation",scan when time allows,2024-08-07T19:52:50Z +issue,132028,torch.sspaddmm: self.is_sparse() INTERNAL ASSERT FAILED on torch 2.5.0.dev20240708+cu121,https://github.com/pytorch/pytorch/issues/132028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-08-07T15:21:34Z +issue,132759,Delete ProxyTorchDispatchMode.enable_tracing,https://github.com/pytorch/pytorch/issues/132759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: proxytensor,scan when time allows,2024-08-07T13:37:30Z +issue,124351,BUG Report using `backend:cudaMallocAsync`,https://github.com/pytorch/pytorch/issues/124351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-08-07T00:49:02Z +issue,129122,Questions about CVE-2024-31583 and CVE-2024-31580,https://github.com/pytorch/pytorch/issues/129122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-08-06T12:51:25Z +issue,132612,bitwise_xor seems to be not optimized,https://github.com/pytorch/pytorch/issues/132612,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: python frontend",scan when time allows,2024-08-06T08:51:21Z +issue,132614,BUG: `torch.special.gammainc`: returns finite results with NaN input,https://github.com/pytorch/pytorch/issues/132614,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: special",scan when time allows,2024-08-05T20:21:24Z +issue,132600,Inquiry about Test Error after Static Compilation,https://github.com/pytorch/pytorch/issues/132600,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: static linking",scan when time allows,2024-08-05T20:20:26Z +issue,131463,Cannot subclass _TensorBase directly error,https://github.com/pytorch/pytorch/issues/131463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-08-05T20:15:23Z +issue,132608,topk with mask - efficent,https://github.com/pytorch/pytorch/issues/132608,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: masked operators, module: performance, module: sorting and selection",scan when time allows,2024-08-05T19:17:30Z +issue,132599,[BUG+FIX] LibTorch silently fails to find cublas_v2.h for NVHPC,https://github.com/pytorch/pytorch/issues/132599,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cublas, module: cuda",scan when time allows,2024-08-05T18:27:00Z +issue,132602,"Is torch.compile imprecise, as this 4-year-old comment in amp.grad_scaler claims?",https://github.com/pytorch/pytorch/issues/132602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: numerical-stability",scan when time allows,2024-08-05T15:56:26Z +issue,132534,triton import inside function doesn't work,https://github.com/pytorch/pytorch/issues/132534,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: custom-operators,scan when time allows,2024-08-05T15:55:53Z +issue,132649,nn.Parallel,https://github.com/pytorch/pytorch/issues/132649,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-08-05T15:50:48Z +issue,113743,The cuda batched GEMM has a poor performance for bigger batch size with smaller matrix size,https://github.com/pytorch/pytorch/issues/113743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2024-08-05T10:53:12Z +issue,127391,Unable to quantize resnet50 model with post training static quantization,https://github.com/pytorch/pytorch/issues/127391,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-08-05T06:43:58Z +issue,91566,Build Error: OpenMP library could not be found. Proceeding might lead to highly sub-optimal performance.,https://github.com/pytorch/pytorch/issues/91566,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn, module: third_party",scan when time allows,2024-08-04T08:53:19Z +issue,74454,torch has no attribute sparse_csr_tensor,https://github.com/pytorch/pytorch/issues/74454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-08-04T08:25:51Z +issue,131307,allow tuples as keys for ModuleDict,https://github.com/pytorch/pytorch/issues/131307,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-04T06:56:55Z +issue,105728,Compile error PyTorch 2.0.1 / GCC 13.1.0,https://github.com/pytorch/pytorch/issues/105728,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-08-02T19:09:56Z +issue,132485,lazy_clone does not work for pinned memory,https://github.com/pytorch/pytorch/issues/132485,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: copy on write, module: cuda",scan when time allows,2024-08-02T14:27:30Z +issue,73829,"Run libtorch examples, export error ""undefined reference to xxx""",https://github.com/pytorch/pytorch/issues/73829,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: binaries, module: build",scan when time allows,2024-08-02T14:04:00Z +issue,132026,"Incompatibility with Numpy,JAX and MLX for int and bool type inputs",https://github.com/pytorch/pytorch/issues/132026,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python frontend",scan when time allows,2024-08-02T08:23:07Z +issue,25888,Forward/backward hooks for C++ torch::nn modules,https://github.com/pytorch/pytorch/issues/25888,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpp",scan when time allows,2024-08-02T08:04:10Z +issue,132130,BuildExtension breaks build because of absolute path,https://github.com/pytorch/pytorch/issues/132130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2024-08-02T07:42:04Z +issue,108158,TORCHELASTIC_RESTART_COUNT doesn't seem to be broadcasted to all worker,https://github.com/pytorch/pytorch/issues/108158,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: r2p",scan when time allows,2024-08-01T18:23:26Z +issue,125389,Support for Kolmogorov-Arnold Networks (KANs),https://github.com/pytorch/pytorch/issues/125389,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-08-01T11:45:32Z +issue,46809,Including AdaBound in the list of Optimizers.,https://github.com/pytorch/pytorch/issues/46809,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-08-01T09:18:59Z +issue,132041,`torch.nn.functional.max_unpool2d`'s behavior is different on cpu and gpu on torch 2.5.0.dev20240708+cu121,https://github.com/pytorch/pytorch/issues/132041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: python frontend",scan when time allows,2024-07-31T20:57:42Z +issue,132300,CUDA Memory Explosion -- Conv3D,https://github.com/pytorch/pytorch/issues/132300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2024-07-31T20:30:20Z +issue,126856,TorchDynamo ONNX Export does not work as expected with masking (ScatterElements),https://github.com/pytorch/pytorch/issues/126856,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-07-31T17:01:46Z +issue,131688,Drop Connect Layer,https://github.com/pytorch/pytorch/issues/131688,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-07-31T15:20:31Z +issue,132045,`torch.scatter`'s behavior is different on cpu and gpu on torch 2.5.0.dev20240708+cu121,https://github.com/pytorch/pytorch/issues/132045,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: python frontend",scan when time allows,2024-07-30T22:05:15Z +issue,131753,PyTorch complains ```A leaf node was passed to _will_engine_execute_node``` when calculating the gradient of a model warpped by ```FlopCounterMode```.,https://github.com/pytorch/pytorch/issues/131753,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: autograd",scan when time allows,2024-07-30T15:46:11Z +issue,70529,BatchNorm on variable-length sequences or batches,https://github.com/pytorch/pytorch/issues/70529,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-30T07:53:47Z +issue,119123,"`DataLoader` generic argument is dropped in `__iter__`, making it non type-safe",https://github.com/pytorch/pytorch/issues/119123,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: typing",scan when time allows,2024-07-29T23:36:01Z +issue,129459,Add pmap to torch.func for Parallel Function Execution,https://github.com/pytorch/pytorch/issues/129459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, oncall: distributed",scan when time allows,2024-07-29T20:06:51Z +issue,124990,[DTensor] Sharding strategy not implemented error should be thrown earlier,https://github.com/pytorch/pytorch/issues/124990,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2024-07-29T19:48:07Z +issue,131769,FlopCounterMode requires `tabulate` dependency,https://github.com/pytorch/pytorch/issues/131769,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-07-29T19:06:00Z +issue,132034,"Implementation of a Q-RNN layer nn.QRNN(filters,kernel,stride,n_cnn,)",https://github.com/pytorch/pytorch/issues/132034,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2024-07-29T18:59:21Z +issue,132047,Inconsistent behavior of gumbel_softmax on cpu vs. gpu,https://github.com/pytorch/pytorch/issues/132047,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: python frontend",scan when time allows,2024-07-29T16:49:48Z +issue,132052,Make C++ torch::autograd::Function support functorch,https://github.com/pytorch/pytorch/issues/132052,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-07-29T16:27:01Z +issue,132013,failed to build v2.3.0 docker image,https://github.com/pytorch/pytorch/issues/132013,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docker",scan when time allows,2024-07-29T16:23:24Z +issue,132024,Refactor serialization IO infrastructure and support HDFS/HTTP,https://github.com/pytorch/pytorch/issues/132024,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-07-29T14:52:29Z +issue,131931,Start processes in _MultiProcessingDataLoaderIter on a ThreadPool,https://github.com/pytorch/pytorch/issues/131931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: performance",scan when time allows,2024-07-29T04:31:04Z +issue,131967,Layer normalization on Nested Tensor ragged dimension fails when `lengths is not None`,https://github.com/pytorch/pytorch/issues/131967,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-07-29T04:16:45Z +issue,111806,Revisit security implications of #31875,https://github.com/pytorch/pytorch/issues/111806,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-29T00:39:06Z +issue,105982,"vmap and rnn/lstm ""accessing '.data' under vmap transform is not allowed""",https://github.com/pytorch/pytorch/issues/105982,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-07-28T13:52:30Z +issue,32167,Support in-place pinning of memory,https://github.com/pytorch/pytorch/issues/32167,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-07-27T19:05:15Z +issue,131962,Distributed tests failing on Amazon2023 AMI,https://github.com/pytorch/pytorch/issues/131962,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-26T22:45:40Z +issue,39357,Enable scripting modules with recursive functions,https://github.com/pytorch/pytorch/issues/39357,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-07-26T10:00:36Z +issue,115819,Allow `input` tensor to have arbitrary tailing extra dimensions than `index` tensor in `torch.gather()` / `torch.scatter()`,https://github.com/pytorch/pytorch/issues/115819,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2024-07-26T02:53:02Z +issue,123555,torch.onnx.errors.UnsupportedOperatorError: Exporting the operator 'quantized::linear_dynamic' to ONNX opset version 17 is not supported. Please feel free to request support or submit a pull request on PyTorch GitHub: https://github.com/pytorch/pytorch/issues,https://github.com/pytorch/pytorch/issues/123555,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-07-25T17:17:05Z +issue,131772,Default argument removal for python dispatch fails for pin_memory on factory functions,https://github.com/pytorch/pytorch/issues/131772,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: library",scan when time allows,2024-07-25T16:25:29Z +issue,131755,CUDA error: an illegal memory access was encountered when use TORCH_NCCL_NAN_CHECK.,https://github.com/pytorch/pytorch/issues/131755,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2024-07-25T15:34:48Z +issue,131736,MPS gives incorrect result when torch.nn.functional.softplus follows moveaxis,https://github.com/pytorch/pytorch/issues/131736,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2024-07-25T14:21:20Z +issue,74935,[RFC]A suggestion of channels last memory format implementation for 3D tensor,https://github.com/pytorch/pytorch/issues/74935,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2024-07-25T13:13:52Z +issue,131695,Keep pytorch/pytorch-canary's main branch in sync with pytorch/pytorch's viable strict,https://github.com/pytorch/pytorch/issues/131695,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-07-25T11:52:53Z +issue,88968,M1 runner i-090e1df32b6f48a20 run out of disk space,https://github.com/pytorch/pytorch/issues/88968,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: ci, module: macos",scan when time allows,2024-07-25T11:28:06Z +issue,97188,Building LibTorch on Ubuntu with Mac M1,https://github.com/pytorch/pytorch/issues/97188,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: macos, module: torchbind",scan when time allows,2024-07-25T11:27:54Z +issue,117094,LBFGS MPS test_can_load_older_state_dict fails on older MacOS versions,https://github.com/pytorch/pytorch/issues/117094,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, module: mps, module: optimizer, module: tests",scan when time allows,2024-07-25T07:00:04Z +issue,36650,state_dict and load_state_dict methods for DataLoader and Sampler to continue training at specific epoch and batch,https://github.com/pytorch/pytorch/issues/36650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-07-25T06:56:11Z +issue,114962,`CUDAExtension` no longer works with `ccache`,https://github.com/pytorch/pytorch/issues/114962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda, module: third_party",scan when time allows,2024-07-25T06:25:20Z +issue,18173,[docs] How to achieve high-order derivation in my .cpp?,https://github.com/pytorch/pytorch/issues/18173,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpp, module: docs",scan when time allows,2024-07-24T20:52:11Z +issue,19453,How to load PyTorch model with LSTM using C++ api,https://github.com/pytorch/pytorch/issues/19453,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-07-24T20:52:00Z +issue,7457,[feature request] torch.nn.DataParallel should work nicely both for cpu and gpu devices,https://github.com/pytorch/pytorch/issues/7457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2024-07-24T20:49:48Z +issue,34375,.detach() behaves differently for dense tensors vs sparse tensors,https://github.com/pytorch/pytorch/issues/34375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-07-24T20:48:14Z +issue,17984,"""unknown builtin op"" error with static library",https://github.com/pytorch/pytorch/issues/17984,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-07-24T20:46:31Z +issue,22122,ASSERT FAILED at /pytorch/torch/csrc/autograd/generated/VariableType_0.cpp:12721,https://github.com/pytorch/pytorch/issues/22122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: assert failure,scan when time allows,2024-07-24T20:46:18Z +issue,131656,Remove CTCLoss limitation of maximum target sequence length 256 (to satisify CuDNN),https://github.com/pytorch/pytorch/issues/131656,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2024-07-24T18:41:30Z +issue,131655,tensor.share_memory_ causes temporarily increased memory consumption,https://github.com/pytorch/pytorch/issues/131655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2024-07-24T18:40:56Z +issue,131631,DataLoader worker exited unexpectedly,https://github.com/pytorch/pytorch/issues/131631,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-07-24T18:40:05Z +issue,101185,Pure virtual function call exception on Python interpreter exit when using debug wheel,https://github.com/pytorch/pytorch/issues/101185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-07-24T04:35:12Z +issue,81856,Compatibility with newest MKL,https://github.com/pytorch/pytorch/issues/81856,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkl",scan when time allows,2024-07-24T01:28:01Z +issue,131274,The empty 0D or more D tensor with exceeding `dim`(`dim=100` or `dim=-100`) for `cummax()` works,https://github.com/pytorch/pytorch/issues/131274,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-23T22:13:26Z +issue,117705,Add GPU runners for Dynamo shards,https://github.com/pytorch/pytorch/issues/117705,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-07-23T21:59:48Z +issue,113854,Timer: Add Standard Deviation for Single Measurement,https://github.com/pytorch/pytorch/issues/113854,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: benchmark,scan when time allows,2024-07-23T21:45:54Z +issue,131491,`torch._assert_async` ignores `assert_msg`,https://github.com/pytorch/pytorch/issues/131491,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",scan when time allows,2024-07-23T21:45:03Z +issue,30723,Distributed Package asynchronous send/receive not working as expected (Gloo),https://github.com/pytorch/pytorch/issues/30723,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-23T21:34:24Z +issue,71203,Keys of a `ModuleDict` cannot have the same name as existing `ModuleDict` class attributes.,https://github.com/pytorch/pytorch/issues/71203,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-07-23T20:54:26Z +issue,101407,Delete old vmap prototype,https://github.com/pytorch/pytorch/issues/101407,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2024-07-23T15:54:09Z +issue,131273,The empty 0D or more D tensor with exceeding `dim`(`dim=100` or `dim=-100`) for `cummin()` works,https://github.com/pytorch/pytorch/issues/131273,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking",scan when time allows,2024-07-23T06:38:46Z +issue,131318,"RuntimeError: INTERNAL ASSERT FAILED at ""../aten/src/ATen/native/sparse/SparseCsrTensorMath.cpp"":1306",https://github.com/pytorch/pytorch/issues/131318,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-07-23T04:21:02Z +issue,131432,Remove all # mypy: ignore-errors,https://github.com/pytorch/pytorch/issues/131432,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2024-07-23T04:12:44Z +issue,131349,[ONNX] opset 11 aten::index_put generating invalid add ops with None args,https://github.com/pytorch/pytorch/issues/131349,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-07-23T00:34:31Z +issue,131357,Follow ups from Increase (4x+) in build times #130744,https://github.com/pytorch/pytorch/issues/131357,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-07-22T21:28:54Z +issue,130807,Generalizing cross product to hodge star operator,https://github.com/pytorch/pytorch/issues/130807,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-07-22T21:17:50Z +issue,130932,Error using torch.func transforms with DDP (broadcast_buffers=True),https://github.com/pytorch/pytorch/issues/130932,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: functorch, oncall: distributed",scan when time allows,2024-07-22T20:47:57Z +issue,131110,Proposal for CUDA-Accelerated Dynamic Time Warping (DTW) Implementation in PyTorch,https://github.com/pytorch/pytorch/issues/131110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-07-22T16:49:10Z +issue,74442,kron has unnecessary and undocumented dependence on memory layout,https://github.com/pytorch/pytorch/issues/74442,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2024-07-22T15:46:34Z +issue,33864,Memory leak in embedding layer and LSTM,https://github.com/pytorch/pytorch/issues/33864,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2024-07-22T10:58:56Z +issue,122957,How to export torch.optim.LBFGS using torch.onnx.export,https://github.com/pytorch/pytorch/issues/122957,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: onnx, module: optimizer",scan when time allows,2024-07-22T09:48:29Z +issue,114723,[RFC] Intel GPU Upstreaming,https://github.com/pytorch/pytorch/issues/114723,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-07-22T07:36:04Z +issue,131113,pytorch dataloader has function that raises error,https://github.com/pytorch/pytorch/issues/131113,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2024-07-19T20:53:03Z +issue,131062,Add some more testing around checkpoint over compile,https://github.com/pytorch/pytorch/issues/131062,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2024-07-19T20:51:18Z +issue,75166,Disable Python mode (torch dispatch mode) inside of mode-induced __torch_dispatch__call,https://github.com/pytorch/pytorch/issues/75166,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2024-07-19T18:01:22Z +issue,131213,Clean up coverage_ignore_functions,https://github.com/pytorch/pytorch/issues/131213,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docs",scan when time allows,2024-07-19T17:53:10Z +issue,131148,"The empty 1D or more D tensor for `median()/nanmedian()` with the deepest `dim` gets errors, not getting NaNs",https://github.com/pytorch/pytorch/issues/131148,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: empty tensor, module: error checking",scan when time allows,2024-07-19T15:37:06Z +issue,44544,torch.distributed server back-connects to clients on random ports leading to firewall problems,https://github.com/pytorch/pytorch/issues/44544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-19T15:25:32Z +issue,130610,dataloader hangs with fork and pin memory=True,https://github.com/pytorch/pytorch/issues/130610,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-07-19T06:45:50Z +issue,131011,xeon.run_cpu will not use numactl when whithin a docker container run with --cpuset-cpus non-0 numa node,https://github.com/pytorch/pytorch/issues/131011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2024-07-19T00:58:47Z +issue,128204,Documentation for DDP-related environment variables,https://github.com/pytorch/pytorch/issues/128204,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-18T21:16:16Z +issue,131054,Shape-aware type hinting,https://github.com/pytorch/pytorch/issues/131054,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2024-07-18T19:45:00Z +issue,124395,Provide TORCH_LIBRARY m.def schema types documentation,https://github.com/pytorch/pytorch/issues/124395,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: library",scan when time allows,2024-07-18T17:06:29Z +issue,121641,torch.isin on GPU vs numpy isin: perf way slower than expected,https://github.com/pytorch/pytorch/issues/121641,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2024-07-18T15:38:03Z +issue,58828,[Feature Pitch] Fast extremal eigensolvers,https://github.com/pytorch/pytorch/issues/58828,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2024-07-18T13:10:38Z +issue,129553,RVV support in PyTorch,https://github.com/pytorch/pytorch/issues/129553,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2024-07-18T06:59:11Z +issue,130792,`with torch.device()` modifies the `as_tensor` intended behavior,https://github.com/pytorch/pytorch/issues/130792,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-18T00:53:46Z +issue,130985,Runtime error when running huggingface pretrained model with torch-xla,https://github.com/pytorch/pytorch/issues/130985,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xla,scan when time allows,2024-07-18T00:30:27Z +issue,130847,Where does torch check for autocast op availability?,https://github.com/pytorch/pytorch/issues/130847,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: docs",scan when time allows,2024-07-18T00:30:08Z +issue,130950,Improve detail / Create tutorial for torch.distributions,https://github.com/pytorch/pytorch/issues/130950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs",scan when time allows,2024-07-18T00:25:19Z +issue,95460,torch.profiler.tensorboard_trace_handler Generates an incorrect JSON file,https://github.com/pytorch/pytorch/issues/95460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",scan when time allows,2024-07-17T23:13:38Z +issue,130512,"The argument of `take()`, `take_along_dim()` and `gather()` should be `index` or `indices` for consistency.",https://github.com/pytorch/pytorch/issues/130512,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-07-17T21:58:57Z +issue,104697,LSTM built-in dropout not reproducible on GPU,https://github.com/pytorch/pytorch/issues/104697,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: determinism, module: nn, module: random",scan when time allows,2024-07-17T15:08:41Z +issue,128020,Make `torch.autograd.Function` support `vmap`,https://github.com/pytorch/pytorch/issues/128020,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2024-07-17T14:57:25Z +issue,91182,onednn(mkldnn) backend support for quantized operators,https://github.com/pytorch/pytorch/issues/91182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, oncall: quantization",scan when time allows,2024-07-17T13:27:29Z +issue,37250,Inconsistency between GPU memory usage in torch.cuda.memory_summary and nvidia-smi,https://github.com/pytorch/pytorch/issues/37250,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2024-07-17T12:52:40Z +issue,50275,torch.transpose is divergent from np.transpose,https://github.com/pytorch/pytorch/issues/50275,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy",scan when time allows,2024-07-17T08:47:34Z +issue,130863,CMake cannot detect non-default fp16 installation path,https://github.com/pytorch/pytorch/issues/130863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-07-17T07:06:12Z +issue,67011,Importing numpy interacts with `tensor.sum` perf,https://github.com/pytorch/pytorch/issues/67011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: performance",scan when time allows,2024-07-17T06:38:07Z +issue,130693,The doc of `vstack()` should say `tuple` or `list` of tensors for `tensors` argument,https://github.com/pytorch/pytorch/issues/130693,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-07-16T20:36:32Z +issue,130621,[Bug Report] Choosing the network interface to use,https://github.com/pytorch/pytorch/issues/130621,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-16T02:31:47Z +issue,130768,Migrate gradcheck with check_batched_grad=True to use the new torch.vmap,https://github.com/pytorch/pytorch/issues/130768,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vmap,scan when time allows,2024-07-16T00:21:10Z +issue,130739,Ineffective Buffer Size Adjustment in Multi-Node NCCL Communication,https://github.com/pytorch/pytorch/issues/130739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nccl,scan when time allows,2024-07-16T00:19:19Z +issue,130701,fx.wrap() doesn't really work for things in torch/*,https://github.com/pytorch/pytorch/issues/130701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-07-15T23:45:22Z +issue,130656,[torchelastic] torchrun with static backend hangs when used with port '0',https://github.com/pytorch/pytorch/issues/130656,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: elastic,scan when time allows,2024-07-15T23:41:32Z +issue,130342,Rescale to zero torch.nn.CrossEntropyLoss with label_smoothing > 0,https://github.com/pytorch/pytorch/issues/130342,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-07-15T21:05:20Z +issue,113948,Unknown CUDA Architecture Name 9.0a in CUDA_SELECT_NVCC_ARCH_FLAGS (compiling from source),https://github.com/pytorch/pytorch/issues/113948,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2024-07-15T19:29:06Z +issue,129477,The ProcessGroupNCCL is not being destructed,https://github.com/pytorch/pytorch/issues/129477,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-15T19:12:56Z +issue,130636,`dims` of `permute()` without `tuple` or with `list` works agaist the doc,https://github.com/pytorch/pytorch/issues/130636,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-13T01:44:43Z +issue,130430,Error occurred when executing KolorsSampler:,https://github.com/pytorch/pytorch/issues/130430,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2024-07-12T23:29:57Z +issue,130608,Setting `shape` and `dtype` without `dtype=` to `view()` works against error message but an `int` tensor is returned instead of a `float` tensor,https://github.com/pytorch/pytorch/issues/130608,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: python frontend",scan when time allows,2024-07-12T17:47:51Z +issue,130524,Unable to compile pytorch with conda python3.8 and cuda 11.4 in Nvidia AGX Jetson Orin aarch64,https://github.com/pytorch/pytorch/issues/130524,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: cuda",scan when time allows,2024-07-12T17:35:51Z +issue,130529,Support tuple of `int`s in `torch.unsqueeze`,https://github.com/pytorch/pytorch/issues/130529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-07-12T16:07:38Z +issue,130514,C++ API Does Not Expose Lazy NVRTC,https://github.com/pytorch/pytorch/issues/130514,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cuda",scan when time allows,2024-07-12T16:04:22Z +issue,130525,[Feature request] Extend index_reduce to allow `index` to be of type List[List[int]],https://github.com/pytorch/pytorch/issues/130525,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: python frontend",scan when time allows,2024-07-12T16:03:54Z +issue,130538,"nested_tensor LayerNorm fails with shape (*,*,j1,*) (but works with (*,j1,*,*))",https://github.com/pytorch/pytorch/issues/130538,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-07-12T16:02:36Z +issue,130448,`n` argument of `randperm()` with a 0D or more D tensor works against the doc,https://github.com/pytorch/pytorch/issues/130448,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: python frontend",scan when time allows,2024-07-12T15:52:39Z +issue,128497,"[bug]When the variable in the dataset is of type cuda, calling the dataloader will result in 0",https://github.com/pytorch/pytorch/issues/128497,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2024-07-12T10:03:12Z +issue,129356,Does it still make sense to prevent saving multiple tensors that have a shared storage but different dtypes?,https://github.com/pytorch/pytorch/issues/129356,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-07-12T07:43:32Z +issue,113053,[CUDA-12.2] cuSPARSE deprecated support for sparse BSR,https://github.com/pytorch/pytorch/issues/113053,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2024-07-11T22:00:30Z +issue,30702,[FR] multidim squeeze and flatten,https://github.com/pytorch/pytorch/issues/30702,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2024-07-11T20:53:42Z +issue,130055,[pipelining] Non-first stage only takes tensors,https://github.com/pytorch/pytorch/issues/130055,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pipelining,scan when time allows,2024-07-11T16:58:24Z +issue,130464,[dynamo] Investigate unexpected success dynamo-wrapped tests from https://github.com/pytorch/pytorch/pull/129185,https://github.com/pytorch/pytorch/issues/130464,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-07-11T12:23:26Z +issue,32034,Extend `ConcatDataset` to return dataset index,https://github.com/pytorch/pytorch/issues/32034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-07-10T21:46:56Z +issue,113329,Mark aten.normal as a core aten op,https://github.com/pytorch/pytorch/issues/113329,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, oncall: pt2",scan when time allows,2024-07-10T19:20:08Z +issue,129625,Triton segfault with compiled flexattention kernel,https://github.com/pytorch/pytorch/issues/129625,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-10T18:02:05Z +issue,114353,add support for linalg lstsq in dynamo,https://github.com/pytorch/pytorch/issues/114353,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-07-10T16:08:51Z +issue,130432,Deferred runtime asserts triggered during backwards should not trigger assert not self.runtime_asserts_frozen,https://github.com/pytorch/pytorch/issues/130432,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes,scan when time allows,2024-07-10T14:33:22Z +issue,130124,Pin memory in subprocess,https://github.com/pytorch/pytorch/issues/130124,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader",scan when time allows,2024-07-10T06:31:55Z +issue,129329,_foreach_addc_,https://github.com/pytorch/pytorch/issues/129329,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer",scan when time allows,2024-07-10T01:21:26Z +issue,129672,Distributed training with NCCL hangs on init_process_group when debug is DETAIL,https://github.com/pytorch/pytorch/issues/129672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-09T18:28:59Z +issue,11372,Suggest: DataLoader add device parameter,https://github.com/pytorch/pytorch/issues/11372,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-07-09T14:34:17Z +issue,130254,"Many types for `start`, `end` and `base` argument of `logspace()` work againt what the doc says",https://github.com/pytorch/pytorch/issues/130254,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-07-09T13:51:05Z +issue,130128,Installation fails with conflicting dependencies (MarkupSafe>=2.0),https://github.com/pytorch/pytorch/issues/130128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-07-09T13:44:44Z +issue,42791,Build fail on Ubuntu 18.04: caffe2/CMakeFiles/init_test.dir/build.make:106: recipe for target 'bin/init_test' failed,https://github.com/pytorch/pytorch/issues/42791,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-07-09T08:08:01Z +issue,130193,Inconsistency `CrossEntropyLoss` weights value between in gpu and in cpu.,https://github.com/pytorch/pytorch/issues/130193,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2024-07-09T00:39:15Z +issue,129749,nccl deadlock,https://github.com/pytorch/pytorch/issues/129749,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-08T20:59:54Z +issue,129442,Volatile GPU-Util keeps 100% when using torch.distributed.isend/irecv,https://github.com/pytorch/pytorch/issues/129442,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-07-08T20:19:51Z +issue,121661,Look at gather + gemv performance.,https://github.com/pytorch/pytorch/issues/121661,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: pt2",scan when time allows,2024-07-08T16:23:09Z +issue,117681,Improved workflow for tracing dynamism through complicated model,https://github.com/pytorch/pytorch/issues/117681,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2024-07-08T16:19:26Z +issue,130078,cuda graph error for torch.isin if the input tensors are too large,https://github.com/pytorch/pytorch/issues/130078,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs",scan when time allows,2024-07-08T14:56:08Z +issue,130217,`unique()` is not intuitively understandable at all.,https://github.com/pytorch/pytorch/issues/130217,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-08T14:50:43Z +issue,130162,caching_allocator_delete does not free up the memory allocated by caching_allocator_alloc,https://github.com/pytorch/pytorch/issues/130162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2024-07-08T14:44:33Z +issue,130096,The time consumption gradually increases when activation checkpoint is iterated and run multiple times,https://github.com/pytorch/pytorch/issues/130096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2024-07-08T14:39:24Z +issue,111159,Wrong onnx model from `torch.onnx.export` when using `index_add_` function with duplicate `index` values.,https://github.com/pytorch/pytorch/issues/111159,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: onnx",scan when time allows,2024-07-08T01:24:08Z +issue,127743,RuntimeError using nested tensor in Apple M1 device MPS,https://github.com/pytorch/pytorch/issues/127743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: nestedtensor",scan when time allows,2024-07-06T05:17:53Z +issue,129310,Upgrade dependencies MKL and Intel OpenMP to 2024.2.0,https://github.com/pytorch/pytorch/issues/129310,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: openmp, module: windows",scan when time allows,2024-07-05T21:10:32Z +issue,109213,inductor/test_max_autotune having timeout issues,https://github.com/pytorch/pytorch/issues/109213,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: tests, oncall: pt2",scan when time allows,2024-07-05T20:32:07Z +issue,110448,Explore Hybrid (CPU+GPU) Graphs in Scalar parameters,https://github.com/pytorch/pytorch/issues/110448,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-07-05T19:54:47Z +issue,111033,`CapabilityBasedPartitioner` returns invalid partitions.,https://github.com/pytorch/pytorch/issues/111033,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: xla, oncall: fx",scan when time allows,2024-07-05T19:39:14Z +issue,124270,PyTorch build fail with GCC 13.2.0 due to aten/src/ATen/core/IListRef_inl.h:171:17: error: possibly dangling reference to a temporary [-Werror=dangling-reference],https://github.com/pytorch/pytorch/issues/124270,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-07-05T07:51:45Z +issue,126980,Long build compilation time (>5 hours) for CUDA ARM build,https://github.com/pytorch/pytorch/issues/126980,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: cuda",scan when time allows,2024-07-03T21:55:25Z +issue,130058,[pipelining] Expose a transform function to inject additional metadata into model forward,https://github.com/pytorch/pytorch/issues/130058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pipelining,scan when time allows,2024-07-03T20:48:24Z +issue,24139,Allow incompatible shapes in load_state_dict(strict=False),https://github.com/pytorch/pytorch/issues/24139,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-07-03T17:37:16Z +issue,130021,[Functionalization] The inplace decompose op is meaningless,https://github.com/pytorch/pytorch/issues/130021,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: xla",scan when time allows,2024-07-03T14:20:55Z +issue,40859,Option to allow loading state dict with mismatching shapes.,https://github.com/pytorch/pytorch/issues/40859,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization",scan when time allows,2024-07-03T10:24:49Z +issue,122568,How do I use exec in a Pytorch Module and train with multiple GPUs?,https://github.com/pytorch/pytorch/issues/122568,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2024-07-03T10:12:51Z +issue,129954,"RuntimeError: NVML_SUCCESS == DriverAPI::get()->nvmlInit_v2_() INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":799",https://github.com/pytorch/pytorch/issues/129954,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-07-03T00:03:40Z +issue,128735,Flaky test page should include retry runs,https://github.com/pytorch/pytorch/issues/128735,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2024-07-02T21:56:26Z +issue,129670,Libtorch does not support SM90,https://github.com/pytorch/pytorch/issues/129670,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-07-02T20:01:56Z +issue,10396,"[feature request] Provide a way to redirect shared memory prefix ""/torch_""",https://github.com/pytorch/pytorch/issues/10396,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-07-02T17:30:55Z +issue,129784,`@pytorchbot merge -i` not ignoring currently failing checks,https://github.com/pytorch/pytorch/issues/129784,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-07-01T21:31:10Z +issue,50010,torch.var and torch.std are not compatible with np.var and np.std,https://github.com/pytorch/pytorch/issues/50010,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy, module: reductions",scan when time allows,2024-07-01T18:52:10Z +issue,106544,"Branch name in double quotes """"",https://github.com/pytorch/pytorch/issues/106544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-07-01T18:23:24Z +issue,124778,Dense-sparse broadcasted multiplication fails in the backward pass,https://github.com/pytorch/pytorch/issues/124778,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-07-01T17:42:26Z +issue,112831,dynamo_export successfully export model but fails at onnx.checker.check_model,https://github.com/pytorch/pytorch/issues/112831,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-07-01T15:31:09Z +issue,129293,forward_ad ignores checkpoints,https://github.com/pytorch/pytorch/issues/129293,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd, module: bc-breaking, module: forward ad",scan when time allows,2024-07-01T05:02:11Z +issue,118499,Autocast on cpu dramatically slow,https://github.com/pytorch/pytorch/issues/118499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2024-07-01T03:59:35Z +issue,128053,Handling reasoning about rationals in symbolic shapes,https://github.com/pytorch/pytorch/issues/128053,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes,scan when time allows,2024-06-30T13:00:27Z +issue,129224,"_If pin_memory_thread is alive but dataqueue is empty, it will fall into a dead loop,",https://github.com/pytorch/pytorch/issues/129224,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2024-06-28T15:54:03Z +issue,109292,aten::squeeze exported to ONNX as an `If` node,https://github.com/pytorch/pytorch/issues/109292,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-06-28T12:59:49Z +issue,123062,Enable UFMT on all files in PyTorch,https://github.com/pytorch/pytorch/issues/123062,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2024-06-28T06:19:13Z +issue,126005,CUDNN not detected in official nightly devel image. We need to use `12.4.1` upstream images,https://github.com/pytorch/pytorch/issues/126005,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cudnn",scan when time allows,2024-06-27T20:37:20Z +issue,102600,Support for activation checkpoint on demand in custom function,https://github.com/pytorch/pytorch/issues/102600,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2024-06-27T20:28:56Z +issue,126216,Allow Lazy Initialization for PipelineStage,https://github.com/pytorch/pytorch/issues/126216,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-27T19:18:05Z +issue,92310,[Releng] Windows AMI needs to be pinned for release,https://github.com/pytorch/pytorch/issues/92310,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-06-27T17:45:26Z +issue,129447,"Pytorch of 2.3.0 version loads bool type mask in triton, encountering error when use tl.where.",https://github.com/pytorch/pytorch/issues/129447,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-06-27T15:17:08Z +issue,129564,Support for 5-D output image tensors in `Col2Im`.,https://github.com/pytorch/pytorch/issues/129564,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend",scan when time allows,2024-06-27T14:43:59Z +issue,116707,AttributeError: `Adam` object has no attribute `param_groups` with Python 3.11,https://github.com/pytorch/pytorch/issues/116707,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: serialization",scan when time allows,2024-06-27T14:36:35Z +issue,59439,NaN values on torch.nn.functional.conv2d (aarch64),https://github.com/pytorch/pytorch/issues/59439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: convolution, module: jetson, module: nn",scan when time allows,2024-06-27T14:09:53Z +issue,116608,unstable test_GroupNorm_general_cpu in test/test_nn.py,https://github.com/pytorch/pytorch/issues/116608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-27T02:16:49Z +issue,129225,Why need to transpose when collate a sequence data in dataloader?,https://github.com/pytorch/pytorch/issues/129225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2024-06-27T02:02:53Z +issue,126472,[NestedTensor] Graph breaks with SDPA + NT constructor,https://github.com/pytorch/pytorch/issues/126472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-06-26T23:02:11Z +issue,129555,Libtorch 2.3.1 requires Glibc_2.28 and can't be used on ubuntu 18.,https://github.com/pytorch/pytorch/issues/129555,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp, module: docs, module: intel",scan when time allows,2024-06-26T21:02:58Z +issue,112956,Optim.step() is significantly SLOW on MPS,https://github.com/pytorch/pytorch/issues/112956,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2024-06-26T20:58:52Z +issue,129565,ChainedScheduler fail on CosineAnnealingWarmRestarts,https://github.com/pytorch/pytorch/issues/129565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2024-06-26T14:47:56Z +issue,129410,"line-length of ruff, black and isort are not the same in pyproject.toml",https://github.com/pytorch/pytorch/issues/129410,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2024-06-26T14:45:46Z +issue,129272,RecursionError for MaskedTensor.where,https://github.com/pytorch/pytorch/issues/129272,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2024-06-26T14:38:17Z +issue,109873,Allow try except check for numpy bfloat16 representation,https://github.com/pytorch/pytorch/issues/109873,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-06-26T08:20:58Z +issue,122321,Batch size can't be varied if you export self attention to onnx,https://github.com/pytorch/pytorch/issues/122321,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: onnx, module: windows",scan when time allows,2024-06-26T06:43:27Z +issue,77165,Automate cleanup of header includes,https://github.com/pytorch/pytorch/issues/77165,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2024-06-26T00:19:53Z +issue,72263,Removing deprecated `cpp_custom_type_hack`,https://github.com/pytorch/pytorch/issues/72263,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: quantization",scan when time allows,2024-06-26T00:15:25Z +issue,46240,"Support ""symmetric"" reflection padding",https://github.com/pytorch/pytorch/issues/46240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: numpy, module: padding, module: tensorflow",scan when time allows,2024-06-25T18:24:30Z +issue,120240,Cannot get deterministic Mask RCNN without running out of CUDA memory,https://github.com/pytorch/pytorch/issues/120240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: determinism",scan when time allows,2024-06-25T18:14:08Z +issue,71872,Clarify the behavior of DataLoader sampler and batch_sampler parameters,https://github.com/pytorch/pytorch/issues/71872,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: docs",scan when time allows,2024-06-25T16:55:39Z +issue,73479,Strange case of empty non-coalesced sparse tensor,https://github.com/pytorch/pytorch/issues/73479,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-06-25T11:52:42Z +issue,101191,custom_op API follow-ups,https://github.com/pytorch/pytorch/issues/101191,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-06-24T18:20:30Z +issue,129330,Fuyou Training Framework Integration for PyTorch,https://github.com/pytorch/pytorch/issues/129330,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-06-24T18:05:10Z +issue,129386,Support a more flexible way to mark all CI jobs in a similar category as unstable,https://github.com/pytorch/pytorch/issues/129386,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: devx",scan when time allows,2024-06-24T16:46:55Z +issue,129336,pytorch-nightly export KINETO_USE_DAEMON=1 Cannot initialize CUDA without ATen_cuda library,https://github.com/pytorch/pytorch/issues/129336,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-06-24T15:55:33Z +issue,129027,Extract some public APIs from torch::cuda::initModule(module) to torch::initModule(),https://github.com/pytorch/pytorch/issues/129027,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2024-06-24T08:44:10Z +issue,129222,ignore all target for CrossEntropyLoss 2d，return nan but 1d return 0,https://github.com/pytorch/pytorch/issues/129222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2024-06-24T07:48:17Z +issue,128963,RuntimeError: NCCL error: invalid usage when deploy LLM model by vllm. (torch version: 2.3.0+cu118),https://github.com/pytorch/pytorch/issues/128963,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-24T04:06:47Z +issue,128495,'make html' will print 'duplicate object description' warnings when there are 1~5 CPUs in the running machine,https://github.com/pytorch/pytorch/issues/128495,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-06-22T18:13:11Z +issue,112089,"Failed to compile: null in call to `__builtin_memmove(__result, __first, sizeof(_Tp) * _Num);` Debian 12, ppc64le, gcc 12.2",https://github.com/pytorch/pytorch/issues/112089,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-06-22T15:28:13Z +issue,128960,[Dynamo x torch_function] magic methods and methods with regular Tensors don't seem to work,https://github.com/pytorch/pytorch/issues/128960,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-06-22T00:17:20Z +issue,15167,cudnn not found,https://github.com/pytorch/pytorch/issues/15167,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cudnn",scan when time allows,2024-06-21T23:28:26Z +issue,129191,RuntimeError when using torch.ops.aten._jagged_to_padded_dense_forward with large jagged tensors,https://github.com/pytorch/pytorch/issues/129191,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: sparse",scan when time allows,2024-06-21T19:12:11Z +issue,109445,Memory usage steadily increasing when using back propagation with sparse CSR parameter matrices on CPU,https://github.com/pytorch/pytorch/issues/109445,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: sparse",scan when time allows,2024-06-21T18:04:45Z +issue,129050,Add comment for label_smoothing parameter in torch.nn.CrossEntropyLoss,https://github.com/pytorch/pytorch/issues/129050,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2024-06-21T14:36:55Z +issue,129085,Adding betainc,https://github.com/pytorch/pytorch/issues/129085,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: special",scan when time allows,2024-06-21T13:39:38Z +issue,128665,[RFC][pipelining] PipelineStage should let user control send/recv endpoints,https://github.com/pytorch/pytorch/issues/128665,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pipelining, oncall: distributed",scan when time allows,2024-06-20T17:19:46Z +issue,23854,Fan out calculation broken for group (depthwise) convolution,https://github.com/pytorch/pytorch/issues/23854,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: initialization",scan when time allows,2024-06-20T13:33:28Z +issue,128706,does FSDP support AMSP (a new DP shard strategy),https://github.com/pytorch/pytorch/issues/128706,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-20T12:19:54Z +issue,73909,SummaryWriter reports encoding error,https://github.com/pytorch/pytorch/issues/73909,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2024-06-20T08:27:01Z +issue,117130,`torch.cuda.memory_summary()` can give `KeyError`,https://github.com/pytorch/pytorch/issues/117130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-06-19T20:35:55Z +issue,128988,Look up tensor device member inside Tensor is_pinned() implementation instead of accepting an outside input,https://github.com/pytorch/pytorch/issues/128988,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-06-19T16:40:45Z +issue,128952,Question about torch.lstsq and torch.linalg.lstsq,https://github.com/pytorch/pytorch/issues/128952,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-06-19T12:05:43Z +issue,128936,Performance degradation for certain input using Conv2D,https://github.com/pytorch/pytorch/issues/128936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: performance",scan when time allows,2024-06-19T11:51:52Z +issue,13058,Backward pass over torch.nn.functional.pad is extremely slow with half tensors,https://github.com/pytorch/pytorch/issues/13058,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half, module: performance",scan when time allows,2024-06-19T11:17:36Z +issue,128821,Questions about TCPStoreLibUV,https://github.com/pytorch/pytorch/issues/128821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-19T05:51:53Z +issue,122711,`torch.sparse.sum` does not support boolean and int when summing over dense dimensions,https://github.com/pytorch/pytorch/issues/122711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-06-18T21:38:05Z +issue,102878,[dynamo] Diffusers - Graph break on OrderedDict,https://github.com/pytorch/pytorch/issues/102878,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamo,scan when time allows,2024-06-18T18:50:06Z +issue,118227,TorchDynamo mistranslates end of tensor slice,https://github.com/pytorch/pytorch/issues/118227,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, oncall: pt2",scan when time allows,2024-06-18T17:05:40Z +issue,128959,caffe2 removal,https://github.com/pytorch/pytorch/issues/128959,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-06-18T15:51:46Z +issue,128791,CUDA error in torch.cdist with compute_mode=donot_use_mm_for_euclid_dist,https://github.com/pytorch/pytorch/issues/128791,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: loss",scan when time allows,2024-06-18T14:08:13Z +issue,126360,Dynamo ignores frame using yield,https://github.com/pytorch/pytorch/issues/126360,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: pt2",scan when time allows,2024-06-18T05:57:08Z +issue,122092,masked_index_add,https://github.com/pytorch/pytorch/issues/122092,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs, module: performance, module: python frontend",scan when time allows,2024-06-17T23:37:15Z +issue,128696,[DDP] DDP bucket memory release during fwd step,https://github.com/pytorch/pytorch/issues/128696,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-17T21:01:43Z +issue,128699,NotImplementedError: Operator aten.native_layer_norm_backward.default does not have a sharding strategy registered.,https://github.com/pytorch/pytorch/issues/128699,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-17T19:36:21Z +issue,124195,torch.save and torch.load is slow. Slower than numpy. Slower even than pickle.,https://github.com/pytorch/pytorch/issues/124195,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-06-17T18:00:54Z +issue,128778,False INTERNAL ASSERT FAILED bug whilst training Neural Network,https://github.com/pytorch/pytorch/issues/128778,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-06-17T15:15:44Z +issue,53785,Support for one-hot of dtypes besides torch.int64,https://github.com/pytorch/pytorch/issues/53785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-17T14:57:31Z +issue,124083,Automatically bind all DispatchKey to Python,https://github.com/pytorch/pytorch/issues/124083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2024-06-17T14:37:24Z +issue,128703,autograd with `is_grads_batched=True` fails on GroupNorm,https://github.com/pytorch/pytorch/issues/128703,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch, module: norms and normalization, module: vmap",scan when time allows,2024-06-17T14:27:19Z +issue,128600,jacrev and jacfwd yield different results if one uses torch.no_grad blocks in module,https://github.com/pytorch/pytorch/issues/128600,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs, module: functorch",scan when time allows,2024-06-17T14:13:39Z +issue,76383,Suggestion to throw a UserWarning when a user forgot .eval() mode during inference,https://github.com/pytorch/pytorch/issues/76383,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: molly-guard, module: nn",scan when time allows,2024-06-16T18:01:58Z +issue,127676,module 'torch.mps' has no attribute 'device',https://github.com/pytorch/pytorch/issues/127676,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2024-06-15T00:13:19Z +issue,110074,ImportError: libc10_cuda.so: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/110074,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2024-06-14T23:45:18Z +issue,128165,`swap_tensors` fail when calling `nn.Module.to` on XLA `DDP` wrapped models.,https://github.com/pytorch/pytorch/issues/128165,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: xla",scan when time allows,2024-06-14T22:19:20Z +issue,128493,Inconsistency `CrossEntropyLoss` vs `BCELoss` regarding logits/probability space,https://github.com/pytorch/pytorch/issues/128493,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-14T18:59:06Z +issue,119557,Python 3.12 CPU Build system cannot find MKL libraries,https://github.com/pytorch/pytorch/issues/119557,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci, module: mkl",scan when time allows,2024-06-14T17:36:32Z +issue,126286,Adopt inductor-perf-test-nightly.yml to be used for release testing,https://github.com/pytorch/pytorch/issues/126286,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-06-14T15:17:37Z +issue,88148,☂️ Issues that trigger crashes due to corner-case API usages,https://github.com/pytorch/pytorch/issues/88148,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2024-06-14T13:44:39Z +issue,107087,RuntimeError with operations on torch.float8_e5m2 and torch.float_e4m3fn data types,https://github.com/pytorch/pytorch/issues/107087,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2024-06-14T09:59:57Z +issue,11982,[feature request] Publish wheels with debug symbols,https://github.com/pytorch/pytorch/issues/11982,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-06-14T07:54:20Z +issue,123238,Fused optimizer kernel for CPU,https://github.com/pytorch/pytorch/issues/123238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-06-14T03:15:34Z +issue,128509,elements of STL get converted to an `IValue` and cannot be converted to an STL.,https://github.com/pytorch/pytorch/issues/128509,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cpp-extensions",scan when time allows,2024-06-14T01:34:29Z +issue,128641,Backward is not supported by `MaskedTensor`'s `torch.maximum`,https://github.com/pytorch/pytorch/issues/128641,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: masked operators",scan when time allows,2024-06-14T01:31:37Z +issue,128548,[torch.compile] Llama2 failure using dynamic shapes with Torch 2.4 nightly,https://github.com/pytorch/pytorch/issues/128548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes, oncall: pt2",scan when time allows,2024-06-13T18:56:04Z +issue,4073,Feature request: Correlation module,https://github.com/pytorch/pytorch/issues/4073,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-06-13T17:51:15Z +issue,128601,Retry on CUBLAS_STATUS_ALLOC_FAILED,https://github.com/pytorch/pytorch/issues/128601,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-06-13T15:49:52Z +issue,124133,ZeroRedundancyOptimizer + AdamW Fused can't load state_dict back,https://github.com/pytorch/pytorch/issues/124133,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: distributed",scan when time allows,2024-06-12T13:45:52Z +issue,45685,"[fx] symbolic trace ""is None"" and ""is not None"" checks",https://github.com/pytorch/pytorch/issues/45685,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-06-12T10:53:59Z +issue,31689,DataLoader does not consider default floating point type,https://github.com/pytorch/pytorch/issues/31689,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-06-12T10:51:45Z +issue,123113,"Please add instructions for ""uninstall compiled PyTorch from source"" in the readme.",https://github.com/pytorch/pytorch/issues/123113,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-06-11T22:02:21Z +issue,93081,[Releng] Optimize PyTorch packaging,https://github.com/pytorch/pytorch/issues/93081,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-06-11T21:57:16Z +issue,128257,can not create custom gradient when fp8 type is used,https://github.com/pytorch/pytorch/issues/128257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2024-06-11T20:55:06Z +issue,118570,[Inductor] Kill Mutation Layout,https://github.com/pytorch/pytorch/issues/118570,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: inductor,scan when time allows,2024-06-11T19:46:16Z +issue,29863,n-dimensional non-constant padding functional,https://github.com/pytorch/pytorch/issues/29863,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",scan when time allows,2024-06-11T15:47:50Z +issue,119740,Devcontainer for the nightly devenv,https://github.com/pytorch/pytorch/issues/119740,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-06-11T13:26:48Z +issue,127535,Can we have a Y-Split module for torch.nn.Sequential?,https://github.com/pytorch/pytorch/issues/127535,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-11T12:19:47Z +issue,50210,"Pytorch 1.7.1 hangs with multi-gpu, while Pytorch 1.6.0 works correctly",https://github.com/pytorch/pytorch/issues/50210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: deadlock",scan when time allows,2024-06-11T09:35:31Z +issue,89158,Support disallowing calls to certain instance methods in TorchDynamo,https://github.com/pytorch/pytorch/issues/89158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamo,scan when time allows,2024-06-10T22:24:46Z +issue,128351,`Tensor.is_shared` behavior is incorrect for tensors created with `torch.from_file`,https://github.com/pytorch/pytorch/issues/128351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: serialization",scan when time allows,2024-06-10T21:53:24Z +issue,105790,torch.sparse.mm() with reduce operator for GPU support and COO matrices,https://github.com/pytorch/pytorch/issues/105790,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: reductions, module: sparse",scan when time allows,2024-06-10T20:51:38Z +issue,100626,Will Deep Implicit Models ever become first class citizens in PyTorch?,https://github.com/pytorch/pytorch/issues/100626,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2024-06-10T20:51:12Z +issue,128296,Proposal to add a CUDA backend for tensor conversion from BSR/BSC tensors to CSR/CSC tensors,https://github.com/pytorch/pytorch/issues/128296,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2024-06-10T20:30:29Z +issue,100241,Interaction of torch.no_grad and torch.autocast context managers with torch.compile,https://github.com/pytorch/pytorch/issues/100241,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-06-10T20:15:32Z +issue,128208,Adam is storing step as cpu tensor,https://github.com/pytorch/pytorch/issues/128208,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-06-10T16:19:24Z +issue,128314,`docker compose up` fails with `RuntimeError: Unexpected error from cudaGetDeviceCount()`,https://github.com/pytorch/pytorch/issues/128314,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2024-06-10T16:18:49Z +issue,68258,Option to configure TorchElastic multiprocessing/api.py PContext.close() timeout,https://github.com/pytorch/pytorch/issues/68258,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-06-09T19:50:14Z +issue,104507,[error] while Implementation of pytorch DistributedParallel,https://github.com/pytorch/pytorch/issues/104507,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-08T18:29:00Z +issue,124906,Many cases in distributed/elastic/multiprocessing/redirects_test.py fails when use pytest,https://github.com/pytorch/pytorch/issues/124906,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, module: tests, oncall: distributed",scan when time allows,2024-06-07T15:32:52Z +issue,128033,Network module memory is not released in libtorch,https://github.com/pytorch/pytorch/issues/128033,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp, module: memory usage",scan when time allows,2024-06-07T05:23:33Z +issue,52289,DataParallel module fails to handle data with size not divisible by number of GPUs,https://github.com/pytorch/pytorch/issues/52289,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2024-06-07T02:24:44Z +issue,105460,Specify version,https://github.com/pytorch/pytorch/issues/105460,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-06-07T01:38:54Z +issue,128066,reduce_scatter_tensor with strided inputs produces corrupted results,https://github.com/pytorch/pytorch/issues/128066,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-06T23:32:09Z +issue,128128,Fake tensor support of distributed ops in FX graph tracing,https://github.com/pytorch/pytorch/issues/128128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-06T23:24:39Z +issue,127964,[pipelining] Supporting skip connections in 1F1B schedules,https://github.com/pytorch/pytorch/issues/127964,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pipelining,scan when time allows,2024-06-06T23:19:44Z +issue,116539,Auto bisect for broken periodic CI jobs,https://github.com/pytorch/pytorch/issues/116539,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-06-06T17:50:43Z +issue,127950,Ensuring reliability of PT2 with TorchRec on PRs,https://github.com/pytorch/pytorch/issues/127950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: pt2",scan when time allows,2024-06-06T16:57:35Z +issue,128044,Request for in-place FFT,https://github.com/pytorch/pytorch/issues/128044,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2024-06-06T15:56:42Z +issue,128126,Stateless random numbers,https://github.com/pytorch/pytorch/issues/128126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2024-06-06T15:43:07Z +issue,126952,Add integrity check in torch.save,https://github.com/pytorch/pytorch/issues/126952,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-06-06T02:22:00Z +issue,117883,Flight Recorder - feature requests,https://github.com/pytorch/pytorch/issues/117883,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-06T00:06:57Z +issue,122931,Stack smashing detected on nccl initialization,https://github.com/pytorch/pytorch/issues/122931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2024-06-05T22:58:21Z +issue,126008,Torch distribution blocking ports and failing to connect,https://github.com/pytorch/pytorch/issues/126008,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: elastic, oncall: distributed",scan when time allows,2024-06-05T22:50:32Z +issue,127873,[libtorch] `CppCommInterface` subclasses cannot be instantiated,https://github.com/pytorch/pytorch/issues/127873,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2024-06-05T22:24:45Z +issue,126773,Give`wait_tensor()` a schema that reflects its side effect,https://github.com/pytorch/pytorch/issues/126773,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-06-05T22:11:54Z +issue,121207,can not find all the deps when install the native CUDA-toolkit and torch in Redhat-like systems,https://github.com/pytorch/pytorch/issues/121207,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2024-06-04T11:36:44Z +issue,127750,aot_function cannot handle index_add,https://github.com/pytorch/pytorch/issues/127750,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-06-04T08:18:46Z +issue,107026,a bug about tensor stride,https://github.com/pytorch/pytorch/issues/107026,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2024-06-04T05:25:08Z +issue,127788,ghstack and pytorch-bot can race,https://github.com/pytorch/pytorch/issues/127788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-06-03T20:16:34Z +issue,127754,profile not record at::meta_xxx time,https://github.com/pytorch/pytorch/issues/127754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2024-06-03T18:00:45Z +issue,127733,torch.nn.Linear does not check the value of the out_features parameter,https://github.com/pytorch/pytorch/issues/127733,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-03T17:58:30Z +issue,127737,"`torch.tensor()` can also be used with a list of the `float`, `complex` or `bool` type tensors which have a single element each",https://github.com/pytorch/pytorch/issues/127737,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2024-06-03T17:56:47Z +issue,127732,torch.nn.ConvTranspose2d/torch.nn.ConvTranspose1d lacks a check for the parameter dilation in the model building process,https://github.com/pytorch/pytorch/issues/127732,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-03T17:54:50Z +issue,127734,nn.ConvTranspose2d does not check for parameter combinations resulting in convolution output with dimension 0,https://github.com/pytorch/pytorch/issues/127734,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-03T17:54:34Z +issue,127701,The type of exception thrown by torch.nn.Conv2d does not match the actual situation,https://github.com/pytorch/pytorch/issues/127701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-06-03T17:52:24Z +issue,127606,` to be a quantized tensor. Is this likely due to missing support for quantized `onnx::Pad`,https://github.com/pytorch/pytorch/issues/127606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-06-03T17:42:47Z +issue,32407,Build without MKL is not possible when MKL is installed,https://github.com/pytorch/pytorch/issues/32407,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkl",scan when time allows,2024-06-03T15:27:43Z +issue,127194,Should `torch.Size` convert np.ndarrays to lists of ints?,https://github.com/pytorch/pytorch/issues/127194,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-06-03T11:24:40Z +issue,119343,Use Dtensor and ColwiseParallel api to do Tensor Parallel for llama2 model but got communication error.,https://github.com/pytorch/pytorch/issues/119343,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2024-06-03T09:04:07Z +issue,127541,RuntimeError: derivative for aten::_spdiags is not implemented,https://github.com/pytorch/pytorch/issues/127541,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2024-06-03T07:55:39Z +issue,93188,Why is AvgPool2D taking longer than Conv2D for the same input?,https://github.com/pytorch/pytorch/issues/93188,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2024-06-03T06:52:36Z +issue,127072,PyTorch DataLoader improvements for Iterable Dataset,https://github.com/pytorch/pytorch/issues/127072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-06-01T09:33:40Z +issue,107611,Previous version not found,https://github.com/pytorch/pytorch/issues/107611,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-05-31T18:42:12Z +issue,126755,inplace parameter in dropouts should function as expected regardless of the value of training(or train) paramter,https://github.com/pytorch/pytorch/issues/126755,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-05-31T17:00:26Z +issue,91737,[MPS] Improve the performance of torch.linear(),https://github.com/pytorch/pytorch/issues/91737,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: mps",scan when time allows,2024-05-31T09:50:57Z +issue,127530,Run rstcheck on modified docstrings and docs as additional linter,https://github.com/pytorch/pytorch/issues/127530,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2024-05-31T01:59:03Z +issue,127518,THPVariable_Check(list_elem) INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/127518,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-31T00:05:33Z +issue,69364,"[feature request] `quantized::linear_dynamic` on CUDA/eager, and other quantized and low-level int8 operators (matmul, gemm etc) on CUDA + integrate LLM.int8 + integrate ZeroQuant?",https://github.com/pytorch/pytorch/issues/69364,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: quantization",scan when time allows,2024-05-30T23:05:17Z +issue,96073,Pytorch 2.0 installation tutorial does not work under Macbook,https://github.com/pytorch/pytorch/issues/96073,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: macos",scan when time allows,2024-05-30T19:39:34Z +issue,127555,"Reserving this issue to test something, don't mind me",https://github.com/pytorch/pytorch/issues/127555,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-30T18:53:49Z +issue,127270,Not supporting higher protobuf versions,https://github.com/pytorch/pytorch/issues/127270,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: protobuf,scan when time allows,2024-05-30T16:57:05Z +issue,127497,torch.compiler docstrings can be derived from _dynamo and _inductor docs,https://github.com/pytorch/pytorch/issues/127497,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: pt2",scan when time allows,2024-05-30T16:53:59Z +issue,127422,index_select op not implemented on Vulkan backend,https://github.com/pytorch/pytorch/issues/127422,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: vulkan",scan when time allows,2024-05-30T16:14:58Z +issue,127221,lacking checking for ConvTranspose's parameters when running with GPUs,https://github.com/pytorch/pytorch/issues/127221,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: error checking",scan when time allows,2024-05-30T16:08:33Z +issue,127223,[BUG] torch.linalg.lstsq returning wrong result.,https://github.com/pytorch/pytorch/issues/127223,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-05-30T14:58:17Z +issue,127525,Multiple unhandled exceptions in weights_only unpickler,https://github.com/pytorch/pytorch/issues/127525,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-05-30T14:35:54Z +issue,127387,"Backwards pass through Beta distribution rsample gives inf for 4 < alpha - 2**16 < 1040, beta = 3/2",https://github.com/pytorch/pytorch/issues/127387,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-05-30T13:42:51Z +issue,127375,Different tensor strides can result in surprisingly large discrepancies in Conv2d outputs,https://github.com/pytorch/pytorch/issues/127375,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: tf32",scan when time allows,2024-05-30T13:33:50Z +issue,48389,"Add parameter ""half_pixel_center =False"" to the Bilinear function",https://github.com/pytorch/pytorch/issues/48389,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2024-05-30T11:46:19Z +issue,29398,[feature request] torch.kthvalue to support a new argument largest,https://github.com/pytorch/pytorch/issues/29398,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2024-05-30T01:36:46Z +issue,118507,RuntimeError in torch.istft with center=False: Window Overlap Add Issue,https://github.com/pytorch/pytorch/issues/118507,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2024-05-29T19:57:46Z +issue,49298,[question] How hard would it be to implement 4-bit precision training?,https://github.com/pytorch/pytorch/issues/49298,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2024-05-29T19:02:17Z +issue,41571,"torch.cuda.BoolTensor uses 8 bits per element, not 1 bit as reported by element_size()",https://github.com/pytorch/pytorch/issues/41571,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-05-29T19:02:13Z +issue,125467,Inaccurate filename due to test class metaprogramming that doesn't set __file__ and __name__,https://github.com/pytorch/pytorch/issues/125467,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devx, module: tests",scan when time allows,2024-05-29T16:48:00Z +issue,127196,torch.topk results differ on CPU and CUDA,https://github.com/pytorch/pytorch/issues/127196,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sorting and selection",scan when time allows,2024-05-29T01:58:35Z +issue,126714,move label_to_label.yml to an issue?,https://github.com/pytorch/pytorch/issues/126714,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-05-29T00:27:07Z +issue,127352,`flake8: noqa` disables flake8 linter for the whole file and it's not obvious,https://github.com/pytorch/pytorch/issues/127352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2024-05-29T00:05:57Z +issue,80411,Tracker: Slow gradcheck failures possibly indicating incorrect gradients,https://github.com/pytorch/pytorch/issues/80411,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2024-05-28T14:54:29Z +issue,126899,function signature for multiprocessing.spawn is multiprocessing.spawn.spawn,https://github.com/pytorch/pytorch/issues/126899,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-05-28T13:46:52Z +issue,127282,torch.fx.symbolic_trace doesn't support many Callable types,https://github.com/pytorch/pytorch/issues/127282,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-05-28T12:50:33Z +issue,127189,`store_param_remainders` from Apex DistributedFusedAdam,https://github.com/pytorch/pytorch/issues/127189,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: distributed",scan when time allows,2024-05-28T12:00:24Z +issue,127164,randn generates different output for 4x4 tensor size sliced to match shape of direct 2x4 or 4x2 and compare output,https://github.com/pytorch/pytorch/issues/127164,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2024-05-28T11:59:57Z +issue,56360,Support for stereo audio data in from torch.utils.tensorboard.SummaryWriter,https://github.com/pytorch/pytorch/issues/56360,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: tensorboard",scan when time allows,2024-05-27T10:39:57Z +issue,22281,Box constraints for optimizers,https://github.com/pytorch/pytorch/issues/22281,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-27T09:36:08Z +issue,122068,"NVML_SUCCESS == r INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":830, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/122068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-05-27T09:15:42Z +issue,127075,What is the processing principle when the complex64 input tensor contains nan or inf for addition?,https://github.com/pytorch/pytorch/issues/127075,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2024-05-27T03:59:52Z +issue,101273,IPEX as TorchDynamo Backend Performance Dashboard,https://github.com/pytorch/pytorch/issues/101273,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",scan when time allows,2024-05-27T01:33:33Z +issue,127031,Make grid_sample a bit more intuitive to use,https://github.com/pytorch/pytorch/issues/127031,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2024-05-24T17:46:44Z +issue,127083,"Setting a `float`, `complex` or `bool` type value to `precision` argument of `set_printoptions()` has problem when creating a `float`, `complex` or `bool` type tensor",https://github.com/pytorch/pytorch/issues/127083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-24T17:39:35Z +issue,126947,`dsplit()` with `indices_or_sections=` doesn't work while `dsplit()` without `indices_or_sections=` works,https://github.com/pytorch/pytorch/issues/126947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-24T17:05:23Z +issue,127010,"`indices_or_sections` parameter of `tensor_split()` doesn't work while `sections`, `indices` or `tensor_indices_or_sections` parameter works",https://github.com/pytorch/pytorch/issues/127010,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-24T16:42:01Z +issue,127009,`indices_or_sections` parameter of `dsplit()` doesn't work while `sections` or `indices` parameter works,https://github.com/pytorch/pytorch/issues/127009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-24T16:41:56Z +issue,127019,`indices_or_sections` parameter of `vsplit()` doesn't work while `sections` or `indices` parameter works,https://github.com/pytorch/pytorch/issues/127019,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-24T16:41:51Z +issue,127020,`indices_or_sections` parameter of `hsplit()` doesn't work while `sections` or `indices` parameter works,https://github.com/pytorch/pytorch/issues/127020,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-24T16:40:14Z +issue,63337,Property based testing like AFL,https://github.com/pytorch/pytorch/issues/63337,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2024-05-24T09:12:24Z +issue,63971,benchmark.Compare raises: TypeError: object of type 'NoneType' has no len(),https://github.com/pytorch/pytorch/issues/63971,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: benchmark, module: error checking",scan when time allows,2024-05-24T04:55:04Z +issue,77880,Make the appropriate backend `DimensionNode` visible to LTC core,https://github.com/pytorch/pytorch/issues/77880,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-24T00:10:50Z +issue,22573,Batched symeig and qr are very slow on GPU,https://github.com/pytorch/pytorch/issues/22573,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",scan when time allows,2024-05-23T23:02:23Z +issue,126660,UnsupportedOperatorError: aten::scatter_reduce when include_self=False,https://github.com/pytorch/pytorch/issues/126660,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-23T22:34:51Z +issue,100914,[MPS] Track failures of test_module.py for MPS backend,https://github.com/pytorch/pytorch/issues/100914,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: mps",scan when time allows,2024-05-23T20:25:30Z +issue,127001,Add ability to pass `MAP_NORESERVE` to`mmap(2)` call in `torch.serialization.set_default_mmap_options`,https://github.com/pytorch/pytorch/issues/127001,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-05-23T19:56:12Z +issue,126544,ncclCommWatchdog always terminates the process and prevents error handling if CUDA context is corrupted,https://github.com/pytorch/pytorch/issues/126544,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2024-05-23T17:26:03Z +issue,70540,cumcount / cumulative count,https://github.com/pytorch/pytorch/issues/70540,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-23T17:24:15Z +issue,52915,Broadcasting behaviour for linear algebra solvers,https://github.com/pytorch/pytorch/issues/52915,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-05-22T21:27:13Z +issue,124566,Can't download pytorch/ timeouterror,https://github.com/pytorch/pytorch/issues/124566,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-05-22T20:47:22Z +issue,126549,Error: Exporting the operator 'aten::searchsorted' to ONNX opset version 17,https://github.com/pytorch/pytorch/issues/126549,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-21T18:29:48Z +issue,126748,"""torch.geqrf"" performs differently on cpu and gpu",https://github.com/pytorch/pytorch/issues/126748,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nans and infs",scan when time allows,2024-05-21T18:01:30Z +issue,126758,The unexpected behavior of `min()`,https://github.com/pytorch/pytorch/issues/126758,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2024-05-21T17:53:32Z +issue,126776,One complex number or boolean value of a 1D or more D tensor with `argmin()` and `dim=` works,https://github.com/pytorch/pytorch/issues/126776,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2024-05-21T17:13:29Z +issue,126777,One complex number or boolean value of a 1D or more D tensor with `argmax()` and `dim=` works,https://github.com/pytorch/pytorch/issues/126777,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2024-05-21T17:13:03Z +issue,125844,`WeightedRandomSampler` convert weights `dtype=torch.double` internally but `MPS` only support `dtype=torch.float32`,https://github.com/pytorch/pytorch/issues/125844,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: mps",scan when time allows,2024-05-21T15:20:15Z +issue,125234,torch.Library can easily cause segfault on loading/unloading,https://github.com/pytorch/pytorch/issues/125234,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: library,scan when time allows,2024-05-21T13:47:21Z +issue,126708,Inductor Channels-last slowdown,https://github.com/pytorch/pytorch/issues/126708,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-05-20T22:30:13Z +issue,126659,`topk()` cannot be used with `complex` or `bool` type of a 0D(only `cuda`) or more D tensor but `topk()` can be used with `complex` or `bool` type of a 0D(only `cpu`) tensor,https://github.com/pytorch/pytorch/issues/126659,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2024-05-20T21:08:03Z +issue,126658,`kthvalue()` cannot be used with `complex` or `bool` type of a 1D or more D tensor but `kthvalue()` can be used with `complex` or `bool` type of a 0D tensor,https://github.com/pytorch/pytorch/issues/126658,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2024-05-20T16:36:24Z +issue,33520,Reverse Cumulative Sum,https://github.com/pytorch/pytorch/issues/33520,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-20T15:31:04Z +issue,126404,Add `int` and `str` type to `device` parameter of `torch.tensor()`,https://github.com/pytorch/pytorch/issues/126404,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-20T15:23:15Z +issue,126386,lintrunner failures on main,https://github.com/pytorch/pytorch/issues/126386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2024-05-20T15:17:49Z +issue,126519,ambiguous overload for TORCH_LIBRARY `def`,https://github.com/pytorch/pytorch/issues/126519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2024-05-20T15:16:02Z +issue,126620,dataloader crashes after several epochs if the trained model contains triton-based operators,https://github.com/pytorch/pytorch/issues/126620,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-05-20T14:55:17Z +issue,126638,Test failures with new configs in test_grad_scaling_autocast in test_torch.py,https://github.com/pytorch/pytorch/issues/126638,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-05-20T12:34:21Z +issue,86162,torch.nn.functional.one_hot only works for int64,https://github.com/pytorch/pytorch/issues/86162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-05-19T14:33:04Z +issue,32912,Optional seq_len argument to torch.nn.utils.rnn.pad_sequence,https://github.com/pytorch/pytorch/issues/32912,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: padding,scan when time allows,2024-05-18T13:32:36Z +issue,125893,dynamo_export fails to produce a graph from nn.GLU : no tensor operations were found,https://github.com/pytorch/pytorch/issues/125893,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-17T18:31:06Z +issue,42665,Optimizer support via Libtorch C++ on Android,https://github.com/pytorch/pytorch/issues/42665,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: optimizer, oncall: mobile",scan when time allows,2024-05-17T09:56:01Z +issue,126253,`dtype` of `type()` doesn't accept a string,https://github.com/pytorch/pytorch/issues/126253,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-17T09:11:35Z +issue,20551,torch.norm produces incorrect results,https://github.com/pytorch/pytorch/issues/20551,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization, module: numerical-reproducibility",scan when time allows,2024-05-17T08:35:36Z +issue,55207,[RFC] Model Sharding for distributed training,https://github.com/pytorch/pytorch/issues/55207,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-05-17T06:54:28Z +issue,105335,Enable SLEEF on ARM,https://github.com/pytorch/pytorch/issues/105335,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: sleef",scan when time allows,2024-05-17T04:51:55Z +issue,125902,How to export onnx with fixed shape output ?,https://github.com/pytorch/pytorch/issues/125902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-17T04:35:24Z +issue,126453,Add testing for NJT serialization,https://github.com/pytorch/pytorch/issues/126453,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: serialization",scan when time allows,2024-05-16T21:02:12Z +issue,125622,Wrapper subclass serialization should preserve all _make_wrapper_subclass() args,https://github.com/pytorch/pytorch/issues/125622,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-05-16T15:48:00Z +issue,117275,deepcopy of `nn.Parameter` does not transfer __dict__,https://github.com/pytorch/pytorch/issues/117275,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-05-16T13:47:43Z +issue,75662,"multiprocessing and torch.tensor, Cannot allocate memory error",https://github.com/pytorch/pytorch/issues/75662,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-05-16T11:40:01Z +issue,126115,filtering on CPU slower in pytorch than scipy,https://github.com/pytorch/pytorch/issues/126115,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2024-05-16T05:31:57Z +issue,126323,Move pytorch_sphinx_theme out of the container build,https://github.com/pytorch/pytorch/issues/126323,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docs",scan when time allows,2024-05-16T00:58:43Z +issue,68944,Pytorch to ONNX model conversion,https://github.com/pytorch/pytorch/issues/68944,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-16T00:39:21Z +issue,98727,Pytorch member variable not working after converting to onnx format,https://github.com/pytorch/pytorch/issues/98727,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-16T00:35:34Z +issue,126347,Break the dependency between torch.nn and torch.distributed,https://github.com/pytorch/pytorch/issues/126347,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-05-15T22:52:35Z +issue,126244,"`dtype` of `torch.tensor()` also accepts `int()`, `float()` and `bool()` which are python built-in functions.",https://github.com/pytorch/pytorch/issues/126244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-15T22:21:19Z +issue,126250,"`dtype` of `to()` also accepts `int()`, `float()` and `bool()` which are python built-in functions.",https://github.com/pytorch/pytorch/issues/126250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-15T22:21:11Z +issue,126247,"`type1` and `type2` of `torch.promote_types()` also accept `int()`, `float()` and `bool()` which are python built-in functions.",https://github.com/pytorch/pytorch/issues/126247,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-15T22:21:02Z +issue,106485,Increasing batch size makes network forward 1000 times slower,https://github.com/pytorch/pytorch/issues/106485,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: nn",scan when time allows,2024-05-15T22:19:42Z +issue,119368,"Support nn.Module.to(..., copy=True)",https://github.com/pytorch/pytorch/issues/119368,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-05-15T18:15:19Z +issue,82139,Re-enable DynamicQuantModule in iOS simulator tests,https://github.com/pytorch/pytorch/issues/82139,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: ios",scan when time allows,2024-05-15T14:43:14Z +issue,98006,Create a plan on removing conda dependency from CI/CD,https://github.com/pytorch/pytorch/issues/98006,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-05-15T14:41:44Z +issue,85499,Execute smoke test for Better Transformer feature,https://github.com/pytorch/pytorch/issues/85499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-05-15T14:32:28Z +issue,87782,Libtorch windows binaries publishing,https://github.com/pytorch/pytorch/issues/87782,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-05-15T14:32:17Z +issue,107780,Add caffe2 ideep/onednn tests to OSS CI,https://github.com/pytorch/pytorch/issues/107780,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: mkldnn, oncall: releng",scan when time allows,2024-05-15T14:31:19Z +issue,126021,Incorrect output from `torch.histc` when min and max are equal.,https://github.com/pytorch/pytorch/issues/126021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-15T13:50:25Z +issue,126164,Context manager for determinism,https://github.com/pytorch/pytorch/issues/126164,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2024-05-15T12:31:19Z +issue,109017,torch.sparse_coo_tensor argname quirks + [feature request] `.numpy()`/`from_numpy` method for sparse_coo_tensor/sparse_csr_tensor (or maybe name them as `.scipy()`/`.from_scipy()` or at least under some `torch.utils.*` namespace for zero-copy conversions to<>from scipy sparse arrays,https://github.com/pytorch/pytorch/issues/109017,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: scipy compatibility, module: sparse",scan when time allows,2024-05-15T11:41:40Z +issue,105641,Turn indexing with a scalar tensor into an copy into a view and avoid a D2H synchronization.,https://github.com/pytorch/pytorch/issues/105641,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: bc-breaking, module: numpy",scan when time allows,2024-05-14T21:29:03Z +issue,126006,The arguments `tesnor1` and `tensor2` of `result_type()` don't work but `tensor` and `other` work,https://github.com/pytorch/pytorch/issues/126006,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-14T18:26:09Z +issue,126014,The unexpected behavior of `can_cast()`,https://github.com/pytorch/pytorch/issues/126014,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-14T18:25:22Z +issue,126034,A scalar with `tensor` argument of `result_type()` doesn't work,https://github.com/pytorch/pytorch/issues/126034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-14T18:21:33Z +issue,126041,A scalar with `input` argument of `remainder()` doesn't work,https://github.com/pytorch/pytorch/issues/126041,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-14T18:21:02Z +issue,126037,A scalar with `input` argument of `pow()` doesn't work,https://github.com/pytorch/pytorch/issues/126037,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-14T18:20:51Z +issue,126039,A scalar with `input` argument of `float_power()` doesn't work,https://github.com/pytorch/pytorch/issues/126039,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-05-14T18:20:37Z +issue,126190,Centralized Mechanism To Pytorch Query Global Flags,https://github.com/pytorch/pytorch/issues/126190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-05-14T18:08:45Z +issue,126065,[ONNX] `SymbolicValueError` with `onnx::quantized_gather`,https://github.com/pytorch/pytorch/issues/126065,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-14T17:02:08Z +issue,126000,Inform when weights loading promotes types,https://github.com/pytorch/pytorch/issues/126000,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-05-14T16:54:17Z +issue,125940,[NestedTensor] multiply batch and ragged dimension to get shape of values tensor,https://github.com/pytorch/pytorch/issues/125940,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-05-14T04:23:34Z +issue,29115,"crash when call dist.new_group(ranks=local_ranks, backend='gloo')",https://github.com/pytorch/pytorch/issues/29115,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-05-14T03:23:26Z +issue,87501,Synchronize domain builds to be executed after core build have completed,https://github.com/pytorch/pytorch/issues/87501,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-05-14T00:26:25Z +issue,125471,Build failure when using CUDAGraph,https://github.com/pytorch/pytorch/issues/125471,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cuda, module: docs",scan when time allows,2024-05-13T22:24:06Z +issue,125800,"[BE][DTensor] get rid of @with_comms decorator for tests, and use run_test to init pg",https://github.com/pytorch/pytorch/issues/125800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-05-13T20:35:58Z +issue,125895,Support for third-party use of pipeline,https://github.com/pytorch/pytorch/issues/125895,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: privateuse1, oncall: distributed",scan when time allows,2024-05-13T20:35:26Z +issue,109833,Implement Copy-on-write (COW) tensors,https://github.com/pytorch/pytorch/issues/109833,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: copy on write, module: internals, module: viewing and reshaping",scan when time allows,2024-05-13T14:54:37Z +issue,125837,Support for PyTorch on iOS Following PEP 730,https://github.com/pytorch/pytorch/issues/125837,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: ios",scan when time allows,2024-05-11T04:13:17Z +issue,65140,profiler_legacy `_parse_legacy_records` deduplication logic has false-positive.,https://github.com/pytorch/pytorch/issues/65140,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-10T20:40:19Z +issue,124974,[Releng] Triton version for minor releases starting with 2.4,https://github.com/pytorch/pytorch/issues/124974,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2024-05-10T15:47:43Z +issue,111908,Inconsistent Keyword Arguments behaviors in torch.triangular_solve(),https://github.com/pytorch/pytorch/issues/111908,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: python frontend",scan when time allows,2024-05-10T08:53:43Z +issue,7733,Inserting a tensor into a python dict causes strange behavior,https://github.com/pytorch/pytorch/issues/7733,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-05-09T23:06:33Z +issue,125861,`torch.nn.LazyLinear` custom weight initialization,https://github.com/pytorch/pytorch/issues/125861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-05-09T20:47:20Z +issue,104045,"""TorchDynamo Deeper Dive"" doc is missing some part",https://github.com/pytorch/pytorch/issues/104045,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-09T17:00:40Z +issue,125531,`LRScheduler` Refactoring,https://github.com/pytorch/pytorch/issues/125531,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2024-05-09T15:11:59Z +issue,125547,PyTorch model slows down 30x when binding threads with fewer than half the CPU cores.,https://github.com/pytorch/pytorch/issues/125547,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: docs, module: performance",scan when time allows,2024-05-09T15:08:48Z +issue,125502,C++ libtorch x86_64 Darwin deprecation note missing,https://github.com/pytorch/pytorch/issues/125502,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp, module: macos",scan when time allows,2024-05-09T13:07:12Z +issue,125835,"When using torch.onnx.export, the result of torch.scatter_add may be wrong for opset11",https://github.com/pytorch/pytorch/issues/125835,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-09T12:52:45Z +issue,125539,`triton` package that ships with torch is not compatible with `torch.ops.matmul`,https://github.com/pytorch/pytorch/issues/125539,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2024-05-09T12:43:03Z +issue,53095,Reset mask for torch.cumsum?,https://github.com/pytorch/pytorch/issues/53095,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: reductions",scan when time allows,2024-05-09T10:37:02Z +issue,125746,DefaultCPUAllocator: not enough memory in matrix multiplication broadcasting,https://github.com/pytorch/pytorch/issues/125746,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2024-05-09T02:00:30Z +issue,125738,[FSDP2] _sharded_param_data is sitll on meta while sharded_param moved to cuda after calling initialize_parameters(),https://github.com/pytorch/pytorch/issues/125738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2024-05-08T22:55:04Z +issue,125744,Improve oneDNN memory alloction performance for pytorch Windows,https://github.com/pytorch/pytorch/issues/125744,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: windows",scan when time allows,2024-05-08T09:57:43Z +issue,123169,Need better documentation for running doctest locally,https://github.com/pytorch/pytorch/issues/123169,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docs, module: tests",scan when time allows,2024-05-07T16:25:54Z +issue,125581,fx hook failed,https://github.com/pytorch/pytorch/issues/125581,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-05-07T15:54:05Z +issue,50339,Function request: scipy.interpolate.griddata,https://github.com/pytorch/pytorch/issues/50339,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2024-05-07T09:11:51Z +issue,125051,MPS backend thinks that subnormals are equal to zero,https://github.com/pytorch/pytorch/issues/125051,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: mps",scan when time allows,2024-05-07T05:02:31Z +issue,125626,[BE] Autoclose issues that haven't been updated in X months,https://github.com/pytorch/pytorch/issues/125626,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-05-06T22:17:08Z +issue,120534,FLOP Counter does not handle kwargs,https://github.com/pytorch/pytorch/issues/120534,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-05-06T22:13:21Z +issue,112942,Can't build PyTorch 2.1 from source by GCC 13.2 on M1 MacOS,https://github.com/pytorch/pytorch/issues/112942,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs, module: macos",scan when time allows,2024-05-06T20:54:29Z +issue,125620,[BE] DTensor optimizer to utilize OptimizerInfo Constructs,https://github.com/pytorch/pytorch/issues/125620,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-05-06T19:18:56Z +issue,62396,"Make interpolation output size compatible with opencv, scikit-image and scipy for floating scale factor",https://github.com/pytorch/pytorch/issues/62396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2024-05-06T17:03:43Z +issue,125591,PyTorch gradient checkpointing consumes a lot more memory than JAX,https://github.com/pytorch/pytorch/issues/125591,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: memory usage",scan when time allows,2024-05-06T15:40:49Z +issue,104981,NotImplementedError: Could not run 'aten::_spdiags' with arguments from the 'CUDA' backend.,https://github.com/pytorch/pytorch/issues/104981,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2024-05-05T17:52:02Z +issue,106141,Sparse COO indices are torch.Int64 -- is this necessary?,https://github.com/pytorch/pytorch/issues/106141,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-05-04T16:43:39Z +issue,37554,Add name to Class Parameter(),https://github.com/pytorch/pytorch/issues/37554,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rpc, oncall: distributed",scan when time allows,2024-05-04T13:40:54Z +issue,102999,Preserve weight_g/weight_v accessors on new weight_norm,https://github.com/pytorch/pytorch/issues/102999,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: nn.utils.parametrize",scan when time allows,2024-05-04T12:09:51Z +issue,35441,"torch::normal only supports (double, double), but at::normal supports (double, double) / (double, Tensor) / (Tensor, double) / (Tensor, Tensor)",https://github.com/pytorch/pytorch/issues/35441,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-05-04T06:41:18Z +issue,3473,Feature Request: Distributed send arbitrary objects,https://github.com/pytorch/pytorch/issues/3473,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pickle, module: serialization, oncall: distributed",scan when time allows,2024-05-04T00:40:02Z +issue,125493,Error calling to_sparse_coo() on any subclass of Tensor,https://github.com/pytorch/pytorch/issues/125493,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-05-03T20:02:46Z +issue,125390,[typing] Make arguments to `__getitem__`/`__setitem__`/ etc. positional only.,https://github.com/pytorch/pytorch/issues/125390,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2024-05-03T13:11:35Z +issue,124726,Foreach tests should xfail on all dtypes that are not supported,https://github.com/pytorch/pytorch/issues/124726,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: tests",scan when time allows,2024-05-03T08:49:47Z +issue,67240,Feature: Add derivative for channel_shuffle,https://github.com/pytorch/pytorch/issues/67240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2024-05-02T10:37:31Z +issue,59168,[CUDA] Add channels_last_3d support for commonly used modules,https://github.com/pytorch/pytorch/issues/59168,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory format",scan when time allows,2024-05-02T08:05:46Z +issue,108342,ONNX export constant folding messes up with shared weight deduplication,https://github.com/pytorch/pytorch/issues/108342,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-02T00:43:34Z +issue,110131,Some ONNX tests have been disabled because of new tensor.split signature,https://github.com/pytorch/pytorch/issues/110131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-01T22:02:34Z +issue,116965,Support GridSample 5D,https://github.com/pytorch/pytorch/issues/116965,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-01T20:48:38Z +issue,70446,Exporting the operator prim_DictConstruct to ONNX opset version 13 is not supported,https://github.com/pytorch/pytorch/issues/70446,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-05-01T20:11:36Z +issue,19092,[RFC] Memory format (aka layout aka NHWC) support,https://github.com/pytorch/pytorch/issues/19092,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: mkldnn",scan when time allows,2024-05-01T19:39:21Z +issue,21120,[RFC] Adding MKL-DNN Int8 functions to PyTorch/Aten/JIT backend,https://github.com/pytorch/pytorch/issues/21120,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, oncall: jit, oncall: quantization",scan when time allows,2024-05-01T19:39:14Z +issue,105582,RFC: Integrating oneDNN Graph Compiler into Inductor C++/OpenMP Backend for Enhanced Graph Fusion and Performance,https://github.com/pytorch/pytorch/issues/105582,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",scan when time allows,2024-05-01T19:38:57Z +issue,48246,[RFC] XPU device for PyTorch,https://github.com/pytorch/pytorch/issues/48246,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: intel,scan when time allows,2024-05-01T19:37:38Z +issue,94322,RFC: Enabling AVX512 dispatch for compute-intensive ATen ops,https://github.com/pytorch/pytorch/issues/94322,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: intel, module: performance",scan when time allows,2024-05-01T19:37:31Z +issue,114835,[RFC] Building system for SYCL and limited number of SYCL kernels for ATen fallbacks of TorchInductor,https://github.com/pytorch/pytorch/issues/114835,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, module: intel",scan when time allows,2024-05-01T19:36:57Z +issue,125298,Tensorboard add_custom_scalar does not work,https://github.com/pytorch/pytorch/issues/125298,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2024-05-01T14:05:44Z +issue,125146,Race condition in FileTimerServerTest.test_expired_timers,https://github.com/pytorch/pytorch/issues/125146,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: testing, module: tests, oncall: r2p",scan when time allows,2024-05-01T02:51:40Z +issue,119442,pytest test/onnx/test_pytorch_onnx_onnxruntime_cuda.py -k test_embedding_bag_2d_per_sample_weights is failing,https://github.com/pytorch/pytorch/issues/119442,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-04-30T23:47:28Z +issue,120822,Problem differentiating through a torch.layer_norm() call when exporting to torch.jit.trace,https://github.com/pytorch/pytorch/issues/120822,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-04-30T23:15:16Z +issue,124915,Exporting torch slice_scatter to onnx Identity,https://github.com/pytorch/pytorch/issues/124915,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-04-30T21:50:44Z +issue,124523,optim.apply_optimizer_in_backward does not account for gradient accumulation,https://github.com/pytorch/pytorch/issues/124523,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: distributed",scan when time allows,2024-04-30T21:28:18Z +issue,125243,omp.h not found on macOS during install from source at github repo,https://github.com/pytorch/pytorch/issues/125243,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2024-04-30T20:25:29Z +issue,125239,Improved strategy for dealing with deterministically flaky tests which are order sensitive,https://github.com/pytorch/pytorch/issues/125239,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",scan when time allows,2024-04-30T19:41:20Z +issue,125183,torch/_refs/__init__.py is not autoreloadable in bento / jupyter notebook,https://github.com/pytorch/pytorch/issues/125183,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: library,scan when time allows,2024-04-30T19:30:41Z +issue,125116,Does torch.nn.Linear check the weights shape before assignment?,https://github.com/pytorch/pytorch/issues/125116,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-04-30T19:29:27Z +issue,124788,torch.nn.checkpoint.checkpoint ignores default device in backward() call,https://github.com/pytorch/pytorch/issues/124788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2024-04-30T19:13:24Z +issue,75512,Connection closed by peer when using dist.isend in gloo backend,https://github.com/pytorch/pytorch/issues/75512,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-30T17:12:19Z +issue,98208,Add a deterministic version of reflection_pad2d_backward_cuda,https://github.com/pytorch/pytorch/issues/98208,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: nn",scan when time allows,2024-04-30T16:16:13Z +issue,68102,Can not guarantee reproducible for `nn.ReflectionPad2d`,https://github.com/pytorch/pytorch/issues/68102,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility,scan when time allows,2024-04-30T16:16:11Z +issue,28233,torch native functions cannot be used with inspect.signature,https://github.com/pytorch/pytorch/issues/28233,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: language binding, module: pybind",scan when time allows,2024-04-30T10:48:18Z +issue,72582,"Avoid using thrust:: directly, use THRUST_NS_QUALIFIER:: instead",https://github.com/pytorch/pytorch/issues/72582,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2024-04-30T04:04:32Z +issue,75242,"large model, low memory: need `torch.load` that loads one submodule at a time",https://github.com/pytorch/pytorch/issues/75242,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-04-29T22:27:24Z +issue,103354,Add a requirements.txt for windows pip packages,https://github.com/pytorch/pytorch/issues/103354,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-04-29T18:26:30Z +issue,124934,Tensor's storage changes computation outcome for CPU tensors.,https://github.com/pytorch/pytorch/issues/124934,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2024-04-29T13:40:51Z +issue,114040,IndexError: map::at with MPI CUDA collectives,https://github.com/pytorch/pytorch/issues/114040,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2024-04-29T11:05:33Z +issue,53534,[FX] Ability to wrap functions in other modules for symbolic tracing,https://github.com/pytorch/pytorch/issues/53534,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-04-29T09:01:56Z +issue,104516,vec_test_all_types_xxx with dtype c10::complex and c10::complex has failures on division,https://github.com/pytorch/pytorch/issues/104516,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cpu, module: vectorization",scan when time allows,2024-04-29T00:46:25Z +issue,124998,"When I set the operator weights all to 1, the operator weights change as soon as an input is made to the operator",https://github.com/pytorch/pytorch/issues/124998,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-04-27T14:27:07Z +issue,125028,OpInfo testing is inadequate for `nextafter`,https://github.com/pytorch/pytorch/issues/125028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2024-04-26T22:00:41Z +issue,125064,ValueError: weight_norm of 'weight' not found in ParametrizedConvTranspose1d,https://github.com/pytorch/pytorch/issues/125064,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn.utils.parametrize,scan when time allows,2024-04-26T21:58:26Z +issue,124973,Export llama v3 to ONNX,https://github.com/pytorch/pytorch/issues/124973,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-04-26T18:02:08Z +issue,68892,fusion in fx graph mode did not take care of direct attribute access,https://github.com/pytorch/pytorch/issues/68892,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-04-26T14:20:12Z +issue,124262,CUBLAS_STATUS_EXECUTION_FAILED when calling cublasGemmEx,https://github.com/pytorch/pytorch/issues/124262,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2024-04-26T09:17:49Z +issue,124989,[DTensor] Keep track of data-dependent ops and skip _propagate_tensor_meta to go thru fake tensor,https://github.com/pytorch/pytorch/issues/124989,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2024-04-26T00:39:03Z +issue,97856,make tensor data const correct,https://github.com/pytorch/pytorch/issues/97856,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: copy on write, module: internals",scan when time allows,2024-04-26T00:00:26Z +issue,102227,lintrunner should fail on badly formatted docstrings,https://github.com/pytorch/pytorch/issues/102227,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devx, oncall: pt2",scan when time allows,2024-04-25T19:34:14Z +issue,99372,PyTorch 2.0.0 encountered CUDA error: an illegal memory access was encountered,https://github.com/pytorch/pytorch/issues/99372,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multithreading",scan when time allows,2024-04-25T19:26:23Z +issue,124953,PRs with unbalanced quotes can not be merged,https://github.com/pytorch/pytorch/issues/124953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2024-04-25T19:17:52Z +issue,124956,torch.onnx.export generates an incorrect bias shape for certain conv transpose models,https://github.com/pytorch/pytorch/issues/124956,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: quantization",scan when time allows,2024-04-25T19:17:44Z +issue,98724,Conflict between ``torch.func`` transformations and ``torch.jit.trace``,https://github.com/pytorch/pytorch/issues/98724,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-04-25T15:55:11Z +issue,88136,Placing LSTM model on bfloat16 on GPU causes error,https://github.com/pytorch/pytorch/issues/88136,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: rnn",scan when time allows,2024-04-25T07:26:19Z +issue,124838,Forwardpropagation of high-dimensional tensors through the nn.Linear module becomes multiple times slower since pytorch 2.1.0,https://github.com/pytorch/pytorch/issues/124838,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-04-25T03:32:14Z +issue,120989,input.is_sparse() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/120989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-04-24T21:06:41Z +issue,51020,arm64-v8a not compiling due to libpytorch_jni.so,https://github.com/pytorch/pytorch/issues/51020,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: arm, module: build",scan when time allows,2024-04-24T20:08:52Z +issue,116506,[discussion] Route pointwise Conv1d/Conv2d to matmul? (also in eager),https://github.com/pytorch/pytorch/issues/116506,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: mkldnn, module: performance",scan when time allows,2024-04-24T18:39:53Z +issue,124857,Use lintrunner-adapters for our adapters when possible,https://github.com/pytorch/pytorch/issues/124857,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2024-04-24T17:27:47Z +issue,124783,Scatter_add limitation when accumulate beyond 2^24 under float32 precision,https://github.com/pytorch/pytorch/issues/124783,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2024-04-24T16:16:48Z +issue,123835,Package manager install on Nvidia Grace Hopper does not make cuda available,https://github.com/pytorch/pytorch/issues/123835,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: cuda",scan when time allows,2024-04-24T16:04:46Z +issue,124719,"Questions about parameter initialization, especially with torch.bfloat16 precision",https://github.com/pytorch/pytorch/issues/124719,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: initialization, module: nn",scan when time allows,2024-04-24T15:35:10Z +issue,85533,"[functorch] transforms like jacrev, jacfwd, grad, etc don't work with BatchNorm",https://github.com/pytorch/pytorch/issues/85533,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-04-24T13:23:33Z +issue,124456,CudaHostAlloc takes a lot of time during training,https://github.com/pytorch/pytorch/issues/124456,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: dataloader",scan when time allows,2024-04-24T02:31:02Z +issue,124572,Variant of slice that doesn't perform clamp behavior,https://github.com/pytorch/pytorch/issues/124572,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes,scan when time allows,2024-04-23T15:08:44Z +issue,124222,Bot to recomment on high pri issues if no activity,https://github.com/pytorch/pytorch/issues/124222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-04-22T23:24:47Z +issue,124417,ghstack mergeability check seems wrong,https://github.com/pytorch/pytorch/issues/124417,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-04-22T18:07:44Z +issue,124630,TORCH_SHOW_DISPATCH_TRACE should give information about PythonDispatcher,https://github.com/pytorch/pytorch/issues/124630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2024-04-22T17:59:14Z +issue,34279,`torch.norm` is 113x slower than `torch.sqrt(a**2 + b**2)`,https://github.com/pytorch/pytorch/issues/34279,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2024-04-22T10:56:16Z +issue,123528,Support integer parameters in FullyShardedDataParallel,https://github.com/pytorch/pytorch/issues/123528,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-22T10:04:25Z +issue,122592,build libtorch in windows with intel oneapi base 2024 error,https://github.com/pytorch/pytorch/issues/122592,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: windows",scan when time allows,2024-04-22T07:33:40Z +issue,113546,"P0: Improve failure trace back when crashed to identify the cause of a crash and the ranks that the crash, output the real traceback at last.",https://github.com/pytorch/pytorch/issues/113546,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: elastic",scan when time allows,2024-04-21T19:07:09Z +issue,110820,"""Expected a 'mps:0' generator device but found 'cpu'"" using shuffle=True on DataLoader",https://github.com/pytorch/pytorch/issues/110820,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: mps",scan when time allows,2024-04-21T13:32:24Z +issue,116515,mobile model has very different inference result,https://github.com/pytorch/pytorch/issues/116515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,scan when time allows,2024-04-20T16:12:23Z +issue,124465,"Matmul with broadcast makes copy because of reshape to bmm, and it is slow",https://github.com/pytorch/pytorch/issues/124465,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2024-04-19T15:28:30Z +issue,124383,llvm disabled in official builds from 2.1.0 onwards,https://github.com/pytorch/pytorch/issues/124383,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-04-19T15:14:37Z +issue,82843,Tensor operation hangs when used with multiprocessing,https://github.com/pytorch/pytorch/issues/82843,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: multiprocessing",scan when time allows,2024-04-19T13:55:21Z +issue,88775,Batched Random Number Generators,https://github.com/pytorch/pytorch/issues/88775,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2024-04-18T18:04:27Z +issue,124291,vmap over getitem indexing raises a RunTime error,https://github.com/pytorch/pytorch/issues/124291,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2024-04-18T02:53:44Z +issue,111454,[ONNX][dynamo] Failed to export cumsum with dtype=float16,https://github.com/pytorch/pytorch/issues/111454,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: onnx",scan when time allows,2024-04-17T21:22:47Z +issue,123120,TD distributed CUDA Rollout list,https://github.com/pytorch/pytorch/issues/123120,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-04-17T20:36:47Z +issue,124194,Initial state of RNNs: h_0 or hx?,https://github.com/pytorch/pytorch/issues/124194,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2024-04-17T14:20:26Z +issue,34837,[torch.jit.script] Support tensor indexing after an Ellipsis,https://github.com/pytorch/pytorch/issues/34837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-04-17T07:13:52Z +issue,77159,"A somewhat cryptic error message (for newcomers) - ""Cannot re-initialize CUDA in forked subprocess"" - report and suggestion for a possible solution",https://github.com/pytorch/pytorch/issues/77159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-04-17T01:04:16Z +issue,124228,[DeviceMesh] Current cache for submesh result is not thread save,https://github.com/pytorch/pytorch/issues/124228,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-16T22:43:00Z +issue,99143,No documentation to show how to implement aten::view for custom backend,https://github.com/pytorch/pytorch/issues/99143,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: docs",scan when time allows,2024-04-16T16:18:30Z +issue,117961,ReplicatePad2d cuda not support bfloat16,https://github.com/pytorch/pytorch/issues/117961,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bfloat16,scan when time allows,2024-04-16T06:22:58Z +issue,124145,Use online softmax for just softmax by itself,https://github.com/pytorch/pytorch/issues/124145,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: inductor,scan when time allows,2024-04-16T04:44:59Z +issue,110758,"Fused Adamw RuntimeError: params, grads, exp_avgs, and exp_avg_sqs must have same dtype, device, and layout",https://github.com/pytorch/pytorch/issues/110758,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-04-16T04:18:47Z +issue,52291,"""LayerNormKernelImpl"" not implemented for 'Half' - CPU",https://github.com/pytorch/pytorch/issues/52291,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: half, module: nn",scan when time allows,2024-04-15T23:21:06Z +issue,123996,MSE loss triggers ZeroTensor immutable error inside `torch.func`,https://github.com/pytorch/pytorch/issues/123996,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2024-04-15T18:34:35Z +issue,123284,We should walk torch._C and require everything have type annotations,https://github.com/pytorch/pytorch/issues/123284,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: typing",scan when time allows,2024-04-15T17:19:06Z +issue,124047,dlopen: cannot load any more object with static TLS,https://github.com/pytorch/pytorch/issues/124047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-04-15T15:28:16Z +issue,33859,Possibility to support int4 data type,https://github.com/pytorch/pytorch/issues/33859,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-04-15T10:07:15Z +issue,499,Feature Request: Locally Connected Layer,https://github.com/pytorch/pytorch/issues/499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-04-15T09:24:01Z +issue,115152,scatter_ op after permute op not working correctly on mps device,https://github.com/pytorch/pytorch/issues/115152,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: scatter & gather ops",scan when time allows,2024-04-12T02:34:51Z +issue,123765,Accurately typed ValueRanges / convenient OpsHandler dtype inference,https://github.com/pytorch/pytorch/issues/123765,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamic shapes,scan when time allows,2024-04-11T17:53:05Z +issue,1249,Dice Loss PR,https://github.com/pytorch/pytorch/issues/1249,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2024-04-11T16:28:24Z +issue,35472,Multiprocessing map gets stuck if doing inference on loaded model,https://github.com/pytorch/pytorch/issues/35472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-04-11T09:19:15Z +issue,123421,nonzero with dtensor hang,https://github.com/pytorch/pytorch/issues/123421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-11T05:22:30Z +issue,105299,Support ONNX opset 20 to export GELU to one single op,https://github.com/pytorch/pytorch/issues/105299,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-04-10T20:59:39Z +issue,116120,bfloat16 on 'grid_sampler_3d_cuda' not implemented,https://github.com/pytorch/pytorch/issues/116120,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bfloat16,scan when time allows,2024-04-10T18:33:04Z +issue,58267,Pytorch build issues on PowerPC,https://github.com/pytorch/pytorch/issues/58267,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2024-04-10T15:52:41Z +issue,37488,undefined reference to pthreadpool_compute*,https://github.com/pytorch/pytorch/issues/37488,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: xnnpack",scan when time allows,2024-04-10T12:35:17Z +issue,123533,Memory allocation for Conv2d,https://github.com/pytorch/pytorch/issues/123533,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda",scan when time allows,2024-04-10T11:23:38Z +issue,123506,[RFE] Disallow submodule updates without successful minimal binary builds,https://github.com/pytorch/pytorch/issues/123506,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",scan when time allows,2024-04-10T11:22:22Z +issue,123702,Issue with Vulkan Device for nn.Conv2d Operation,https://github.com/pytorch/pytorch/issues/123702,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2024-04-10T11:12:08Z +issue,72791,"why ram memory surges while loading model, with change in torch load device from CPU to GPU",https://github.com/pytorch/pytorch/issues/72791,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2024-04-10T03:02:32Z +issue,8069,Document torch.acos() behavior near -1 and 1,https://github.com/pytorch/pytorch/issues/8069,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-04-09T20:56:16Z +issue,123245,torch.overrides.get_overridable_functions list functions that are not overridable via __torch_function__,https://github.com/pytorch/pytorch/issues/123245,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2024-04-09T10:56:15Z +issue,45139,Mention accessor/data_ptr for raw memory access in Libtorch index API document and discuss performance implications,https://github.com/pytorch/pytorch/issues/45139,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: performance",scan when time allows,2024-04-09T04:49:54Z +issue,123604,"Can't initialize NVML, H100 gpus on ubuntu",https://github.com/pytorch/pytorch/issues/123604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-04-09T02:30:33Z +issue,119181,[inductor][cpu]Background_Matting fP32 static default wrapper multiple threads performance regression,https://github.com/pytorch/pytorch/issues/119181,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor,scan when time allows,2024-04-09T00:00:58Z +issue,123500,"Cannot add 2 sparse_csc_tensor directly, raising RuntimeError",https://github.com/pytorch/pytorch/issues/123500,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-04-08T22:19:46Z +issue,123466,Undefined behavior when an OpOverload's Library is deleted but the OpOverload continues to be used,https://github.com/pytorch/pytorch/issues/123466,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-04-08T22:19:02Z +issue,22400,Sparse allreduce for ProcessGroupNCCL,https://github.com/pytorch/pytorch/issues/22400,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-07T07:22:00Z +issue,102936,torch.cuda.is_available() returns False on GTX 1650 with cuda 11.7 and torch==2.0.0+cpu,https://github.com/pytorch/pytorch/issues/102936,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-04-06T20:19:33Z +issue,113600,Segmentation fault in RPC worker when DataLoader has num_workers > 0,https://github.com/pytorch/pytorch/issues/113600,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",scan when time allows,2024-04-06T07:02:16Z +issue,66707,`layer_norm` needs to be done in fp32 for fp16 inputs,https://github.com/pytorch/pytorch/issues/66707,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: norms and normalization, module: numerical-stability",scan when time allows,2024-04-06T04:17:18Z +issue,121960,"RuntimeError: ""max_pool2d_channels_last"" not implemented for 'Half'",https://github.com/pytorch/pytorch/issues/121960,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: memory format, module: mkldnn, module: nn",scan when time allows,2024-04-05T17:15:49Z +issue,118478,Cannot formatted print at::Tensor in gdb,https://github.com/pytorch/pytorch/issues/118478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-04-05T17:14:27Z +issue,55902,Warn users when cross entropy is called after softmax,https://github.com/pytorch/pytorch/issues/55902,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-04-05T10:59:17Z +issue,103089,"[OOM] Unable to convert 30B model to ONNX, using 4x A100's",https://github.com/pytorch/pytorch/issues/103089,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-04-04T23:00:14Z +issue,123344,document torch.ops.import_module on the website; also make sure to use this in tutorials,https://github.com/pytorch/pytorch/issues/123344,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-04-04T17:09:21Z +issue,123262,mmap error - no such device (19),https://github.com/pytorch/pytorch/issues/123262,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-04-04T08:13:51Z +issue,111704,Add more flexibility on print / output console,https://github.com/pytorch/pytorch/issues/111704,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-04-04T04:19:32Z +issue,121650,add scatter to functional collectives,https://github.com/pytorch/pytorch/issues/121650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-03T23:47:20Z +issue,103805,Error when building with USE_TENSORRT=1,https://github.com/pytorch/pytorch/issues/103805,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-04-03T16:25:49Z +issue,123157,Add support for CUBLAS_COMPUTE_16F for GEMM opeartions in cudaBLAS,https://github.com/pytorch/pytorch/issues/123157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2024-04-02T20:26:47Z +issue,120820,Problems differentiating through certain operations when exporting to TorchScript,https://github.com/pytorch/pytorch/issues/120820,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-04-02T17:24:22Z +issue,123172,`torch.utils.data.Sampler`: constructor's `data_source` argument should be removed,https://github.com/pytorch/pytorch/issues/123172,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: deprecation",scan when time allows,2024-04-02T16:02:07Z +issue,120139,"On `Dataset`, `IterableDataset` inheritance",https://github.com/pytorch/pytorch/issues/120139,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2024-04-02T16:00:57Z +issue,122410,On Lack of Default `__len__` in Python Abstract Base Classes,https://github.com/pytorch/pytorch/issues/122410,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: docs",scan when time allows,2024-04-02T15:56:14Z +issue,27421,[dataloader] Sampler abstract constructor API minor proposal,https://github.com/pytorch/pytorch/issues/27421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-04-02T15:51:09Z +issue,57278,Attributes consistency in Sampler and DistributedSampler,https://github.com/pytorch/pytorch/issues/57278,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-02T15:50:28Z +issue,121797,Can’t save SparseTensor for custom device,https://github.com/pytorch/pytorch/issues/121797,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-04-02T12:35:18Z +issue,91950,RuntimeError: philox_cuda_state for an unexpected CUDA generator used during capture,https://github.com/pytorch/pytorch/issues/91950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: random",scan when time allows,2024-04-02T10:01:11Z +issue,115138,ImportError: /lib64/libstdc++.so.6: version `GLIBCXX_3.4.20' not found,https://github.com/pytorch/pytorch/issues/115138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-04-02T09:25:14Z +issue,123091,build error: memory_desc.cpp:716:1: internal compiler error: Segmentation fault,https://github.com/pytorch/pytorch/issues/123091,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn, module: third_party",scan when time allows,2024-04-02T09:01:45Z +issue,97990,Relax version dependencies on CUDA pip wheels?,https://github.com/pytorch/pytorch/issues/97990,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2024-04-02T06:31:45Z +issue,123082,Torch.cdist calculation problem,https://github.com/pytorch/pytorch/issues/123082,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-04-01T16:23:18Z +issue,10996,Multiprocess Deadlock when using np.transpose and torch.stack,https://github.com/pytorch/pytorch/issues/10996,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-04-01T16:20:49Z +issue,123075,LibTorch Release version 2.2.2+cpu loads traced model only in Debug configuration,https://github.com/pytorch/pytorch/issues/123075,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-04-01T16:02:07Z +issue,122756,How to reduce memory usage for large matrix calculations？,https://github.com/pytorch/pytorch/issues/122756,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-04-01T15:59:16Z +issue,123070,CUDA out of memory still exist after using FSDP,https://github.com/pytorch/pytorch/issues/123070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-04-01T15:57:02Z +issue,91353,Perf reduction due to munmap with dataloader pinning thread ?,https://github.com/pytorch/pytorch/issues/91353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-04-01T11:42:11Z +issue,98002,DataLoader with collate_fn that returns tensors in GPU memory raises warnings when deleted,https://github.com/pytorch/pytorch/issues/98002,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-03-30T23:52:16Z +issue,100637,Optimal Batch Size Selection in Torchdynamo Benchmarks for Different GPU Memory Sizes,https://github.com/pytorch/pytorch/issues/100637,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-03-29T23:54:13Z +issue,89417,OSError: libcublas.so.11: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/89417,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-03-29T13:30:52Z +issue,65156,[feature request] Jagged / padding version of torch.stack / torch.cat + some general nested tensor discussion,https://github.com/pytorch/pytorch/issues/65156,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: padding, module: sparse",scan when time allows,2024-03-29T03:32:36Z +issue,67760,Add an LRScheduler interface for torch schedulers.,https://github.com/pytorch/pytorch/issues/67760,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2024-03-29T02:24:40Z +issue,72106,Update MKL version used with MAGMA,https://github.com/pytorch/pytorch/issues/72106,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: magma",scan when time allows,2024-03-28T19:12:43Z +issue,122814,torchdim as torch_function subclass exploration,https://github.com/pytorch/pytorch/issues/122814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2024-03-28T04:41:56Z +issue,122699,Bug of `svd_lowrank` when passing parameter `M`,https://github.com/pytorch/pytorch/issues/122699,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2024-03-28T04:32:44Z +issue,76510,"Docs source link broken for all methods defined on Tensor in _tensor.py, e.g. `Tensor.register_hook()`",https://github.com/pytorch/pytorch/issues/76510,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-03-27T22:27:59Z +issue,60149,backward compatibility - need a way to find out when a certain API was added/modified/etc.,https://github.com/pytorch/pytorch/issues/60149,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: deprecation, module: doc infra, module: docs",scan when time allows,2024-03-27T19:35:12Z +issue,122803,Enabling `swap_tensors` path by default for all tensor subclasses errors on some transformer tests,https://github.com/pytorch/pytorch/issues/122803,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-03-27T17:44:45Z +issue,106121,Got error when train models with more than one param_group in torch2.0,https://github.com/pytorch/pytorch/issues/106121,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-03-27T17:44:42Z +issue,103483,torch._dynamo.exc.Unsupported: Tensor.backward with aten_graph=True,https://github.com/pytorch/pytorch/issues/103483,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: export, oncall: pt2",scan when time allows,2024-03-27T17:34:13Z +issue,122081,[Feature Request] add a new argument `storage_offset=None` to `empty_strided`,https://github.com/pytorch/pytorch/issues/122081,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2024-03-27T14:25:58Z +issue,91309,iSTFT produces RuntimeError with center=False and Blackman/Bartlett/Hann windows,https://github.com/pytorch/pytorch/issues/91309,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2024-03-27T10:23:16Z +issue,95708,Dynamo + MacOS: fatal error: 'omp.h' file not found,https://github.com/pytorch/pytorch/issues/95708,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, oncall: cpu inductor, oncall: pt2",scan when time allows,2024-03-27T04:18:17Z +issue,67979,Feature Request: Support prelu_cuda for BFloat16,https://github.com/pytorch/pytorch/issues/67979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: bootcamp, module: cuda, module: nn",scan when time allows,2024-03-27T00:18:49Z +issue,122467,RuntimeError: out_ptr == out_accessor[thread_count_nonzero[tid + 1]].data() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/122467,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: multiprocessing, module: sparse",scan when time allows,2024-03-26T18:13:29Z +issue,122170,torch.cumsum() super slow on 2d input,https://github.com/pytorch/pytorch/issues/122170,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2024-03-26T17:03:25Z +issue,122552,Could not run 'aten::empty_strided' with arguments from the 'CUDA' backend.,https://github.com/pytorch/pytorch/issues/122552,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-03-26T15:52:52Z +issue,77821,cannot convert to channels last format for conv2d conv3d hybrid model,https://github.com/pytorch/pytorch/issues/77821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2024-03-25T20:00:52Z +issue,122571,some official docker images has mismatch nccl version with pytorch,https://github.com/pytorch/pytorch/issues/122571,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docker",scan when time allows,2024-03-25T15:19:09Z +issue,122287,Tolerance setting for test_modules.py -- TestModuleCUDA.test_cpu_cuda_parity,https://github.com/pytorch/pytorch/issues/122287,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: testing",scan when time allows,2024-03-25T08:53:51Z +issue,112575,"RuntimeError: ""grid_sampler_2d_cuda"" not implemented for 'BFloat16'",https://github.com/pytorch/pytorch/issues/112575,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda",scan when time allows,2024-03-23T09:51:12Z +issue,109850,torch._export has no logging,https://github.com/pytorch/pytorch/issues/109850,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: logging, oncall: export, oncall: pt2",scan when time allows,2024-03-23T03:11:00Z +issue,93719,Use bytecode rewriting to offer a __future__ feature,https://github.com/pytorch/pytorch/issues/93719,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-03-22T16:38:18Z +issue,54147,Implement Truly Parallel Ensemble Layers,https://github.com/pytorch/pytorch/issues/54147,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: nn",scan when time allows,2024-03-22T13:15:40Z +issue,122359,Onnx export memory leak with qlora+deepspeed stage 2+ORT,https://github.com/pytorch/pytorch/issues/122359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-03-22T02:37:38Z +issue,93482,[inductor] Add more matmul configurations to `TORCHINDUCTOR_MAX_AUTOTUNE=1` mode,https://github.com/pytorch/pytorch/issues/93482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-03-22T02:23:20Z +issue,117930,C++ Tensor.clone() changes max threads in OpenMP,https://github.com/pytorch/pytorch/issues/117930,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: multithreading",scan when time allows,2024-03-21T19:55:01Z +issue,95785,Fully quantized model (`torch.quantization.convert`) produces incorrect output compared to analytical solution,https://github.com/pytorch/pytorch/issues/95785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-03-21T01:49:47Z +issue,122107,Can't export an onnx with dynamic axes if it contains backpropagation,https://github.com/pytorch/pytorch/issues/122107,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-03-20T18:45:03Z +issue,120879,FSDP: unsharded gradients accessor,https://github.com/pytorch/pytorch/issues/120879,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2024-03-20T17:31:59Z +issue,77537,batch Kronecker product,https://github.com/pytorch/pytorch/issues/77537,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-03-20T04:04:54Z +issue,122187,CuDNN version for pre-built package conflicts with the pip installed version (pytorch 2.2.1 + cuda 12.1),https://github.com/pytorch/pytorch/issues/122187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-03-19T22:26:27Z +issue,122186,Deep-copied model cannot be sent through multiprocessing `Pipe`,https://github.com/pytorch/pytorch/issues/122186,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-03-19T22:09:28Z +issue,122185,Numerical issues on positive semidefiniteness constraint,https://github.com/pytorch/pytorch/issues/122185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-03-19T22:07:40Z +issue,82397,Bilinear interpolation with antialiasing is slow in performance,https://github.com/pytorch/pytorch/issues/82397,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2024-03-19T21:06:26Z +issue,121842,gradcheck fails when inputs are dependent on each other,https://github.com/pytorch/pytorch/issues/121842,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2024-03-19T15:02:02Z +issue,32097,logsumexp with subtraction,https://github.com/pytorch/pytorch/issues/32097,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2024-03-19T13:48:01Z +issue,121926,"Call `torch.pow(int_tensor, torch.tensor(-1))` does not raise error like NumPy.",https://github.com/pytorch/pytorch/issues/121926,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2024-03-19T13:14:48Z +issue,48163,nn.functional.interpolate backward in fp16 is extremely slow,https://github.com/pytorch/pytorch/issues/48163,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half, module: interpolation, module: nn",scan when time allows,2024-03-19T08:42:34Z +issue,121917,"[PythonDispatch] when the value of a argument is the same as its default value, __torch_dispatch fails to capture it.",https://github.com/pytorch/pytorch/issues/121917,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2024-03-19T03:07:14Z +issue,115706,[export][functionalize] Complicated subgraph from functionalizing inplace op on sliced tensors,https://github.com/pytorch/pytorch/issues/115706,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2024-03-19T03:04:54Z +issue,121719,index_copy change value when use_deterministic (true),https://github.com/pytorch/pytorch/issues/121719,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: boolean tensor, module: cuda, module: determinism",scan when time allows,2024-03-19T02:26:05Z +issue,122017,"Got ""KeyError: 'classname'"" while runing pytorch unit tests",https://github.com/pytorch/pytorch/issues/122017,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2024-03-19T01:47:02Z +issue,122080,graph.owning_module does not point to the correct module after dataparallel forwarding.,https://github.com/pytorch/pytorch/issues/122080,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2024-03-19T01:35:34Z +issue,122021,Wishart: problems with singular samples.,https://github.com/pytorch/pytorch/issues/122021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-03-18T21:52:49Z +issue,116478,The Python module installs cmake files and headers under the Python's site-packages directory that can't be used,https://github.com/pytorch/pytorch/issues/116478,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2024-03-18T12:10:04Z +issue,6934,Feature request: SSIM/MS-SSIM,https://github.com/pytorch/pytorch/issues/6934,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2024-03-17T19:13:37Z +issue,23850,[quantization] jit::class_ for packed weights,https://github.com/pytorch/pytorch/issues/23850,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-03-17T19:08:30Z +issue,87559,diagonal of Jacobian matrix,https://github.com/pytorch/pytorch/issues/87559,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2024-03-17T11:45:34Z +issue,122027,PyTorch fails to build with mkl if cmake is installed from pip,https://github.com/pytorch/pytorch/issues/122027,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: mkl",scan when time allows,2024-03-16T18:47:13Z +issue,117236,```torch.linalg.lstsq```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117236,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2024-03-16T10:39:41Z +issue,117242,```torch.tensordot```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117242,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-03-16T08:27:40Z +issue,121737,Unable to install pytorch from source for version 2.0.0,https://github.com/pytorch/pytorch/issues/121737,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-03-16T05:43:46Z +issue,1329,[proposed feature] Eve: Improving Stochastic Gradient Descent with Feedback,https://github.com/pytorch/pytorch/issues/1329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-03-16T01:10:04Z +issue,121715,Depthwise conv2d op produces nan values with non-nan input,https://github.com/pytorch/pytorch/issues/121715,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: edge cases, module: nans and infs, module: nn",scan when time allows,2024-03-15T18:46:33Z +issue,12760,[feature request] Spectral norm support in torch.norm or factor out Power Iteration from spectralnormalization in some other place and orthogonalization from PowerSGD hook,https://github.com/pytorch/pytorch/issues/12760,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2024-03-15T18:43:32Z +issue,108432,Wrong result of first run with torch.compile() when backend is using torch.jit.trace() and model has inplace operators,https://github.com/pytorch/pytorch/issues/108432,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-03-15T09:23:42Z +issue,106991,Add cutlass as an alternative backend of PT2 Inductor,https://github.com/pytorch/pytorch/issues/106991,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-03-15T08:50:32Z +issue,67682,Feature request: bfloat16 support for CUDA matmuls,https://github.com/pytorch/pytorch/issues/67682,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda",scan when time allows,2024-03-14T23:29:28Z +issue,65510,nn.conv1d padding='same',https://github.com/pytorch/pytorch/issues/65510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: docs",scan when time allows,2024-03-14T13:37:23Z +issue,121789,Precision Inconsistent for `half` type compared with NumPy and TensorFlow,https://github.com/pytorch/pytorch/issues/121789,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: numpy",scan when time allows,2024-03-14T06:52:45Z +issue,105281,Optimize PyTorch C++ part with Profile-Guided Optimization (PGO),https://github.com/pytorch/pytorch/issues/105281,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: performance",scan when time allows,2024-03-14T02:35:28Z +issue,105092,[RFC] Proposal to upgrade LLVM version,https://github.com/pytorch/pytorch/issues/105092,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-03-14T02:24:41Z +issue,104531,[RFC] Optional Modular Representation for FX Graph from `torch.compile()`,https://github.com/pytorch/pytorch/issues/104531,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: pt2,scan when time allows,2024-03-14T01:51:44Z +issue,121683,linalg.ldl_factor raises an error with zero tensor,https://github.com/pytorch/pytorch/issues/121683,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: mkl",scan when time allows,2024-03-13T21:20:26Z +issue,95225,Remove conda virtualenv from the docker image,https://github.com/pytorch/pytorch/issues/95225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docker",scan when time allows,2024-03-13T13:14:11Z +issue,77646,Werror=nonnull in dataloader.cpp (part of tests),https://github.com/pytorch/pytorch/issues/77646,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-03-13T09:08:08Z +issue,102360,PyTorch hangs at import when used together with TensorFlow,https://github.com/pytorch/pytorch/issues/102360,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-03-13T00:09:11Z +issue,71028,Skip LSTM quantization by default in get_default_qconfig_dict and get_default_qat_qconfig_dict,https://github.com/pytorch/pytorch/issues/71028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-03-12T18:52:36Z +issue,80874,C++ extensions inject a bunch of compilation flags,https://github.com/pytorch/pytorch/issues/80874,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp-extensions",scan when time allows,2024-03-12T13:02:46Z +issue,117233,Submit a bug when trying to assign arg 'out=' in api torch._foreach_add_,https://github.com/pytorch/pytorch/issues/117233,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: python frontend",scan when time allows,2024-03-12T05:53:05Z +issue,121261,Convolution and norm with vmap jacobian failing,https://github.com/pytorch/pytorch/issues/121261,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: forward ad, module: functorch, module: norms and normalization",scan when time allows,2024-03-11T22:28:31Z +issue,121596,Memory Usage of Conv2d Rapidly Increases when Params >1G,https://github.com/pytorch/pytorch/issues/121596,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: memory usage",scan when time allows,2024-03-11T18:48:37Z +issue,52439,[proposal] Way to export/display the convolution algorithm found with cudnn.benchmark = True + allow to override cudnn algo choice with conv/matmul module/function algo/hint arguments,https://github.com/pytorch/pytorch/issues/52439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2024-03-11T18:23:53Z +issue,121588,logistic sigmoid function outputs NaN indeterminate on complex input,https://github.com/pytorch/pytorch/issues/121588,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cpu, module: nans and infs",scan when time allows,2024-03-11T16:18:00Z +issue,121629,Pad operator produces non-nan value with nan input,https://github.com/pytorch/pytorch/issues/121629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-03-11T13:51:53Z +issue,50820,NCCL_BLOCKING_WAIT=1 makes training extremely slow (but if not set then OOM on one device will hang training),https://github.com/pytorch/pytorch/issues/50820,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-03-11T12:10:46Z +issue,114945,torch.multinomial - Unexpected (incorrect) results when replacement=True in version 2.1.1+cpu,https://github.com/pytorch/pytorch/issues/114945,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2024-03-10T17:01:26Z +issue,75391,Pytorch linalg test failure with cuda 11.6,https://github.com/pytorch/pytorch/issues/75391,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2024-03-09T23:35:51Z +issue,121430,Getting a CUDA tensor through a Queue from within the same process causes a segfault,https://github.com/pytorch/pytorch/issues/121430,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-03-08T16:49:45Z +issue,121389,`torch.gather` backward performance extremely low on `torch.bfloat16`,https://github.com/pytorch/pytorch/issues/121389,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: performance, module: scatter & gather ops",scan when time allows,2024-03-08T16:49:23Z +issue,65418,Sending tensors from short-lived multiprocessing process fails on Linux,https://github.com/pytorch/pytorch/issues/65418,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-03-08T07:50:04Z +issue,119873,Different numerical results during training with CUDA graphs,https://github.com/pytorch/pytorch/issues/119873,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: numerical-reproducibility",scan when time allows,2024-03-07T16:55:23Z +issue,121131,Add flags for device that does not support fused or foreach ops,https://github.com/pytorch/pytorch/issues/121131,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: mta",scan when time allows,2024-03-06T17:44:16Z +issue,121308,"installed pycuda, loss in pytorch has bug",https://github.com/pytorch/pytorch/issues/121308,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: protobuf",scan when time allows,2024-03-06T17:42:37Z +issue,121226,input_data.to produce inconsistent results,https://github.com/pytorch/pytorch/issues/121226,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: python frontend",scan when time allows,2024-03-06T17:42:09Z +issue,121271,`ReduceLROnPlateau` does not call `super().__init__()`,https://github.com/pytorch/pytorch/issues/121271,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2024-03-06T15:41:26Z +issue,121203,How to clear GPU memory without restarting kernel when using a PyTorch model,https://github.com/pytorch/pytorch/issues/121203,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-03-06T15:21:20Z +issue,120708,ELU quantization parameters (scale + zp) missing from model's state_dict,https://github.com/pytorch/pytorch/issues/120708,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: serialization, oncall: quantization",scan when time allows,2024-03-06T11:24:41Z +issue,108739,"DDP Elastic ""master_addr"" resolution error in environment variables.",https://github.com/pytorch/pytorch/issues/108739,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2024-03-06T08:39:54Z +issue,82886,CUDA graph capturing fails for nn.Embedding and large batch sizes,https://github.com/pytorch/pytorch/issues/82886,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs, module: embedding",scan when time allows,2024-03-06T05:11:38Z +issue,89254,"`torch.nn.ReplicationPad2D` Report ""invalid configuration argument"" Error under Compute Sanitizer",https://github.com/pytorch/pytorch/issues/89254,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn, module: padding",scan when time allows,2024-03-06T01:51:58Z +issue,121215,"Build torch From Source is failed on Mac, maybe caused by protobuf",https://github.com/pytorch/pytorch/issues/121215,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2024-03-06T00:46:38Z +issue,71631,Wrapping make_graphed_callables with autocast issue,https://github.com/pytorch/pytorch/issues/71631,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: cuda graphs",scan when time allows,2024-03-05T16:59:50Z +issue,121090,Pytorch nightly docker images set PYTORCH_VERSION variable that doesn't match the installed torch version,https://github.com/pytorch/pytorch/issues/121090,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docker",scan when time allows,2024-03-05T15:13:47Z +issue,115331,"Please increase ""Number of CUDA devices"" recognized by pytorch.",https://github.com/pytorch/pytorch/issues/115331,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-03-05T11:46:46Z +issue,103652,inductor: support horizontal reduction with vec_transpose to improve TIMM swin_base_patch4_window7_224 dynamic shape performance,https://github.com/pytorch/pytorch/issues/103652,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: cpu inductor, oncall: pt2",scan when time allows,2024-03-05T07:02:53Z +issue,120930,make_graphed_callables doesn't turn off autocast for backward,https://github.com/pytorch/pytorch/issues/120930,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda graphs",scan when time allows,2024-03-04T20:16:31Z +issue,107433,No checks when running torch.nn.functional.ctc_loss with bogus inputs,https://github.com/pytorch/pytorch/issues/107433,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: loss",scan when time allows,2024-03-04T08:51:46Z +issue,62066,Memory Leak Found in Persistent DataLoader,https://github.com/pytorch/pytorch/issues/62066,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: memory usage",scan when time allows,2024-03-03T23:33:32Z +issue,13505,[feature request] Singular values and spectral norm for convolutional layers,https://github.com/pytorch/pytorch/issues/13505,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: fft, module: linear algebra",scan when time allows,2024-03-03T02:40:22Z +issue,120994,OOM with mixed precision with OPT1.3B Decoder,https://github.com/pytorch/pytorch/issues/120994,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2024-03-01T17:18:29Z +issue,120973,TORCH_CHECK level changes in flatten,https://github.com/pytorch/pytorch/issues/120973,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: viewing and reshaping",scan when time allows,2024-03-01T16:59:44Z +issue,108341,`adaptive_avg_pool2d_backward_cuda` is not deterministic,https://github.com/pytorch/pytorch/issues/108341,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2024-03-01T16:53:25Z +issue,67324,torch.stft - fill_cuda not implemented for ComplexHalf,https://github.com/pytorch/pytorch/issues/67324,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cuda, module: half",scan when time allows,2024-03-01T09:51:58Z +issue,72766,Feature Request: Deterministic MaxPool3d and AvgPool3d,https://github.com/pytorch/pytorch/issues/72766,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: nn, module: pooling",scan when time allows,2024-02-29T16:00:55Z +issue,120847,[C10D] DeviceMesh + _set_default_timeout doesn't work,https://github.com/pytorch/pytorch/issues/120847,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-02-28T23:16:23Z +issue,120689,Tracing UpSample get RuntimeError when use_determinitic_algorithms=True,https://github.com/pytorch/pytorch/issues/120689,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, oncall: jit",scan when time allows,2024-02-27T21:44:27Z +issue,54393,[docs] torch.optim.lr_scheduler,https://github.com/pytorch/pytorch/issues/54393,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2024-02-27T21:21:32Z +issue,44593,Docs include the verbose arg of `CosineAnnealingWarmRestarts` but the actual docstring in code doesn't,https://github.com/pytorch/pytorch/issues/44593,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2024-02-27T20:42:01Z +issue,120572,FCD and tensor subclasses,https://github.com/pytorch/pytorch/issues/120572,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: first class dims, module: functorch",scan when time allows,2024-02-27T15:33:55Z +issue,82073,CUDACachingAllocator should be cuda memory merge/compact friendly,https://github.com/pytorch/pytorch/issues/82073,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2024-02-27T12:40:47Z +issue,120209,`torch.put_along_dim` and `torch.put_along_dim_` like `np.put_along_axis`,https://github.com/pytorch/pytorch/issues/120209,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python frontend",scan when time allows,2024-02-27T09:48:29Z +issue,110029,Dataloader resetting with num_workers=1 and persistent_workers=True,https://github.com/pytorch/pytorch/issues/110029,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-02-27T09:40:33Z +issue,120291,Cannot download older version of libtorch ( libtorch-macos-1.9.0.zip ),https://github.com/pytorch/pytorch/issues/120291,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-02-27T00:21:13Z +issue,120466,faster communication,https://github.com/pytorch/pytorch/issues/120466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-02-26T21:50:52Z +issue,117099,torch-1.11 Cross-Compiling for QNX 7.1.0 with aarch64le issue: no matching function for call to 'isnan(c10::BFloat16&)',https://github.com/pytorch/pytorch/issues/117099,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-02-26T16:29:29Z +issue,119190,TensorList Python binding triggering C++ TypeError,https://github.com/pytorch/pytorch/issues/119190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2024-02-25T03:54:30Z +issue,120463,Make upsample_bicubic2d_out_frame support BFloat16,https://github.com/pytorch/pytorch/issues/120463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bfloat16,scan when time allows,2024-02-24T04:53:15Z +issue,120541,[c10d] distributed.breakpoint fallback when pg is not inited,https://github.com/pytorch/pytorch/issues/120541,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-24T00:11:52Z +issue,120354,Performance degradation for `aten::fill_` and `aten::add_` in backward pass,https://github.com/pytorch/pytorch/issues/120354,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2024-02-23T21:33:58Z +issue,60548,Ensure warnings relate to user code with stacklevel,https://github.com/pytorch/pytorch/issues/60548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deprecation,scan when time allows,2024-02-23T18:48:04Z +issue,113370,Incorrect stride when permuting shapes where a zero dimension is present.,https://github.com/pytorch/pytorch/issues/113370,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: empty tensor",scan when time allows,2024-02-23T18:06:34Z +issue,39656,Broadcasting for torch.cross,https://github.com/pytorch/pytorch/issues/39656,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python array api",scan when time allows,2024-02-22T23:16:11Z +issue,120396,Reinitialize parameters in TransformerEncoder,https://github.com/pytorch/pytorch/issues/120396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-02-22T15:10:07Z +issue,120319,The CUDA download should include cusparseLt.dll since xformers now wants to use it in the context of torch,https://github.com/pytorch/pytorch/issues/120319,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: windows",scan when time allows,2024-02-22T15:01:05Z +issue,55865,Don't define __assert_fail on systems using `musl-dev` headers,https://github.com/pytorch/pytorch/issues/55865,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-02-21T23:32:49Z +issue,120255,Add documention to the AT_DISPATCH functionality,https://github.com/pytorch/pytorch/issues/120255,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-02-21T16:48:43Z +issue,120237,FFT operations on MPS are less accurate than on CPU,https://github.com/pytorch/pytorch/issues/120237,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: mps",scan when time allows,2024-02-20T22:08:19Z +issue,120000,Update Protobuf submodule to v3.21.X,https://github.com/pytorch/pytorch/issues/120000,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: protobuf",scan when time allows,2024-02-20T21:18:02Z +issue,11340,Better user experience for using Generator object,https://github.com/pytorch/pytorch/issues/11340,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: random",scan when time allows,2024-02-19T01:34:58Z +issue,106408,PyTorchMPS not showing up in Instruments for `torch.mps.profiler`,https://github.com/pytorch/pytorch/issues/106408,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2024-02-17T19:52:12Z +issue,118580,[NestedTensor] Support for operations found commonly in point cloud analysis architectures,https://github.com/pytorch/pytorch/issues/118580,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-02-17T17:53:40Z +issue,44878,"[ONNX] Support Tuple(List(Tensors),...) in input/ouput",https://github.com/pytorch/pytorch/issues/44878,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2024-02-17T00:06:12Z +issue,120104,[memory profiler] keep consistent MB/MiB between y-axis and CUDACachingAllocator size,https://github.com/pytorch/pytorch/issues/120104,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-16T19:55:40Z +issue,119714,[NestedSubTensor] add communication operators support,https://github.com/pytorch/pytorch/issues/119714,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, module: nestedtensor, oncall: distributed",scan when time allows,2024-02-16T19:00:20Z +issue,117146,`test_triton_bsr_scatter_mm` fails on V100,https://github.com/pytorch/pytorch/issues/117146,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-02-16T17:44:23Z +issue,120085,Collate function in libtorch c++ API,https://github.com/pytorch/pytorch/issues/120085,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: dataloader",scan when time allows,2024-02-16T17:08:10Z +issue,78720,ValueError during `yaml.dump(dtype)`,https://github.com/pytorch/pytorch/issues/78720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-02-16T12:35:17Z +issue,120037,[FSDP][BE] print kwargs in run_subtests,https://github.com/pytorch/pytorch/issues/120037,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-16T01:14:24Z +issue,119452,"RuntimeError: !block->expandable_segment_ INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":2814, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/119452,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2024-02-15T22:07:16Z +issue,120002,[c10d] torch.distributed.breakpoint stuck with multi-thread process group,https://github.com/pytorch/pytorch/issues/120002,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-15T18:26:30Z +issue,24089,torch.nn.functional.grid_sample with 'circular' border conditions,https://github.com/pytorch/pytorch/issues/24089,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-02-15T03:29:12Z +issue,43579,"For the same complex dtype and same value, comparing a PyTorch tensor with a NumPy array results in False",https://github.com/pytorch/pytorch/issues/43579,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2024-02-14T20:12:09Z +issue,68961,Deterministic implementation for upsample_bilinear2d_backward_cuda,https://github.com/pytorch/pytorch/issues/68961,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism",scan when time allows,2024-02-14T19:03:59Z +issue,119843,"Docker image pytorch/pytorch:2.1.2-cuda12.1-cudnn8-runtime uses python 3.10, but default for Ubuntu 20.04 is python 3.8 => lots of issues",https://github.com/pytorch/pytorch/issues/119843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2024-02-14T14:24:48Z +issue,119281,Support for dynamic DataLoader (DataLoader.__getitem__),https://github.com/pytorch/pytorch/issues/119281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-02-14T11:35:56Z +issue,119711,torch.sparse.spdiags must support integer offsets,https://github.com/pytorch/pytorch/issues/119711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-02-13T17:16:14Z +issue,119624,Submatrix multiplication selected by rows or columns,https://github.com/pytorch/pytorch/issues/119624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2024-02-13T17:12:19Z +issue,119633,CUDA-GCC version mapping for CUDA 11.7+,https://github.com/pytorch/pytorch/issues/119633,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda",scan when time allows,2024-02-13T17:07:17Z +issue,50689,RandomSampler is very slow with huge dataset,https://github.com/pytorch/pytorch/issues/50689,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-02-13T15:23:58Z +issue,73665,Nice way to override string representation of tensor subclasses,https://github.com/pytorch/pytorch/issues/73665,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-13T13:55:37Z +issue,40972,len of dataloader when using iterable dataset does not reflect batch size,https://github.com/pytorch/pytorch/issues/40972,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: nn",scan when time allows,2024-02-12T23:07:59Z +issue,119431,FSDP convergence issue with CPU offload,https://github.com/pytorch/pytorch/issues/119431,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2024-02-12T19:37:51Z +issue,118891,Add an option for torchrun c10d store to be collocated with rank 0 training node,https://github.com/pytorch/pytorch/issues/118891,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-02-12T17:03:24Z +issue,119486,Conversion of 2d and 4d vectors in c++ to torch::tensor doesn't have the right data,https://github.com/pytorch/pytorch/issues/119486,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-02-09T18:42:36Z +issue,118918,Expectile Loss?,https://github.com/pytorch/pytorch/issues/118918,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2024-02-09T18:17:30Z +issue,119524,"Consistency in using (mean, std) vs. (loc, scale) terminology in Normal distributions",https://github.com/pytorch/pytorch/issues/119524,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2024-02-09T16:49:06Z +issue,116626,`test_modules.py` does not test for half precision or complex types,https://github.com/pytorch/pytorch/issues/116626,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: nn",scan when time allows,2024-02-08T20:51:06Z +issue,119473,Document how to override autocast rules properly,https://github.com/pytorch/pytorch/issues/119473,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2024-02-08T20:43:22Z +issue,119387,Lazy Tensor GetBackendDevice is not working at all,https://github.com/pytorch/pytorch/issues/119387,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-08T18:18:35Z +issue,119462,Python 3.12 CPU TestBinaryUfuncsCPU.test_cumulative_trapezoid_cpu is failing,https://github.com/pytorch/pytorch/issues/119462,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: scipy compatibility, module: tests",scan when time allows,2024-02-08T18:04:03Z +issue,119464,Python 3.12 CPU TestNamedTensor::test_using_unseen_interned/uninterned tests are failing,https://github.com/pytorch/pytorch/issues/119464,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: named tensor, module: tests",scan when time allows,2024-02-08T18:03:14Z +issue,119206,load_inline should always recompile a kernel if it failed,https://github.com/pytorch/pytorch/issues/119206,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2024-02-08T15:26:52Z +issue,99584,Training Faster R-CNN model with COCO dataset has been consistently unsuccessful.,https://github.com/pytorch/pytorch/issues/99584,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2024-02-08T13:34:58Z +issue,119418,"[c10d][NCCL] better error msg than ""Software caused connection abort""",https://github.com/pytorch/pytorch/issues/119418,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-07T22:14:38Z +issue,119417,[c10d] accept [] in torch.distributed.all_gather_object,https://github.com/pytorch/pytorch/issues/119417,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-07T22:09:25Z +issue,108861,RuntimeError: DataLoader worker (pid 11011) is killed by signal: Bus error. It is possible that dataloader's workers are out of shared memory. Please try to raise your shared memory limit,https://github.com/pytorch/pytorch/issues/108861,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2024-02-07T18:47:45Z +issue,119377,Asked to report bug by a `UserWarning`,https://github.com/pytorch/pytorch/issues/119377,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-07T17:15:04Z +issue,93197,Add Support for RockChip NPUs (RKNN(2)),https://github.com/pytorch/pytorch/issues/93197,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: arm,scan when time allows,2024-02-07T12:14:39Z +issue,12146,Misleading step method in lr_scheduler.ReduceLROnPlateau,https://github.com/pytorch/pytorch/issues/12146,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-02-07T09:56:15Z +issue,117245,```at::batch_norm```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117245,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cpp, module: nn",scan when time allows,2024-02-07T07:14:24Z +issue,118829,The inference result is wrong when I apply the torch.quantization.fuse_modules api to resnet50 model.,https://github.com/pytorch/pytorch/issues/118829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-02-07T05:21:05Z +issue,119279,MODULE.bazel support,https://github.com/pytorch/pytorch/issues/119279,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: build",scan when time allows,2024-02-06T22:40:50Z +issue,117748,[NCCL 2.19.3] Performance reduction when using NVLSTree algorithm when running nccl-tests all-reduce,https://github.com/pytorch/pytorch/issues/117748,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nccl,scan when time allows,2024-02-05T20:13:49Z +issue,118015,Disable tests UX suggestions,https://github.com/pytorch/pytorch/issues/118015,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-02-05T19:33:44Z +issue,117647,Easier way to add a lot of tests,https://github.com/pytorch/pytorch/issues/117647,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-02-05T19:23:02Z +issue,118970,"When concatenating empty variable with non-empty variable, I get CUDA error: illegal memory access encounter",https://github.com/pytorch/pytorch/issues/118970,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-02-05T18:26:07Z +issue,118634,symbolically traced variables cannot be used as inputs to control flow,https://github.com/pytorch/pytorch/issues/118634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-02-05T16:57:43Z +issue,27504,[JIT] List Comprehensions With Ifs not Supported,https://github.com/pytorch/pytorch/issues/27504,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-02-05T14:59:21Z +issue,67092,Sparse matrix multiplication (torch.sparse.mm) NotImplementedError,https://github.com/pytorch/pytorch/issues/67092,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-02-05T07:26:17Z +issue,118863,Pytorch with DSA enabled? Are there binaries? Perhaps the nightly?,https://github.com/pytorch/pytorch/issues/118863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-02-03T16:56:24Z +issue,50273,torch.cross is divergent from np.cross,https://github.com/pytorch/pytorch/issues/50273,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy",scan when time allows,2024-02-03T10:00:20Z +issue,55267,Improve CUDA extension building experience,https://github.com/pytorch/pytorch/issues/55267,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2024-02-02T21:57:39Z +issue,44343,Asynchronous Execution on CPU,https://github.com/pytorch/pytorch/issues/44343,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2024-02-02T21:43:37Z +issue,118723,"Libtorch torch::load(tensor, ""tensor.pt""); error",https://github.com/pytorch/pytorch/issues/118723,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: serialization, oncall: jit",scan when time allows,2024-02-02T18:41:38Z +issue,118922,torch.export(strict=False) fails with tensor constants,https://github.com/pytorch/pytorch/issues/118922,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2024-02-02T18:06:00Z +issue,118883,Torch Sparse Sampled AddMM Backwards Memory Usage Blowup,https://github.com/pytorch/pytorch/issues/118883,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2024-02-02T01:02:49Z +issue,118861,"Wrong typing for `argnums` in `torch.func.jacrev`, `torch.func.jacfwd` and possibly `torch.func.vmap`.",https://github.com/pytorch/pytorch/issues/118861,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: typing",scan when time allows,2024-02-01T22:49:01Z +issue,72117,Force PyTorch to clear CUDA cache,https://github.com/pytorch/pytorch/issues/72117,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-02-01T22:26:09Z +issue,59074,c++ use pybind11 to import torch free(): invalid pointer,https://github.com/pytorch/pytorch/issues/59074,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: pybind",scan when time allows,2024-02-01T19:20:13Z +issue,110250,Accessing Particular Nightly Builds Don't Work,https://github.com/pytorch/pytorch/issues/110250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-02-01T17:57:32Z +issue,118859,"[W CudaIPCTypes.cpp:95] Producer process tried to deallocate over 1000 memory blocks referred by consumer processes. Deallocation might be significantly slowed down. We assume it will never going to be the case, but if it is, please file but to https://github.com/pytorch/pytorch",https://github.com/pytorch/pytorch/issues/118859,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2024-02-01T13:01:45Z +issue,100419,Subgraph rewriter: Unable to match constant args,https://github.com/pytorch/pytorch/issues/100419,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-01-31T21:22:30Z +issue,96396,[dynamo] add hook to modify instructions before/after instructions be generated,https://github.com/pytorch/pytorch/issues/96396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dynamo,scan when time allows,2024-01-31T20:07:56Z +issue,96386,"[export] ""strict subset of traced input/output"" error when huggingface `ModelOutput` is returned",https://github.com/pytorch/pytorch/issues/96386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2024-01-31T20:06:42Z +issue,95021,Custom operations in inductor,https://github.com/pytorch/pytorch/issues/95021,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-01-31T19:58:00Z +issue,118736,[typing] `torch.Tensor` subclass types (eg `nn.Parameter`) broken by 2.2 release,https://github.com/pytorch/pytorch/issues/118736,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: typing",scan when time allows,2024-01-31T16:02:23Z +issue,112716,Quantized model's size gets doubled on optimizing and saving it for mobile,https://github.com/pytorch/pytorch/issues/112716,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: mobile, oncall: quantization",scan when time allows,2024-01-31T07:14:40Z +issue,116567,[MPS] `sort` incorrectly handles 'negative' NaNs,https://github.com/pytorch/pytorch/issues/116567,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2024-01-30T21:28:11Z +issue,74842,error in quantization by quantize_fx.prepare_fx,https://github.com/pytorch/pytorch/issues/74842,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2024-01-30T14:37:54Z +issue,46169,Squared 2-norm pdist (as available in SciPy / Faiss),https://github.com/pytorch/pytorch/issues/46169,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: norms and normalization, module: numpy",scan when time allows,2024-01-30T10:03:59Z +issue,74614,compilation error with PyTorch v1.11 for CPU on ppc64le,https://github.com/pytorch/pytorch/issues/74614,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: power,scan when time allows,2024-01-30T08:23:23Z +issue,118390,Audit all global mutable state in PT2 to see if it should be thread local instead,https://github.com/pytorch/pytorch/issues/118390,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: inductor, oncall: pt2",scan when time allows,2024-01-30T00:25:05Z +issue,43760,Error : lib/libstdc++.so.6: version `CXXABI_1.3.11 not found,https://github.com/pytorch/pytorch/issues/43760,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2024-01-30T00:06:31Z +issue,117976,Unify test_torchinductor.copy_tests and make_test_cls_with_patches,https://github.com/pytorch/pytorch/issues/117976,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2024-01-29T23:12:08Z +issue,118373,Torch + CUDA installation,https://github.com/pytorch/pytorch/issues/118373,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2024-01-29T20:42:58Z +issue,118517,ValueError when setting tensor type to match MPS tensor type,https://github.com/pytorch/pytorch/issues/118517,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2024-01-29T19:51:00Z +issue,118488,RFC: Enable disallow_untyped_defs by default,https://github.com/pytorch/pytorch/issues/118488,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: typing",scan when time allows,2024-01-29T19:14:51Z +issue,118508,[Bug] Bugs when installing `torch-sparse` and `torch-scatter` in docker.,https://github.com/pytorch/pytorch/issues/118508,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2024-01-29T19:12:50Z +issue,118461,[DTensor] DTensor fails to execute backward of index_put op.,https://github.com/pytorch/pytorch/issues/118461,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2024-01-29T19:06:48Z +issue,118095,NAN value for truthfulqa_mc2 on full finetuned model TinyLlama,https://github.com/pytorch/pytorch/issues/118095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2024-01-29T19:03:54Z +issue,115806,Merge all 3 mypy lint jobs,https://github.com/pytorch/pytorch/issues/115806,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2024-01-27T22:31:04Z +issue,44156,CUDA memory leak in multi-processing,https://github.com/pytorch/pytorch/issues/44156,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: performance",scan when time allows,2024-01-27T20:16:06Z +issue,114392,[BUG][pytree] equal `dict`s do not imply equal leaves and equal treespecs,https://github.com/pytorch/pytorch/issues/114392,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pytree,scan when time allows,2024-01-27T17:33:43Z +issue,118370,Tests aren't cleaning up state in test_optim.py,https://github.com/pytorch/pytorch/issues/118370,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: tests",scan when time allows,2024-01-26T16:56:07Z +issue,118352,Unify TEST_WITH_TORCHDYNAMO and PYTORCH_TEST_WITH_DYNAMO,https://github.com/pytorch/pytorch/issues/118352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2024-01-26T16:34:23Z +issue,118294,Failures in test_nn depending on test order,https://github.com/pytorch/pytorch/issues/118294,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: testing",scan when time allows,2024-01-26T16:23:43Z +issue,117826,[MPS] Probable incorrect gradient for einsum,https://github.com/pytorch/pytorch/issues/117826,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2024-01-25T06:30:47Z +issue,118245,[FSDP] NO_SHARD numerics with cpu offloading and orig params,https://github.com/pytorch/pytorch/issues/118245,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2024-01-24T23:26:44Z +issue,118058,torch.export doesn't handle functools.partial correctly.,https://github.com/pytorch/pytorch/issues/118058,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, oncall: export",scan when time allows,2024-01-24T20:27:56Z +issue,117437,_register_state_dict_hook docstring is not accurate?,https://github.com/pytorch/pytorch/issues/117437,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2024-01-24T19:38:40Z +issue,107534,NNPACK slow down M1/M2 Mac CPU,https://github.com/pytorch/pytorch/issues/107534,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nnpack,scan when time allows,2024-01-24T13:54:53Z +issue,117967,"Manually modified parameter during trainning, gradients are not being correctly updated. Help",https://github.com/pytorch/pytorch/issues/117967,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-01-24T10:57:51Z +issue,5580,"[feature request] F.interpolate to support integral data types: bool, int8, int32, int16, int64 ||| support uint8 on CUDA",https://github.com/pytorch/pytorch/issues/5580,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2024-01-24T10:22:49Z +issue,97772,consider bumping `DEFAULT_PROTOCOL`,https://github.com/pytorch/pytorch/issues/97772,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-01-24T02:03:42Z +issue,117576,Warning for unusual log sizes in CI,https://github.com/pytorch/pytorch/issues/117576,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-01-23T21:37:15Z +issue,116896,keep-going UX is not good,https://github.com/pytorch/pytorch/issues/116896,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-01-23T21:16:24Z +issue,115659,Clean up AWS credentials from PyTorch CI and replace them with OIDC,https://github.com/pytorch/pytorch/issues/115659,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-01-23T20:07:43Z +issue,118093,torch.where only accepts boolean tensor as condition deviating from np.where where you can also use binary input condition of dtype float,https://github.com/pytorch/pytorch/issues/118093,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2024-01-23T19:56:24Z +issue,64327,RFC: multi-part `torch.load`/`torch.save` to support huge models and/or low CPU memory,https://github.com/pytorch/pytorch/issues/64327,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization",scan when time allows,2024-01-23T19:22:27Z +issue,83540,I have the same issue as @samgelman on my MacOS.,https://github.com/pytorch/pytorch/issues/83540,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, module: openmp, module: third_party",scan when time allows,2024-01-23T12:24:12Z +issue,61457,using DistributedSampler occur RuntimeError: Expected a 'cuda:0' generator device but found 'cpu',https://github.com/pytorch/pytorch/issues/61457,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",scan when time allows,2024-01-23T12:09:22Z +issue,4392,"Cache CuDNN benchmark selection, turn it on by default, use it across PyTorch runs",https://github.com/pytorch/pytorch/issues/4392,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2024-01-23T07:40:06Z +issue,116052,Add a capturable impl to ASGD single tensor,https://github.com/pytorch/pytorch/issues/116052,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-01-22T22:38:00Z +issue,118018,Enable all optimizers (except SparseAdam + LBFGS) to be capturable,https://github.com/pytorch/pytorch/issues/118018,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs, module: optimizer",scan when time allows,2024-01-22T21:33:14Z +issue,117645,Segfault linear_search_fdes when calling c10::get_backtrace,https://github.com/pytorch/pytorch/issues/117645,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2024-01-22T20:54:53Z +issue,117981,[Feature] Foreach view,https://github.com/pytorch/pytorch/issues/117981,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: performance",scan when time allows,2024-01-22T17:44:40Z +issue,117791,[RFC][Export] Capture module attribute/annotation in `torch.export.unflatten`,https://github.com/pytorch/pytorch/issues/117791,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2024-01-22T17:26:29Z +issue,117938,Libtorch Don't Support Multi-thread Inference,https://github.com/pytorch/pytorch/issues/117938,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2024-01-22T15:08:02Z +issue,117963,Does Libtorch c++ convolution operation use optimization?,https://github.com/pytorch/pytorch/issues/117963,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cpp",scan when time allows,2024-01-22T14:09:50Z +issue,105068,[linalg] test_ops.py::test_python_ref_meta__refs_linalg_svd_cpu_complex failing,https://github.com/pytorch/pytorch/issues/105068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: meta tensors",scan when time allows,2024-01-22T13:55:07Z +issue,94966,CSR matrix add_ error with RuntimeError: CUDA error: kernel launch failure when calling cusparseXcsrgeam2Nnz,https://github.com/pytorch/pytorch/issues/94966,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2024-01-21T10:31:36Z +issue,117910,canonical way to normalize torch.device?,https://github.com/pytorch/pytorch/issues/117910,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: python frontend",scan when time allows,2024-01-20T04:29:17Z +issue,117832,amp need a whitelist,https://github.com/pytorch/pytorch/issues/117832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2024-01-20T04:23:12Z +issue,117728,[DTensor] addmm strategy cannot handle 0-element input tensors,https://github.com/pytorch/pytorch/issues/117728,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2024-01-19T19:33:43Z +issue,76108,RuntimeError: [enforce fail at C:\actions-runner\_work\pytorch\pytorch\builder\windows\pytorch\caffe2\serialize\inline_container.cc:300],https://github.com/pytorch/pytorch/issues/76108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2024-01-19T12:52:34Z +issue,117429,Memory leak while initializing tensors in child thread,https://github.com/pytorch/pytorch/issues/117429,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: memory usage",scan when time allows,2024-01-18T14:44:06Z +issue,111621,maximum Python version supported is not indicated,https://github.com/pytorch/pytorch/issues/111621,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-01-18T10:15:23Z +issue,35603,Performance bug with convolutions with weights and inputs of similar spatial size,https://github.com/pytorch/pytorch/issues/35603,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: performance",scan when time allows,2024-01-18T09:08:46Z +issue,114632,[DeviceMesh] back DeviceMesh initialization by custom_pg,https://github.com/pytorch/pytorch/issues/114632,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devicemesh, oncall: distributed",scan when time allows,2024-01-17T22:19:23Z +issue,116569,Python version inconsistency between Linux and Windows in the getting started,https://github.com/pytorch/pytorch/issues/116569,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-01-17T17:12:59Z +issue,65313,openSUSE Build: CMake Generate step failed. Build files cannot be regenerated correctly.,https://github.com/pytorch/pytorch/issues/65313,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2024-01-17T07:15:26Z +issue,117601,[FSDP] more actionable error logging in meta init,https://github.com/pytorch/pytorch/issues/117601,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2024-01-17T02:10:35Z +issue,115058,[DTensor] enforcing foreach for optimizer causes jittering,https://github.com/pytorch/pytorch/issues/115058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2024-01-16T23:37:39Z +issue,117467,Is this an expected memory profile?,https://github.com/pytorch/pytorch/issues/117467,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: multiprocessing",scan when time allows,2024-01-16T14:50:04Z +issue,117508,Consistency between torch.ao.quantization _LearnableFakeQuantize PTQ toggle and FakeQuantize PTQ toggle names.,https://github.com/pytorch/pytorch/issues/117508,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-01-16T14:48:19Z +issue,42594,support LSTM for quantization aware training,https://github.com/pytorch/pytorch/issues/42594,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-01-16T02:46:42Z +issue,117246,```at::smooth_l1_loss```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117246,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: loss",scan when time allows,2024-01-16T01:44:42Z +issue,108399,Generalize weight prepacking during quantized model deserialization,https://github.com/pytorch/pytorch/issues/108399,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2024-01-15T06:44:37Z +issue,117424,Various CI configurations are missing a slow equivalent,https://github.com/pytorch/pytorch/issues/117424,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2024-01-13T00:36:46Z +issue,113543,[PTD] Logging Improvements Main Task,https://github.com/pytorch/pytorch/issues/113543,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2024-01-12T21:12:52Z +issue,34306,[docs] Missing docs online for conv_tbc,https://github.com/pytorch/pytorch/issues/34306,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2024-01-12T20:29:45Z +issue,116691,torch.distributed doesn't have is_nccl_available() as the documentation suggests,https://github.com/pytorch/pytorch/issues/116691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-01-12T16:01:45Z +issue,109194,"`torch.mean` not supported for `torch.sparse_coo_tensor`, but `torch.sum` is supported (`scipy.sparse.coo_matrix` does support both `mean` and `sum`)",https://github.com/pytorch/pytorch/issues/109194,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-12T06:54:43Z +issue,40635,[JIT] Recursive compilation doesn't apply for types used in type expressions but not value expressions,https://github.com/pytorch/pytorch/issues/40635,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-01-12T03:05:48Z +issue,35108,"zip(list, tuple) throws an non-actionable error message",https://github.com/pytorch/pytorch/issues/35108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2024-01-12T03:05:35Z +issue,117259,```torch.Tensor.sum```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117259,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-11T19:57:55Z +issue,117256,```torch.Tensor.round_```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117256,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-11T19:57:22Z +issue,117251,"After model compilation and run on second gpu card (cuda:1) , 154 MB still goes to the first gpu card (cuda:0).Using pytorch ignite framework",https://github.com/pytorch/pytorch/issues/117251,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: inductor,scan when time allows,2024-01-11T19:56:26Z +issue,117250,```torch.nn.functional.normalize```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-11T19:55:13Z +issue,117249,```torch.Tensor.triangular_solve```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117249,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: sparse",scan when time allows,2024-01-11T19:55:03Z +issue,117248,```torch.Tensor.matmul```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-11T19:54:49Z +issue,117244,```torch.Tensor.index_copy_```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-11T19:54:03Z +issue,117243,```torch.Tensor.cov```: Internal assertion failed,https://github.com/pytorch/pytorch/issues/117243,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-11T19:53:25Z +issue,117232,```torch.Tensor.renorm```: Internal assert failed,https://github.com/pytorch/pytorch/issues/117232,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2024-01-11T19:43:56Z +issue,117247,```torch.nn.Linear``` class: Internal assertion failed,https://github.com/pytorch/pytorch/issues/117247,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-11T19:11:04Z +issue,116861,test_serialization_offset may fail with small $TMP,https://github.com/pytorch/pytorch/issues/116861,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization, module: tests",scan when time allows,2024-01-10T12:26:28Z +issue,92398,[Inductor] support complex dtypes,https://github.com/pytorch/pytorch/issues/92398,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: inductor, module: random, oncall: cpu inductor",scan when time allows,2024-01-10T12:10:17Z +issue,116633,Sparse tensor element-wise multiplication bug when broadcast needed,https://github.com/pytorch/pytorch/issues/116633,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-10T01:17:14Z +issue,116775,CI on the main branch should run with keep-going,https://github.com/pytorch/pytorch/issues/116775,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2024-01-09T21:55:14Z +issue,116870,cuda FP16 default accumulation type differs between TensorIterator vs. torch.mm,https://github.com/pytorch/pytorch/issues/116870,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability, module: performance",scan when time allows,2024-01-09T20:58:37Z +issue,114388,Respect user-specified USE_ROCM/USE_CUDA,https://github.com/pytorch/pytorch/issues/114388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: rocm",scan when time allows,2024-01-09T19:14:06Z +issue,96643,PyTorch SGEMV is using 1 single core on AMD CPUs (very slow),https://github.com/pytorch/pytorch/issues/96643,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2024-01-09T13:46:47Z +issue,94909,[numpy] mean & nanmean should support int dtypes,https://github.com/pytorch/pytorch/issues/94909,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: reductions",scan when time allows,2024-01-09T13:01:40Z +issue,116980,Sharded shape attributes for DTensor,https://github.com/pytorch/pytorch/issues/116980,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-01-09T05:15:50Z +issue,110387,Change unsqueeze(0) to preserve memory layout contiguity of original tensor,https://github.com/pytorch/pytorch/issues/110387,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: numerical-reproducibility",scan when time allows,2024-01-09T02:33:15Z +issue,111509,Sparse Tensor Sum Still Does Not Work for PyTorch Geometric,https://github.com/pytorch/pytorch/issues/111509,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-08T21:23:50Z +issue,116987,[pytree] class based registration api,https://github.com/pytorch/pytorch/issues/116987,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pytree,scan when time allows,2024-01-08T20:05:12Z +issue,116320,RuntimeError: CUDA error: an illegal memory access was encountered using vmap and model ensembling call for cuda system,https://github.com/pytorch/pytorch/issues/116320,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda, module: functorch, module: vmap",scan when time allows,2024-01-08T18:59:23Z +issue,116018,Inconsistent behavior of torch.einsum on float32 and complex128,https://github.com/pytorch/pytorch/issues/116018,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: linear algebra",scan when time allows,2024-01-08T18:07:52Z +issue,116403,pin_memory() for a custom Python object,https://github.com/pytorch/pytorch/issues/116403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2024-01-06T09:42:14Z +issue,116543,Multiprocessing using Gunicorn causing CUDA Error,https://github.com/pytorch/pytorch/issues/116543,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2024-01-05T12:49:08Z +issue,112509,Add a function to torch.nested to create nested tensors from a buffer and sizes,https://github.com/pytorch/pytorch/issues/112509,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: serialization",scan when time allows,2024-01-05T11:57:35Z +issue,113024,Upsample bilinear 2d decomposition does not match native implementation for uint8,https://github.com/pytorch/pytorch/issues/113024,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: decompositions,scan when time allows,2024-01-05T09:37:02Z +issue,110819,Segmentation fault on aarch64 (Rpi4) using Pytorch 2.1.0 & torchaudio,https://github.com/pytorch/pytorch/issues/110819,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2024-01-05T02:58:49Z +issue,96595,Why doesn't PyTorch install the REAL nvidia cuDNN pip package?,https://github.com/pytorch/pytorch/issues/96595,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2024-01-04T17:27:58Z +issue,114090,Nested tensors fail on Conv2D,https://github.com/pytorch/pytorch/issues/114090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2024-01-04T17:09:13Z +issue,116762,parametrized should fail to generate tests with weird names,https://github.com/pytorch/pytorch/issues/116762,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2024-01-04T10:37:26Z +issue,116533,`backward` implicit conversion from tuple to `torch.Tensor` results in an indexing error message,https://github.com/pytorch/pytorch/issues/116533,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: molly-guard, module: python frontend",scan when time allows,2024-01-03T09:54:47Z +issue,116576,Build with MKL and OpenMP produces libtorch_global_deps.so with undefined symbol omp_get_max_active_levels,https://github.com/pytorch/pytorch/issues/116576,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: dependency bug",scan when time allows,2024-01-03T08:30:15Z +issue,116565,`test_autograd_dense_output_addmm` fails if matrix sizes are square,https://github.com/pytorch/pytorch/issues/116565,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2024-01-03T08:25:22Z +issue,78067,Installation on Jetson target board,https://github.com/pytorch/pytorch/issues/78067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: jetson,scan when time allows,2024-01-03T08:05:20Z +issue,116610,Align `torch.sign` with Array API,https://github.com/pytorch/pytorch/issues/116610,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2024-01-02T15:57:41Z +issue,116400,Pytorch Transformer Distributed Training,https://github.com/pytorch/pytorch/issues/116400,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2024-01-02T14:31:25Z +issue,116592,Support float64 step in fused optimizers,https://github.com/pytorch/pytorch/issues/116592,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2024-01-01T20:03:15Z +issue,116558,verbose stack trace for default collate function,https://github.com/pytorch/pytorch/issues/116558,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: error checking",scan when time allows,2024-01-01T00:44:13Z +issue,115257,Compiling source code reported an error,https://github.com/pytorch/pytorch/issues/115257,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: nccl",scan when time allows,2023-12-31T03:05:39Z +issue,116461,Numpy like Masked/where operations,https://github.com/pytorch/pytorch/issues/116461,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2023-12-31T03:02:05Z +issue,111368,Implement device parameter in Dropout2d,https://github.com/pytorch/pytorch/issues/111368,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-12-30T23:48:33Z +issue,115821,[BUG] missing automatic dtype promotion for out-of-place `torch.scatter_*()` functions,https://github.com/pytorch/pytorch/issues/115821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2023-12-30T03:21:59Z +issue,105319,Implementation of torch.sparse.sampled_baddmm,https://github.com/pytorch/pytorch/issues/105319,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-12-28T16:38:21Z +issue,116479,Address huge memory usage on large dataset,https://github.com/pytorch/pytorch/issues/116479,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-12-28T16:28:32Z +issue,116502,Add BC test for load_state_dict on OptimizerInfos and ModuleInfos,https://github.com/pytorch/pytorch/issues/116502,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer, module: tests",scan when time allows,2023-12-28T11:06:17Z +issue,114175,Unable to build on CUDA 11.8 due to cutlass incompatibility,https://github.com/pytorch/pytorch/issues/114175,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2023-12-27T12:38:10Z +issue,1642,feature request: On-the-fly Operation Batching in Dynamic Computation Graphs,https://github.com/pytorch/pytorch/issues/1642,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-12-26T05:49:11Z +issue,115610,[Bug] Big difference between the output of Conv float precision and double precision,https://github.com/pytorch/pytorch/issues/115610,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn",scan when time allows,2023-12-23T23:58:18Z +issue,116184,Potential cublas handle leaking,https://github.com/pytorch/pytorch/issues/116184,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2023-12-23T23:48:18Z +issue,100411,Sparse Matrix nnz Overflow when casting from COO to CSR,https://github.com/pytorch/pytorch/issues/100411,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-12-23T14:28:41Z +issue,75500,RuntimeError: cuDNN error: CUDNN_STATUS_EXECUTION_FAILED,https://github.com/pytorch/pytorch/issues/75500,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn, module: dependency bug",scan when time allows,2023-12-23T12:06:41Z +issue,113378,[pytree] implement key path API,https://github.com/pytorch/pytorch/issues/113378,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pytree,scan when time allows,2023-12-21T20:00:35Z +issue,116255,PrivateUseOne backend does't support tensor shollow copy,https://github.com/pytorch/pytorch/issues/116255,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2023-12-21T16:10:29Z +issue,74420,`torch.histogram` has wrong output dtype and doesn't support integer inputs,https://github.com/pytorch/pytorch/issues/74420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: sorting and selection",scan when time allows,2023-12-21T09:34:29Z +issue,115309,"[distributed] add a ""small"" transformer model to the test suite for testing usage",https://github.com/pytorch/pytorch/issues/115309,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-12-21T02:12:28Z +issue,98600,Wrong illustration in README.md,https://github.com/pytorch/pytorch/issues/98600,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-12-21T00:50:31Z +issue,114188,Unsupported operator error: `aten::to_mkldnn` export to ONNX not supported,https://github.com/pytorch/pytorch/issues/114188,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, module: onnx",scan when time allows,2023-12-20T21:54:40Z +issue,116012,Issue with Protoc while building PyTorch for RISC-V,https://github.com/pytorch/pytorch/issues/116012,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-12-20T18:58:47Z +issue,116117,WeightedRandomSampler doesnt keep expected ratio betwen samples,https://github.com/pytorch/pytorch/issues/116117,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-12-20T18:54:49Z +issue,116101,[DTensor] Support API to shard to parent mesh,https://github.com/pytorch/pytorch/issues/116101,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2023-12-19T23:31:33Z +issue,115820,Question on XNNPACK support for PowerPC,https://github.com/pytorch/pytorch/issues/115820,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: power, module: third_party, module: xnnpack",scan when time allows,2023-12-19T21:28:19Z +issue,115785,arctan2 fp16 error when optimising,https://github.com/pytorch/pytorch/issues/115785,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: half, module: optimizer",scan when time allows,2023-12-19T18:04:57Z +issue,62466,Support gathering string in DataParallel,https://github.com/pytorch/pytorch/issues/62466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2023-12-19T11:50:46Z +issue,22169,[RFC] NestedTensor - 0.0.1,https://github.com/pytorch/pytorch/issues/22169,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: batching,scan when time allows,2023-12-19T02:41:42Z +issue,91263,Make quant_min/quant_max required for observer/fake_quant,https://github.com/pytorch/pytorch/issues/91263,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-12-19T01:44:53Z +issue,91604,wrong assert message,https://github.com/pytorch/pytorch/issues/91604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-12-19T01:43:43Z +issue,110515,DynamicQuantizedLinear shows incorrect qscheme after applying eager mode dynamic quantization,https://github.com/pytorch/pytorch/issues/110515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-12-19T01:38:20Z +issue,111902,torchscript file can not be loaded if its saved form the export model produced by torch.export.export,https://github.com/pytorch/pytorch/issues/111902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2023-12-19T01:32:12Z +issue,115978,Implement KV cached attention with NestedTensor,https://github.com/pytorch/pytorch/issues/115978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-12-18T16:21:42Z +issue,78722,AlBert quantization,https://github.com/pytorch/pytorch/issues/78722,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-12-18T09:41:25Z +issue,115966,Adam Eager Benchmark Failures,https://github.com/pytorch/pytorch/issues/115966,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: inductor,scan when time allows,2023-12-17T05:42:02Z +issue,23048,Add automatic tuning flags to utils.data.dataloader,https://github.com/pytorch/pytorch/issues/23048,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-12-16T01:12:29Z +issue,16266,support for multiple torch.cuda.max_memory_allocated() counters,https://github.com/pytorch/pytorch/issues/16266,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-12-15T23:18:45Z +issue,115614,Make test_public_bindings.py a lint,https://github.com/pytorch/pytorch/issues/115614,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-12-14T17:27:41Z +issue,112395,"KeyError during model export while using ""newer"" data types",https://github.com/pytorch/pytorch/issues/112395,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-12-14T16:36:18Z +issue,115672,Latest PyTorch is not buildable against CUDA-11.2,https://github.com/pytorch/pytorch/issues/115672,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: sparse",scan when time allows,2023-12-14T15:42:50Z +issue,90695,`torch.empty` produces incorrect tensors with `layout=sparse_csr|sparse_csc` on the CPU,https://github.com/pytorch/pytorch/issues/90695,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: docs, module: sparse",scan when time allows,2023-12-14T15:05:24Z +issue,43623,[FR] Raise an exception when constructing non-empty index and empty values sparse tensor,https://github.com/pytorch/pytorch/issues/43623,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2023-12-14T14:45:40Z +issue,115735,Document required semantics of allocation functions in `CUDAPluggableAllocator`,https://github.com/pytorch/pytorch/issues/115735,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: docs",scan when time allows,2023-12-14T11:08:24Z +issue,77955,NaN tensor values problem for GTX16xx users (no problem on other devices),https://github.com/pytorch/pytorch/issues/77955,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2023-12-14T10:59:39Z +issue,115624,`amax` fails on `masked_tensor` with multiple negative dims,https://github.com/pytorch/pytorch/issues/115624,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2023-12-13T16:58:41Z +issue,114943,torch.einsum may choose a strategy for which there is not enough memory,https://github.com/pytorch/pytorch/issues/114943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2023-12-13T11:15:54Z +issue,70173,Dynamic Tensor Rematerialization (DTR),https://github.com/pytorch/pytorch/issues/70173,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2023-12-12T22:31:15Z +issue,87085,gradcheck failure with sparse matrix multiplication,https://github.com/pytorch/pytorch/issues/87085,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-12-12T20:19:41Z +issue,115418,max_pool2d_with_indices_backward_out_cuda: remove useless code gradInput.zero_();,https://github.com/pytorch/pytorch/issues/115418,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance, module: pooling",scan when time allows,2023-12-12T20:17:26Z +issue,93477,[Inductor] [CPU] optimize thread parallel and loop collapse,https://github.com/pytorch/pytorch/issues/93477,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: cpu inductor, oncall: pt2",scan when time allows,2023-12-12T04:32:55Z +issue,115077,RuntimeError: cuSPARSELT not supported on your machine. When I was calling: torch._cslt_compress(),https://github.com/pytorch/pytorch/issues/115077,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: sparse",scan when time allows,2023-12-11T18:20:16Z +issue,97156,Implement `torch.distributions.Poisson.cdf()`,https://github.com/pytorch/pytorch/issues/97156,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-12-11T08:43:30Z +issue,115340,RuntimeError: CUDA unknown error even after sudo modprobe -r nvidia_uvm && sudo modprobe nvidia_uvm,https://github.com/pytorch/pytorch/issues/115340,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-12-08T21:30:58Z +issue,115434,torch.kthvalue should take a tensor of `k` values.,https://github.com/pytorch/pytorch/issues/115434,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2023-12-08T20:30:27Z +issue,115342,torch.func.grad not working with list comprehension over dictionary,https://github.com/pytorch/pytorch/issues/115342,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-12-08T20:24:06Z +issue,41119,TORCH_CUDA_ARCH_LIST deprecation warning,https://github.com/pytorch/pytorch/issues/41119,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-12-08T20:09:52Z +issue,115422,string representation method for empty masked tensors fails,https://github.com/pytorch/pytorch/issues/115422,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2023-12-08T20:08:45Z +issue,106780,inf and nan are mapped to quant_min in torch.fake_quantize_per_tensor_affine,https://github.com/pytorch/pytorch/issues/106780,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-12-08T19:00:03Z +issue,67597,torch.kthvalue (and maybe torch.quantile) should support argument descending/largest (=True) like topk/sort,https://github.com/pytorch/pytorch/issues/67597,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2023-12-08T18:10:53Z +issue,115421,Document PyTorch docker release policies in RELEASE.MD,https://github.com/pytorch/pytorch/issues/115421,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, oncall: releng",scan when time allows,2023-12-08T14:59:27Z +issue,8126,PyTorch multiprocessing using single CPU core,https://github.com/pytorch/pytorch/issues/8126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2023-12-08T13:14:35Z +issue,115164,nansum with complex support,https://github.com/pytorch/pytorch/issues/115164,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: masked operators, module: nans and infs",scan when time allows,2023-12-08T09:08:44Z +issue,114871,Sparse block tensors (`torch.sparse`),https://github.com/pytorch/pytorch/issues/114871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-12-08T08:44:16Z +issue,98872,Expand component configurable logging system to C++,https://github.com/pytorch/pytorch/issues/98872,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2023-12-07T20:39:25Z +issue,115350,Enable skipped MPS OptimizerInfo tests,https://github.com/pytorch/pytorch/issues/115350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: optimizer",scan when time allows,2023-12-07T18:45:46Z +issue,115301,inconsistency for enableTiming / FlightRecorder / duration_ms,https://github.com/pytorch/pytorch/issues/115301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nccl,scan when time allows,2023-12-07T15:33:50Z +issue,65663,Model Evaluation Returns Nan Values Sometimes for the Same Input,https://github.com/pytorch/pytorch/issues/65663,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,scan when time allows,2023-12-07T09:39:53Z +issue,115202,Auto skip the entire test when @parametrize takes an empty list,https://github.com/pytorch/pytorch/issues/115202,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2023-12-06T15:13:39Z +issue,47513,ReduceLROnPlateau fails for negative input,https://github.com/pytorch/pytorch/issues/47513,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2023-12-06T11:23:09Z +issue,112608,`grad_fn` is not defined in `__torch_dispatch__` arguments when lazy device is used,https://github.com/pytorch/pytorch/issues/112608,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-12-05T23:08:24Z +issue,114928,`torch.nn.utils.rnn.pack_padded_sequence` heap-buffer-overflow,https://github.com/pytorch/pytorch/issues/114928,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-12-05T21:36:25Z +issue,95423,PT2 Computes Multi Device Backward in a Single Thread,https://github.com/pytorch/pytorch/issues/95423,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2023-12-05T20:49:54Z +issue,80851,[bug][nvfuser] Applying nvfuser to the model leads to runtime error,https://github.com/pytorch/pytorch/issues/80851,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2023-12-05T17:40:03Z +issue,115064,torch.nn.PairwiseDistance，The results vary widely from version to version,https://github.com/pytorch/pytorch/issues/115064,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-12-05T08:37:07Z +issue,113985,[export] Helper function for specifying dynamic batch size,https://github.com/pytorch/pytorch/issues/113985,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2023-12-05T00:53:38Z +issue,115034,libtorch can't switch between with and without layers by is_training(),https://github.com/pytorch/pytorch/issues/115034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-12-04T22:42:44Z +issue,115020,[Libtorch/iOS] Build for iOS as dynamic library fails in linker phase.,https://github.com/pytorch/pytorch/issues/115020,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ios",scan when time allows,2023-12-04T22:27:08Z +issue,43536,"How do I debug ""RuntimeError: trying to initialize the default process group twice!""",https://github.com/pytorch/pytorch/issues/43536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-12-04T14:46:49Z +issue,108445,_foreach_copy_ with scalar second arg,https://github.com/pytorch/pytorch/issues/108445,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer",scan when time allows,2023-12-04T03:18:38Z +issue,114948,conda package _openmp_mutex makes pytorch dataloader slower when set num_workers > 0,https://github.com/pytorch/pytorch/issues/114948,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-12-03T14:34:00Z +issue,114951,Unexpected poor performance of C++ extension / wish for a fast `operator[]`,https://github.com/pytorch/pytorch/issues/114951,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: performance",scan when time allows,2023-12-02T21:12:32Z +issue,47918,Function request: Sparse matrix inverse,https://github.com/pytorch/pytorch/issues/47918,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2023-12-01T19:27:47Z +issue,111676,[export] self.buffer += 1 raises error,https://github.com/pytorch/pytorch/issues/111676,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: graph breaks, oncall: export",scan when time allows,2023-12-01T19:27:15Z +issue,114935,print statements used in torch/utils/cpp_extension.py,https://github.com/pytorch/pytorch/issues/114935,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2023-12-01T16:59:26Z +issue,114949,"Nvidia P100, where to disable upcasting? Plus kernel image missing.",https://github.com/pytorch/pytorch/issues/114949,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2023-12-01T16:50:59Z +issue,94164,Pytorch 2.0: Detection models from torchvision don't work with onnx and tensorrt backends,https://github.com/pytorch/pytorch/issues/94164,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: onnx, oncall: pt2",scan when time allows,2023-12-01T15:30:44Z +issue,113408,Early testing stop logic for CUDA error looks wrong for instantiated_test with pytest,https://github.com/pytorch/pytorch/issues/113408,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: testing",scan when time allows,2023-12-01T07:22:07Z +issue,103475,[Inductor] Optimize More Cases of Int32 -> Int64,https://github.com/pytorch/pytorch/issues/103475,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: inductor,scan when time allows,2023-11-30T19:58:59Z +issue,101117,Barriers to using torch.compile directly in PyTorch library code,https://github.com/pytorch/pytorch/issues/101117,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dynamo, module: inductor, oncall: pt2",scan when time allows,2023-11-30T19:52:59Z +issue,93639,Explore TorchInductor optimization pass to reorder kernel bodies,https://github.com/pytorch/pytorch/issues/93639,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2023-11-30T03:39:47Z +issue,114808,Efficient Cholesky and QR updates,https://github.com/pytorch/pytorch/issues/114808,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-11-30T02:51:42Z +issue,114805,Unsupported: call_method UserDefinedObjectVariable(_profiler_enabled) __call__ [] {},https://github.com/pytorch/pytorch/issues/114805,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: graph breaks,scan when time allows,2023-11-30T02:51:10Z +issue,95604,"Performance bugs exists in multiple convolution operations(e.g., `Convtranspose2d`) when useing the `groups` argument",https://github.com/pytorch/pytorch/issues/95604,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: docs",scan when time allows,2023-11-29T22:44:44Z +issue,31316,quantization - Missing operations needed for object detection,https://github.com/pytorch/pytorch/issues/31316,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-29T17:49:21Z +issue,114093,Found nn.LazyBatchNorm1d(0) has inconsistency bug between GPU and CPU testing,https://github.com/pytorch/pytorch/issues/114093,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-11-29T08:48:09Z +issue,113956,iSTFT gives wrong results for some batched input,https://github.com/pytorch/pytorch/issues/113956,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda, module: fft",scan when time allows,2023-11-29T01:17:11Z +issue,114679,Unexpected Results in PyTorch Tensor Operations with Python Scalars,https://github.com/pytorch/pytorch/issues/114679,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-11-28T23:23:20Z +issue,114672,torch.fft.ifft crashes for empty input,https://github.com/pytorch/pytorch/issues/114672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2023-11-28T23:13:29Z +issue,114581,Flawed testing of onesidedness in istft,https://github.com/pytorch/pytorch/issues/114581,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2023-11-28T23:10:46Z +issue,114440,There seems to be redundant calling of findOp in findSchemaOrThrow method of Dispatcher.cpp,https://github.com/pytorch/pytorch/issues/114440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2023-11-28T22:19:12Z +issue,114652,__torch_dispatch__ fails on functionalization when lazy device and `out=` arg is used,https://github.com/pytorch/pytorch/issues/114652,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: functionalization, module: lazy",scan when time allows,2023-11-28T21:12:36Z +issue,106137,"`export(..., pre_dispatch=True)` for model in eval mode still inserts autograd ops",https://github.com/pytorch/pytorch/issues/106137,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2023-11-28T17:12:28Z +issue,106302,`torch.nn.modules.MultiheadAttention` yields different graph under pre_dispatch tracing,https://github.com/pytorch/pytorch/issues/106302,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, oncall: export",scan when time allows,2023-11-28T17:12:15Z +issue,109753,fp16 parity issue with traced code on GPU,https://github.com/pytorch/pytorch/issues/109753,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, oncall: export, oncall: jit",scan when time allows,2023-11-28T17:11:07Z +issue,112079,The error type `UserError` in dynamo is confusing.,https://github.com/pytorch/pytorch/issues/112079,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2023-11-28T17:08:37Z +issue,112095,torch.export fails on a model with optional parameter,https://github.com/pytorch/pytorch/issues/112095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2023-11-28T17:08:29Z +issue,113507,Revisit test_edge_op_registration on PyTorch CI,https://github.com/pytorch/pytorch/issues/113507,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2023-11-28T17:07:41Z +issue,113552,2.1.0 export module：Why can't export() capture reverse graphs?,https://github.com/pytorch/pytorch/issues/113552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: export,scan when time allows,2023-11-28T17:07:34Z +issue,40134,About torch.backends.cudnn.deterministic issue,https://github.com/pytorch/pytorch/issues/40134,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2023-11-28T10:02:42Z +issue,114604,"lax.cond, lax.switch alternatives",https://github.com/pytorch/pytorch/issues/114604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-11-27T19:59:40Z +issue,114350,The 'out'-parameter in torch.matmul() works for 'cuda' device but not for 'cpu',https://github.com/pytorch/pytorch/issues/114350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cpu, module: error checking, module: intel",scan when time allows,2023-11-27T18:15:34Z +issue,114216,torch.nn.functional.max_pool2d outputs inf,https://github.com/pytorch/pytorch/issues/114216,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: nn",scan when time allows,2023-11-27T15:37:39Z +issue,107854,"""file_descriptor"" multiprocessing sharing strategy works incorrectly in dataloading",https://github.com/pytorch/pytorch/issues/107854,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2023-11-25T15:35:20Z +issue,98187,how can i load seperate pytorch_model.bin?,https://github.com/pytorch/pytorch/issues/98187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-11-24T15:32:58Z +issue,114080,torch.matrix_exp(x) get inf and nan,https://github.com/pytorch/pytorch/issues/114080,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda",scan when time allows,2023-11-23T00:09:42Z +issue,114044,Nan on torch.corrcoef(x.t()),https://github.com/pytorch/pytorch/issues/114044,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-11-22T23:46:50Z +issue,112857,Vendored FindCUDAToolkit.cmake deviates from upstream in splayed installation support,https://github.com/pytorch/pytorch/issues/112857,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-11-22T18:16:52Z +issue,114364,Improve usability of CUDA package by adding description of CUDA,https://github.com/pytorch/pytorch/issues/114364,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-11-22T16:54:25Z +issue,114356,"[WARNING] could not load None, generating random data instead is too spammy",https://github.com/pytorch/pytorch/issues/114356,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: minifier,scan when time allows,2023-11-22T14:50:54Z +issue,99025,Is there a way to get the full call stack of pytorch from python to C/C++?,https://github.com/pytorch/pytorch/issues/99025,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-11-22T08:56:55Z +issue,62616,LazyTensor cuda fallback,https://github.com/pytorch/pytorch/issues/62616,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2023-11-21T20:49:12Z +issue,62619,LazyTensor support for fuser integration investigation,https://github.com/pytorch/pytorch/issues/62619,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2023-11-21T20:48:59Z +issue,20433,Dataloader's memory usage keeps increasing during one single epoch.,https://github.com/pytorch/pytorch/issues/20433,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-11-21T18:03:15Z +issue,98459,Fail to pass test HAVE_XXX_REGEX while building pytorch,https://github.com/pytorch/pytorch/issues/98459,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-11-21T06:13:53Z +issue,102833,LibTorch-Lite 1.13.0.1 Crash on iOS 12 on app startup,https://github.com/pytorch/pytorch/issues/102833,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ios,scan when time allows,2023-11-21T02:43:27Z +issue,2478,ReduceLROnPlateau with a naive Backtracking,https://github.com/pytorch/pytorch/issues/2478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-11-20T21:45:01Z +issue,114155,Lintrunner on all files fails locally even though it passes in CI,https://github.com/pytorch/pytorch/issues/114155,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2023-11-20T20:24:05Z +issue,114131,Inconsistency of state_dict loading across devices,https://github.com/pytorch/pytorch/issues/114131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-11-20T19:15:59Z +issue,113905,Sympy reasoning with true division is broken,https://github.com/pytorch/pytorch/issues/113905,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dependency bug,scan when time allows,2023-11-20T18:48:13Z +issue,58969,Improper use of quantization API for MHA should fail fast,https://github.com/pytorch/pytorch/issues/58969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-20T18:21:00Z +issue,114107,Add the possibility to pass a `Generator` to `gumbel_softmax`,https://github.com/pytorch/pytorch/issues/114107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: random",scan when time allows,2023-11-20T17:41:51Z +issue,84290,Fix convert path for fixed qparam ops (sigmoid and softmax),https://github.com/pytorch/pytorch/issues/84290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-20T16:34:59Z +issue,37343,qnnpack's quantized-add gives wrong result,https://github.com/pytorch/pytorch/issues/37343,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: mobile, oncall: quantization",scan when time allows,2023-11-20T15:26:16Z +issue,28906,Font used in documentation is not always sharp,https://github.com/pytorch/pytorch/issues/28906,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-11-20T15:22:02Z +issue,98929,torch.sparse_csr_tensor() stops gradients,https://github.com/pytorch/pytorch/issues/98929,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-11-20T13:24:06Z +issue,58155,"""uncorrectable NVLink error"" making the tests fail",https://github.com/pytorch/pytorch/issues/58155,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nccl, oncall: distributed",scan when time allows,2023-11-20T08:11:52Z +issue,47885,NCCL 2.7.8 errors on PyTorch distributed process group creation,https://github.com/pytorch/pytorch/issues/47885,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-11-20T06:57:32Z +issue,45352,torch.nn.functional.one_hot should gracefully skip negative and out-of-range indices,https://github.com/pytorch/pytorch/issues/45352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-11-19T12:20:47Z +issue,101972,No pytorch_android 2.0.x builds,https://github.com/pytorch/pytorch/issues/101972,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2023-11-18T14:32:06Z +issue,90613,`torch.inverse` multi-threading RuntimeError: lazy wrapper should be called at most once,https://github.com/pytorch/pytorch/issues/90613,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: multithreading",scan when time allows,2023-11-18T01:18:33Z +issue,90536,Illegal hardware instruction following Real Time Inference on Raspberry Pi 4 tutorial,https://github.com/pytorch/pytorch/issues/90536,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, oncall: quantization",scan when time allows,2023-11-17T23:49:49Z +issue,46749,Quantization - we need a better solution for tracking quantization backend settings in a model,https://github.com/pytorch/pytorch/issues/46749,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T23:02:03Z +issue,32262,models after prepare/prepare_qat contains `qconfig` which prevents pickling,https://github.com/pytorch/pytorch/issues/32262,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:48:14Z +issue,36508,Quantized _out functions don't follow same conventions as other out functions in the codebase,https://github.com/pytorch/pytorch/issues/36508,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:41:21Z +issue,36091,[quantization] torch.quantized_lstm and torch.quantized_gru not documented,https://github.com/pytorch/pytorch/issues/36091,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: quantization",scan when time allows,2023-11-17T22:40:23Z +issue,38569,[quantization] Version support for quantization BC tests,https://github.com/pytorch/pytorch/issues/38569,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:32:26Z +issue,39058,Failure when loading quantized pre-trained weights partially,https://github.com/pytorch/pytorch/issues/39058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:30:55Z +issue,41149,eager mode quantization should remove qconfig in the (non-leaf module of) quantized model,https://github.com/pytorch/pytorch/issues/41149,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:19:27Z +issue,42779,[quant] Quantized AdaptivePool3d is much slower for ChannelsLast3d.,https://github.com/pytorch/pytorch/issues/42779,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: performance, oncall: quantization",scan when time allows,2023-11-17T22:17:29Z +issue,44676,Make quantized::prepack_fp16 op just do prepacking,https://github.com/pytorch/pytorch/issues/44676,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:16:52Z +issue,45635,fx graph mode quantization tutorials,https://github.com/pytorch/pytorch/issues/45635,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:15:32Z +issue,46604,torch.concat doesn't raise an error in a quantized model,https://github.com/pytorch/pytorch/issues/46604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:14:28Z +issue,47857,QAT with DDP should have documentation,https://github.com/pytorch/pytorch/issues/47857,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:10:57Z +issue,48106,quantization - document get_default_qconfig,https://github.com/pytorch/pytorch/issues/48106,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:09:21Z +issue,51152,Make it clearer when a tensor was quantized symmetrically,https://github.com/pytorch/pytorch/issues/51152,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:07:20Z +issue,53026,Ensure that nn.MHA and functional multi-head attention is supported by FX graph mode quantization,https://github.com/pytorch/pytorch/issues/53026,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: quantization",scan when time allows,2023-11-17T22:06:09Z +issue,47535,per channel observer to work for weights of groupwise conv transpose,https://github.com/pytorch/pytorch/issues/47535,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T22:03:08Z +issue,54979,test_quantize_fx.py test_resnet_18_dpp test failure,https://github.com/pytorch/pytorch/issues/54979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: quantization",scan when time allows,2023-11-17T22:01:38Z +issue,56262,torch.nn.quantized.functional.conv_transpose1d/2d/3d support,https://github.com/pytorch/pytorch/issues/56262,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: quantization",scan when time allows,2023-11-17T21:59:18Z +issue,61246,"torch.quantization.fx.fuser forget to pass ""additional_fuser_method_mapping"" to fuse method",https://github.com/pytorch/pytorch/issues/61246,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: quantization",scan when time allows,2023-11-17T21:53:08Z +issue,60210,Quantized model using `boolean_dispatch` not picklable,https://github.com/pytorch/pytorch/issues/60210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:50:53Z +issue,61046,Audit use of __ARM_NEON__ define,https://github.com/pytorch/pytorch/issues/61046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:50:08Z +issue,69158,Fusing module with multiple hooks causes `RuntimeError: OrderedDict mutated during iteration` in fuse_known_modules,https://github.com/pytorch/pytorch/issues/69158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:48:13Z +issue,70480,Dynamic quantified Conv2d have accuracy issue,https://github.com/pytorch/pytorch/issues/70480,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:44:23Z +issue,71889,Make scale/zero_point accessible from TorchScript traced module for QFlaotFunctional modules,https://github.com/pytorch/pytorch/issues/71889,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:42:24Z +issue,71919,Graph Mode Quantization does not keep NamedTuple Information,https://github.com/pytorch/pytorch/issues/71919,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: quantization",scan when time allows,2023-11-17T21:41:47Z +issue,73507,Implement SiLU method for the QuantizedCPU backend,https://github.com/pytorch/pytorch/issues/73507,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:39:36Z +issue,74393,Quantized cannot inference with cuda,https://github.com/pytorch/pytorch/issues/74393,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: quantization",scan when time allows,2023-11-17T21:37:52Z +issue,76304,Quantization fails when padding parameter given as string,https://github.com/pytorch/pytorch/issues/76304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:33:25Z +issue,75785,"[ONNX] Expected quantizer->qscheme() == kPerTensorAffine to be true, but got false.",https://github.com/pytorch/pytorch/issues/75785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:32:00Z +issue,74953,LSTM quantization fails if proj_size > 0,https://github.com/pytorch/pytorch/issues/74953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:30:46Z +issue,76755,Quantization in Libtorch,https://github.com/pytorch/pytorch/issues/76755,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:28:10Z +issue,83929,ModuleNotFoundError: No module named 'torch.ao.quantization.experimental',https://github.com/pytorch/pytorch/issues/83929,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:20:25Z +issue,84064,Better error message for qlinear_prepack,https://github.com/pytorch/pytorch/issues/84064,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:14:41Z +issue,84178,Support setting strides on quantized weights of Embedding,https://github.com/pytorch/pytorch/issues/84178,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T21:13:16Z +issue,85745,[Quant] Remove or clarify the meaning of Nones in QConfig/BackendConfig,https://github.com/pytorch/pytorch/issues/85745,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:59:20Z +issue,85851,Print a warning when user specifies a qconfig for some node and the qconfig is not supported by BackendConfig,https://github.com/pytorch/pytorch/issues/85851,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:59:18Z +issue,86848,.view(dtype) on a quantized tensor throws SegmentationFault,https://github.com/pytorch/pytorch/issues/86848,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:54:05Z +issue,87126,Bug in Histogram Observer Implementation,https://github.com/pytorch/pytorch/issues/87126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:45:38Z +issue,87777,Improvements to fuse optimization,https://github.com/pytorch/pytorch/issues/87777,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: quantization",scan when time allows,2023-11-17T20:29:39Z +issue,88505,quantization: error message when using `convert_fx` on a model on cuda should be better,https://github.com/pytorch/pytorch/issues/88505,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:28:01Z +issue,88579,[Quant] Validate FixedQParams observers in eager mode,https://github.com/pytorch/pytorch/issues/88579,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:26:19Z +issue,88800,Quantization error between fake-quantized model and quantized model using the new observer,https://github.com/pytorch/pytorch/issues/88800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:21:16Z +issue,89303,code sharing for fundamental ops in quantization,https://github.com/pytorch/pytorch/issues/89303,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:11:46Z +issue,89357,Quantizable LSTM has different behavior than LSTM in bidirectional setting,https://github.com/pytorch/pytorch/issues/89357,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rnn, oncall: quantization",scan when time allows,2023-11-17T20:10:47Z +issue,90288,quantization observers: can we relax the default epsilon value?,https://github.com/pytorch/pytorch/issues/90288,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T20:02:02Z +issue,90424,A few functions in fbgemm_utils.cpp are defined in global namespace,https://github.com/pytorch/pytorch/issues/90424,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: quantization",scan when time allows,2023-11-17T19:56:01Z +issue,90553,Embedding dynamic quantization is not documented and hard to use,https://github.com/pytorch/pytorch/issues/90553,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-17T19:52:53Z +issue,113941,"libtorch_cpu.so: Undefined symbol ""__assert_fail""",https://github.com/pytorch/pytorch/issues/113941,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-17T18:48:23Z +issue,113847,RuntimeError in use torch 2.1.0 cuda 11.8,https://github.com/pytorch/pytorch/issues/113847,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-11-17T17:51:35Z +issue,113831,Update _scaled_mm to support addmm with bias.size == out.size,https://github.com/pytorch/pytorch/issues/113831,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: floatx (formerly float8),scan when time allows,2023-11-17T17:34:42Z +issue,113942,Tensorboard list of tensors as input,https://github.com/pytorch/pytorch/issues/113942,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-11-17T17:14:48Z +issue,104195,Torchscript with dynamic quantization produces inconsistent model outputs,https://github.com/pytorch/pytorch/issues/104195,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: quantization",scan when time allows,2023-11-17T15:53:55Z +issue,108224,qnnpack quantized model can not be traced,https://github.com/pytorch/pytorch/issues/108224,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: quantization",scan when time allows,2023-11-17T15:33:39Z +issue,113748,New feature: Balanced Sampler,https://github.com/pytorch/pytorch/issues/113748,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2023-11-16T15:11:19Z +issue,113621,App crashes when I attempt to run it with an iOS xcframework that relies on the LibTorch-Lite CocoaPod,https://github.com/pytorch/pytorch/issues/113621,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: ios",scan when time allows,2023-11-16T02:06:34Z +issue,1489,Optimizer should track parameter names and not id,https://github.com/pytorch/pytorch/issues/1489,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-11-15T19:22:16Z +issue,31742,sparse tensor eliminate_zeros,https://github.com/pytorch/pytorch/issues/31742,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-11-15T14:44:58Z +issue,11202,[pytorch][feature request] Cosine distance / simialrity between samples of own tensor or two tensors,https://github.com/pytorch/pytorch/issues/11202,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: nn, module: numpy",scan when time allows,2023-11-15T11:10:12Z +issue,98860,Sharded Grad Scaler Issue Tracker,https://github.com/pytorch/pytorch/issues/98860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: fsdp, oncall: distributed",scan when time allows,2023-11-15T01:16:48Z +issue,107016,cov onnx error,https://github.com/pytorch/pytorch/issues/107016,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2023-11-15T00:14:39Z +issue,113579,Creating Gaussian Mixture Models with MultivariateNormal,https://github.com/pytorch/pytorch/issues/113579,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: nn",scan when time allows,2023-11-14T18:04:08Z +issue,35642,Caching Support for class Dataset,https://github.com/pytorch/pytorch/issues/35642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-11-14T15:31:59Z +issue,46705,torch.eye not supported for sparse layout,https://github.com/pytorch/pytorch/issues/46705,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-11-14T13:10:27Z +issue,113550,CUDA extension error message doesn't look correct,https://github.com/pytorch/pytorch/issues/113550,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda",scan when time allows,2023-11-14T02:49:07Z +issue,113525,[Feature request] `stft` doesn't have `pad_value` argument,https://github.com/pytorch/pytorch/issues/113525,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2023-11-14T01:46:21Z +issue,113515,torchrun discarding --rdzv-endpoint when it should not,https://github.com/pytorch/pytorch/issues/113515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: elastic,scan when time allows,2023-11-14T01:44:18Z +issue,112833,Failed to remove output annotation in Quantizer for PT2 QAT quantization,https://github.com/pytorch/pytorch/issues/112833,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-11-14T01:02:31Z +issue,73008,Followup requires for MKL link issue / cannot find -lmkl_core,https://github.com/pytorch/pytorch/issues/73008,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",scan when time allows,2023-11-13T23:27:19Z +issue,104502,Get errors after compiling and running PyTorch MINIMAL EXAMPLE for c++ Mac M1 with make,https://github.com/pytorch/pytorch/issues/104502,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: cpp, module: macos",scan when time allows,2023-11-13T19:20:35Z +issue,73487,PyTorch not recognizing GPU on WSL - installed cudnn and cuda,https://github.com/pytorch/pytorch/issues/73487,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: wsl",scan when time allows,2023-11-13T09:50:41Z +issue,72814,Standardize Naming for Workflows/Jobs,https://github.com/pytorch/pytorch/issues/72814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2023-11-11T01:42:15Z +issue,109108,Cannot construct `torch.sparse_coo_tensor` (but `scipy.sparse.coo_matrix` works fine): `TypeError: only integer tensors of a single element can be converted to an index`,https://github.com/pytorch/pytorch/issues/109108,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: scipy compatibility, module: sparse",scan when time allows,2023-11-10T23:29:35Z +issue,113449,cuDNN error: CUDNN_STATUS_MAPPING_ERROR on gtx_1080/A10 when conv1d is called,https://github.com/pytorch/pytorch/issues/113449,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2023-11-10T15:51:42Z +issue,113369,When will Huawei Shengteng atlas be supported,https://github.com/pytorch/pytorch/issues/113369,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2023-11-10T15:37:47Z +issue,113382,TorchXLA - owner @JackCaoG,https://github.com/pytorch/pytorch/issues/113382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xla,scan when time allows,2023-11-10T14:16:56Z +issue,113368,Gradient of image rotation,https://github.com/pytorch/pytorch/issues/113368,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2023-11-10T14:12:57Z +issue,113326,`torch._C._cuda_getDeviceCount` inflates system memory usage,https://github.com/pytorch/pytorch/issues/113326,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-11-09T15:13:49Z +issue,97760,Test Failure: TestUnaryUfuncsCPU.test_reference_numerics_normal_cos_cpu_float32 on s390x,https://github.com/pytorch/pytorch/issues/97760,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-11-08T10:37:09Z +issue,98904,Test failure: TestCommonCPU.test_python_ref__refs_abs_cpu_complex32,https://github.com/pytorch/pytorch/issues/98904,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2023-11-08T10:33:14Z +issue,98566,`F.interpolate` and `F.grid_sample` - documentation error and bug,https://github.com/pytorch/pytorch/issues/98566,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-11-08T09:29:22Z +issue,81868,grid_sample and mode='bilinear' induces errors at discrete pixel locations,https://github.com/pytorch/pytorch/issues/81868,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-11-08T09:20:38Z +issue,113007,`make_fx` failing for `_scaled_dot_product_flash_attention` decomposition,https://github.com/pytorch/pytorch/issues/113007,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, module: fx",scan when time allows,2023-11-08T04:10:24Z +issue,112849,CSR matrix on MPS,https://github.com/pytorch/pytorch/issues/112849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: sparse",scan when time allows,2023-11-07T19:35:06Z +issue,108023,[FSDP] Ignored modules on meta device seem to be initialized on CUDA device,https://github.com/pytorch/pytorch/issues/108023,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2023-11-07T17:01:52Z +issue,113034,make_fx produces incorrect graph when used under FunctionalTensorMode,https://github.com/pytorch/pytorch/issues/113034,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: fx",scan when time allows,2023-11-07T04:17:32Z +issue,112876,How to handle CVE vulnerabilities in underlying operating system?,https://github.com/pytorch/pytorch/issues/112876,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2023-11-06T22:34:04Z +issue,98814,autocast does not work properly on embedding module,https://github.com/pytorch/pytorch/issues/98814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-11-06T19:27:35Z +issue,112847,High dimensional grid sample,https://github.com/pytorch/pytorch/issues/112847,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-11-06T15:17:41Z +issue,88943,`torch.vander` Trigger RuntimeError with UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: linear algebra, module: sanitizers",scan when time allows,2023-11-06T00:55:45Z +issue,27617,[feature request] [dataloader] Pad variable-sized tensors in default_collate,https://github.com/pytorch/pytorch/issues/27617,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-11-03T19:28:11Z +issue,30401,Unexpected difference torch.multiprocessing.manager.queue and torch.multiprocessing.queue,https://github.com/pytorch/pytorch/issues/30401,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2023-11-03T19:15:14Z +issue,112775,Multi-Threaded GraphModule / torch.fx inference raises an exception,https://github.com/pytorch/pytorch/issues/112775,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: multiprocessing",scan when time allows,2023-11-03T17:31:32Z +issue,38650,Implement torch.erf for complex dtypes,https://github.com/pytorch/pytorch/issues/38650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2023-11-03T16:39:56Z +issue,41970,torch.random.fork tries to initialize cuda even when no cuda devices are available,https://github.com/pytorch/pytorch/issues/41970,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: random",scan when time allows,2023-11-03T13:39:50Z +issue,88621,"[FSDP] FSDP produces different gradient norms vs DDP, and w/ grad norm clipping creates different training results",https://github.com/pytorch/pytorch/issues/88621,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-11-03T13:00:52Z +issue,112714,log_softmax() on CPU and GPU has expected numerical error when used with low-precision bfloat16,https://github.com/pytorch/pytorch/issues/112714,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: numerical-stability",scan when time allows,2023-11-03T02:46:10Z +issue,80259,Cudnn batch norm kernel (batchnorm_bwtr_nhwc_semiPersist) gets blocked by overlapping NCCL all_reduce calls,https://github.com/pytorch/pytorch/issues/80259,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: dependency bug, module: memory format, module: nccl",scan when time allows,2023-11-02T22:36:30Z +issue,112691,when huawei NPU,https://github.com/pytorch/pytorch/issues/112691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2023-11-02T14:34:29Z +issue,112666,dispatcher cannot determine dispatch key on tuple input,https://github.com/pytorch/pytorch/issues/112666,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2023-11-02T14:14:27Z +issue,112715,Wrong with code_coverage/readme.md,https://github.com/pytorch/pytorch/issues/112715,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-11-02T14:05:54Z +issue,109112,Unable to install the latest version of PyTorch using mamba.,https://github.com/pytorch/pytorch/issues/109112,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:32Z +issue,100366,some of the enteries in the previous version of pytorch section are invalid,https://github.com/pytorch/pytorch/issues/100366,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docs",scan when time allows,2023-11-02T13:58:29Z +issue,103578,"ImportError: undefined symbol: cublasSetWorkspace_v2, version libcublas.so.11",https://github.com/pytorch/pytorch/issues/103578,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2023-11-02T13:58:28Z +issue,99637,Torch hangs at import if tensorflow is imported first,https://github.com/pytorch/pytorch/issues/99637,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:28Z +issue,98361,pip doesn't install the right version of pytorch when torchtext is involved,https://github.com/pytorch/pytorch/issues/98361,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:27Z +issue,96060,Linking libtorch with QT5 OpenGL application using llvmpipe mesa opengl crashes,https://github.com/pytorch/pytorch/issues/96060,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:26Z +issue,92292,"""Get Started"" tells us to use the anaconda installer for PyTorch 3.x - but this should be python 3.x",https://github.com/pytorch/pytorch/issues/92292,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:26Z +issue,92273,Always install cpu version automatically,https://github.com/pytorch/pytorch/issues/92273,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:25Z +issue,80857,Compatibility List,https://github.com/pytorch/pytorch/issues/80857,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docs",scan when time allows,2023-11-02T13:58:22Z +issue,78681,"Unable to install Preview (Nightly) on M1 macOS: ""Symbol not found""",https://github.com/pytorch/pytorch/issues/78681,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:22Z +issue,75158,Offical libtorch (pytorch c++ frontend) docker image,https://github.com/pytorch/pytorch/issues/75158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-02T13:58:21Z +issue,47439,how to use torch.utils.checkpoint + gru with variable length sequence?,https://github.com/pytorch/pytorch/issues/47439,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2023-11-02T13:26:34Z +issue,92226,Hijacked package names from nightly repository,https://github.com/pytorch/pytorch/issues/92226,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-11-01T21:29:24Z +issue,27970,can't load model on cuda after call cudaDeviceReset functions,https://github.com/pytorch/pytorch/issues/27970,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-11-01T07:13:29Z +issue,50198,deterministic implementation for adaptive_avg_pool2d_backward_cuda,https://github.com/pytorch/pytorch/issues/50198,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism, module: pooling",scan when time allows,2023-11-01T05:28:01Z +issue,111931,[opcheck] Faster gradcheck execution,https://github.com/pytorch/pytorch/issues/111931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: opcheck,scan when time allows,2023-10-31T20:27:32Z +issue,111654,"Static Linking C++, Op not available at runtime",https://github.com/pytorch/pytorch/issues/111654,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: vision",scan when time allows,2023-10-31T18:40:10Z +issue,25032,[FYI] NestedTensor Project Progress,https://github.com/pytorch/pytorch/issues/25032,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-10-31T17:57:12Z +issue,20655,nn.Embedding backwards slow under high row contention,https://github.com/pytorch/pytorch/issues/20655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn, module: performance",scan when time allows,2023-10-31T11:35:00Z +issue,40024,torch::jit::load -> Unhandled exception,https://github.com/pytorch/pytorch/issues/40024,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-10-31T10:12:26Z +issue,94233,unsqueeze a single dimension multiple times,https://github.com/pytorch/pytorch/issues/94233,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2023-10-30T23:59:21Z +issue,2732,support grid_sample with batch=1 but supprting batch affine parameters,https://github.com/pytorch/pytorch/issues/2732,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: vision",scan when time allows,2023-10-30T22:32:14Z +issue,104933,torch.func.jvp fails with BERT training,https://github.com/pytorch/pytorch/issues/104933,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, module: functorch",scan when time allows,2023-10-30T19:35:16Z +issue,112318,IBM AfroHacks at AfroTech Participation,https://github.com/pytorch/pytorch/issues/112318,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-10-30T18:18:16Z +issue,112342,[pytree] Pytree node registration hygeine: deprecate global _register_pytree_node; only allow enabling registered pytree extensions locally,https://github.com/pytorch/pytorch/issues/112342,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pytree,scan when time allows,2023-10-30T15:27:44Z +issue,112330,Library is included twice QNNPACK,https://github.com/pytorch/pytorch/issues/112330,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: mobile",scan when time allows,2023-10-29T16:34:54Z +issue,29981,[feature request] Multivariate normal CDF,https://github.com/pytorch/pytorch/issues/29981,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-10-29T04:41:33Z +issue,111907,Migration from c10::variant to std::variant causes undefined symbols when linking against older pytorch,https://github.com/pytorch/pytorch/issues/111907,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-10-28T00:58:47Z +issue,103962,How to unwrap after auto_wrap in FSDP?,https://github.com/pytorch/pytorch/issues/103962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2023-10-27T15:16:22Z +issue,112251,Optest is unable to xfail on parametrized tests,https://github.com/pytorch/pytorch/issues/112251,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: opcheck,scan when time allows,2023-10-27T14:26:56Z +issue,109440,[FSDP] supports QLora finetuning,https://github.com/pytorch/pytorch/issues/109440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2023-10-27T14:19:04Z +issue,110737,pytorch consuming all cpu cores 100% on ARM,https://github.com/pytorch/pytorch/issues/110737,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: openmp",scan when time allows,2023-10-26T17:55:07Z +issue,112013,RuntimeError for hessian vector product with jvp,https://github.com/pytorch/pytorch/issues/112013,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-10-26T02:23:17Z +issue,49851,[Feature Request] SuperLoss (NeurIPS 2020),https://github.com/pytorch/pytorch/issues/49851,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2023-10-26T02:11:38Z +issue,90544,[Dispatchable Collectives] Follow up tasks,https://github.com/pytorch/pytorch/issues/90544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-10-25T20:17:03Z +issue,91863,torch.Categorical samples indexes with 0 probability when given logits as argument,https://github.com/pytorch/pytorch/issues/91863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-10-25T13:50:01Z +issue,111789,Precisely monitor the collective communication tasks,https://github.com/pytorch/pytorch/issues/111789,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: profiler",scan when time allows,2023-10-25T02:50:40Z +issue,111925,Resize warning in two argument torch.logical_* with broadcasting,https://github.com/pytorch/pytorch/issues/111925,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: safe resize",scan when time allows,2023-10-24T17:21:36Z +issue,111930,[opcheck] Way to reduce Hypothesis sampling when running opcheck,https://github.com/pytorch/pytorch/issues/111930,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: opcheck,scan when time allows,2023-10-24T17:03:12Z +issue,111924,[opcheck] Cannot share failures_dict between multiple tests with differing sets of tests they run,https://github.com/pytorch/pytorch/issues/111924,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: opcheck,scan when time allows,2023-10-24T15:37:29Z +issue,61816,Better argument names for torch.atan2 and other math functions,https://github.com/pytorch/pytorch/issues/61816,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-10-24T08:17:22Z +issue,87350,OpenCL 3.0 support: support every GPU on earth through rusticl,https://github.com/pytorch/pytorch/issues/87350,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-10-24T01:38:43Z +issue,111739,grad is inf/nan when using torch.amp,https://github.com/pytorch/pytorch/issues/111739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-10-23T23:56:51Z +issue,111819,__slots__ + inheriting from torch.Tensor,https://github.com/pytorch/pytorch/issues/111819,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-10-23T23:31:50Z +issue,110966,`torch.is_autocast_enabled()` always False on CPU,https://github.com/pytorch/pytorch/issues/110966,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-10-23T17:23:24Z +issue,107538,[fx] tracing function with in-place mutation results in unexpected behaviour due to local vars becoming persisted in `GraphModule(nn.Module)`,https://github.com/pytorch/pytorch/issues/107538,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-10-22T01:08:11Z +issue,56978,Do padding of weight and activation tensors to match optimized backend implementation,https://github.com/pytorch/pytorch/issues/56978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-10-21T07:25:57Z +issue,111623,Missing `ignored_param` when calling wrapper_cls (FSDP) recursively,https://github.com/pytorch/pytorch/issues/111623,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-10-21T06:50:53Z +issue,109791,Heap-buffer-overflow during tensor unpickling,https://github.com/pytorch/pytorch/issues/109791,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-10-20T18:24:42Z +issue,111525,Functorch FCD breaks with tensor subclasses,https://github.com/pytorch/pytorch/issues/111525,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: first class dims, module: functorch",scan when time allows,2023-10-20T17:12:28Z +issue,111552,[Bug]: some parameters' grad is None when using FSDP with torch2.1.0,https://github.com/pytorch/pytorch/issues/111552,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2023-10-20T12:52:09Z +issue,96608,No matching distribution found for torch==1.13.1+cu117,https://github.com/pytorch/pytorch/issues/96608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-10-20T07:26:55Z +issue,111566,build: failure when building pytorch with TBB,https://github.com/pytorch/pytorch/issues/111566,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: tbb",scan when time allows,2023-10-20T06:17:35Z +issue,111564,misusing percision value in test_cuda function in torch/testing/_internal/common_nn.py.,https://github.com/pytorch/pytorch/issues/111564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2023-10-20T04:36:12Z +issue,110698,max_pool3d_with_indices_backward_cuda and avg_pool3d_backward_cuda does not have a deterministic implementation,https://github.com/pytorch/pytorch/issues/110698,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: nn",scan when time allows,2023-10-20T02:41:40Z +issue,84193,Attach execution time to each node in an fx trace,https://github.com/pytorch/pytorch/issues/84193,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-10-19T20:00:23Z +issue,55363,it is lack of compatibility of CUDA_HOME (or other env),https://github.com/pytorch/pytorch/issues/55363,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda",scan when time allows,2023-10-19T14:06:48Z +issue,87696,Enable PostLocalSGDOptimizer on CUDA tensors,https://github.com/pytorch/pytorch/issues/87696,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2023-10-19T12:40:30Z +issue,86537,Running JIT trace for many times leads to OOM,https://github.com/pytorch/pytorch/issues/86537,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, oncall: jit",scan when time allows,2023-10-19T11:40:40Z +issue,110959,`model.named_buffers()` fails if module not hashable.,https://github.com/pytorch/pytorch/issues/110959,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-10-19T08:56:45Z +issue,111466,yolov5_train,https://github.com/pytorch/pytorch/issues/111466,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism",scan when time allows,2023-10-19T04:04:29Z +issue,47363,FX quantization: we should preserve original model class name through the quantization passes,https://github.com/pytorch/pytorch/issues/47363,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-10-18T21:49:10Z +issue,67773,Instability in `test_input_weight_equalization_activation_values` test for random test values.,https://github.com/pytorch/pytorch/issues/67773,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: quantization",scan when time allows,2023-10-18T20:58:26Z +issue,60008,Dynamic quantized LSTM does not support output projection now.,https://github.com/pytorch/pytorch/issues/60008,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-10-18T20:55:41Z +issue,59730,Quantized conv2d with dilation and groups much slower than float32,https://github.com/pytorch/pytorch/issues/59730,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: performance, oncall: quantization",scan when time allows,2023-10-18T20:55:10Z +issue,110516,Torch Nested Issue With Backward Pass In Transpose,https://github.com/pytorch/pytorch/issues/110516,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-10-18T16:01:03Z +issue,111410,"Use of -Wl,--as-needed in cmake config files can leak into third-party users' code and modify their own private libraries",https://github.com/pytorch/pytorch/issues/111410,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-10-18T14:44:39Z +issue,111210,"RuntimeError: Expected is_sm80 || is_sm90 to be true, but got false. (Using Google Colab)",https://github.com/pytorch/pytorch/issues/111210,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-10-17T16:56:36Z +issue,102142,Pytorch CXX11 ABI version,https://github.com/pytorch/pytorch/issues/102142,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: build",scan when time allows,2023-10-17T11:53:28Z +issue,111363,Simulating lower memory on GPU does not indicate simulated memory in error message,https://github.com/pytorch/pytorch/issues/111363,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: memory usage",scan when time allows,2023-10-17T06:36:29Z +issue,110252,cuda/tf32 docs are outdated,https://github.com/pytorch/pytorch/issues/110252,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: tf32",scan when time allows,2023-10-16T21:01:40Z +issue,111311,There is a performance drop because we have not yet implemented the batching rule for aten::mkldnn_rnn_layer_backward.,https://github.com/pytorch/pytorch/issues/111311,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-10-16T20:09:38Z +issue,111349,CUDA version 12.2 has differential accuracy when executing CPU and GPU,https://github.com/pytorch/pytorch/issues/111349,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-10-16T19:56:46Z +issue,111351,_foreach_copy_ supports fast copy between cpu and cuda devices.,https://github.com/pytorch/pytorch/issues/111351,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mta,scan when time allows,2023-10-16T19:50:28Z +issue,111366,RuntimeError: CUDAPluggableAllocator does not yet support cacheInfo,https://github.com/pytorch/pytorch/issues/111366,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudacachingallocator,scan when time allows,2023-10-16T19:44:08Z +issue,111186,[ONNX][Exporter] Maintain support for exporter arguments export_params and keep_initializers_as_inputs,https://github.com/pytorch/pytorch/issues/111186,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2023-10-16T17:37:32Z +issue,25646,libtorch forward memory leak,https://github.com/pytorch/pytorch/issues/25646,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: memory usage",scan when time allows,2023-10-16T02:57:10Z +issue,110795,Build process failure with torch_shm_manager,https://github.com/pytorch/pytorch/issues/110795,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-10-15T19:23:05Z +issue,103444,Deprecation warning on lr_scheduler.step(num_steps),https://github.com/pytorch/pytorch/issues/103444,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-10-13T20:11:00Z +issue,111157,overloads can perhaps be more performant?,https://github.com/pytorch/pytorch/issues/111157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: dispatch, module: optimizer",scan when time allows,2023-10-13T19:44:53Z +issue,5405,scatter_add_ should support scalar source (including Python scalar),https://github.com/pytorch/pytorch/issues/5405,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2023-10-12T18:22:39Z +issue,110363,"nn.BatchNorm2d (track_running_stats = True) causes ""modified by an in-place operation"" error when in torch.nn.parallel.DistributedDataParallel",https://github.com/pytorch/pytorch/issues/110363,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-10-12T17:07:19Z +issue,111025,Unprompted UserWarning,https://github.com/pytorch/pytorch/issues/111025,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-10-12T13:15:30Z +issue,111024,ncu python conv2d.py runs indefinitely after activating cudnn.benchmark,https://github.com/pytorch/pytorch/issues/111024,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2023-10-12T13:15:04Z +issue,111126,type promotion test for torch.div variants is broken,https://github.com/pytorch/pytorch/issues/111126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2023-10-12T13:02:20Z +issue,94371,"QAT + torch.autocast does not work with default settings, missing fused fake_quant support for half",https://github.com/pytorch/pytorch/issues/94371,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-10-12T03:54:17Z +issue,80067,"FakeTensor: Support torch.tensor([FakeTensor, 0])",https://github.com/pytorch/pytorch/issues/80067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2023-10-12T00:54:59Z +issue,80580,`torch.ops.aten.find` inconsistent with `str.find`,https://github.com/pytorch/pytorch/issues/80580,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: sorting and selection",scan when time allows,2023-10-11T09:17:57Z +issue,105778,https://pytorch.org/docs/stable/backends.html does not describe torch.backends.cpu,https://github.com/pytorch/pytorch/issues/105778,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: docs",scan when time allows,2023-10-10T23:47:07Z +issue,108484,DistributedDataParallel to support __getattr__,https://github.com/pytorch/pytorch/issues/108484,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-10-09T21:26:13Z +issue,28329,No in-place version of where(),https://github.com/pytorch/pytorch/issues/28329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2023-10-09T13:25:08Z +issue,107394,[LibTorch/iOS] Unknown custom class type quantized.Conv2dPackedParamsBase. Please ensure it is registered,https://github.com/pytorch/pytorch/issues/107394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",scan when time allows,2023-10-09T07:51:12Z +issue,16330,support `unique_indices` option for `unique`,https://github.com/pytorch/pytorch/issues/16330,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-10-08T19:02:49Z +issue,110786,.lldbinit formatters only work when building with clang,https://github.com/pytorch/pytorch/issues/110786,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-10-07T05:55:14Z +issue,110605,ValueError issued instead of TypeError when tensor is cast to a scalar,https://github.com/pytorch/pytorch/issues/110605,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: numpy",scan when time allows,2023-10-06T20:41:36Z +issue,39310,Make torch.cross `dim` parameter work intuitively,https://github.com/pytorch/pytorch/issues/39310,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: numpy",scan when time allows,2023-10-05T22:41:22Z +issue,41027,out= resizing (and restriding) behavior is confusing,https://github.com/pytorch/pytorch/issues/41027,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: numpy, module: safe resize",scan when time allows,2023-10-05T22:41:12Z +issue,46291,ProcessGroup::Work API changes,https://github.com/pytorch/pytorch/issues/46291,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: c10d",scan when time allows,2023-10-05T22:40:59Z +issue,50292,Implement dunder getattribute instead of dunder getattr for nn.Module,https://github.com/pytorch/pytorch/issues/50292,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: nn",scan when time allows,2023-10-05T22:40:41Z +issue,50342,torch.any and torch.all map uint8 -> uint8 but should map uint8 -> bool,https://github.com/pytorch/pytorch/issues/50342,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: numpy, module: reductions",scan when time allows,2023-10-05T22:40:31Z +issue,50530,Unable to compile CUDAExtension with Pytorch 1.5 or above,https://github.com/pytorch/pytorch/issues/50530,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: cpp-extensions, module: cuda",scan when time allows,2023-10-05T22:40:25Z +issue,61574,[nn] Passing dtype to `_stacklevel` argument in `log_softmax` silently works,https://github.com/pytorch/pytorch/issues/61574,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: deprecation, module: functional ux, module: nn",scan when time allows,2023-10-05T22:33:05Z +issue,68464,"Changing the type of dtype argument in softmax, log_softmax, and softmin is possibly BC-breaking.",https://github.com/pytorch/pytorch/issues/68464,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: norms and normalization",scan when time allows,2023-10-05T22:32:25Z +issue,70921,`sort` should only return the sorted input,https://github.com/pytorch/pytorch/issues/70921,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: python array api, module: sorting and selection",scan when time allows,2023-10-05T22:32:18Z +issue,70924,change supported arguments for parameter `dim` in `squeeze`,https://github.com/pytorch/pytorch/issues/70924,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: python array api",scan when time allows,2023-10-05T22:32:11Z +issue,72317,Feature request: a mode to disallow calling prototype or beta features,https://github.com/pytorch/pytorch/issues/72317,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: python frontend",scan when time allows,2023-10-05T22:32:06Z +issue,79195,"[bug] fill_, masked_fill_ : fill ops allow lossy downcasting of fill value",https://github.com/pytorch/pytorch/issues/79195,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: primtorch",scan when time allows,2023-10-05T22:31:50Z +issue,92331,backward(inputs= does not need to execute grad_fn of the inputs,https://github.com/pytorch/pytorch/issues/92331,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: bc-breaking",scan when time allows,2023-10-05T22:31:34Z +issue,95116,"When using `ceil_mode=True`, `torch.nn.AvgPool1d` could get negative shape.",https://github.com/pytorch/pytorch/issues/95116,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: shape checking",scan when time allows,2023-10-05T22:31:26Z +issue,95711,`copy.deepcopy` does not copy gradients of nn.Parameter,https://github.com/pytorch/pytorch/issues/95711,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: bc-breaking, module: nn",scan when time allows,2023-10-05T22:31:21Z +issue,109781,"`torch.embedding`, `weight[indices]`, `torch.index_select` returns random data with indices on meta device",https://github.com/pytorch/pytorch/issues/109781,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: embedding, module: meta tensors",scan when time allows,2023-10-05T22:29:17Z +issue,71409,Random Shuffle along Axis,https://github.com/pytorch/pytorch/issues/71409,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2023-10-05T14:07:24Z +issue,110461,Custom tensor attributes not preserved with registered functions,https://github.com/pytorch/pytorch/issues/110461,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: library",scan when time allows,2023-10-05T01:33:32Z +issue,79145,Redundant info are saved when using torch.save to save part of torch.tensor,https://github.com/pytorch/pytorch/issues/79145,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-10-04T23:21:52Z +issue,55279,[Feature Pitch] Full-batch optimization toolkit,https://github.com/pytorch/pytorch/issues/55279,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-10-04T20:26:57Z +issue,110366,Categorical Simplex constraint throws error for valid values,https://github.com/pytorch/pytorch/issues/110366,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-10-04T19:52:18Z +issue,110356,"Dropout signature inconsistent between `torch.dropout`, `torch.nn.Dropout` and `torch.nn.functional.dropout`",https://github.com/pytorch/pytorch/issues/110356,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend",scan when time allows,2023-10-04T19:52:10Z +issue,104296,affine_grid and grid_sample operators merge/accelleration,https://github.com/pytorch/pytorch/issues/104296,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2023-10-04T11:04:08Z +issue,85098,install libtorch cxx11 ABI as default in PyTorch pip installation,https://github.com/pytorch/pytorch/issues/85098,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-10-03T18:31:26Z +issue,71721,Improve performance of index for quantized ops,https://github.com/pytorch/pytorch/issues/71721,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-10-03T00:37:01Z +issue,110249,`torch.func.functional_call` does not work with `__torch_function__ ` Tensor-like objects,https://github.com/pytorch/pytorch/issues/110249,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: functorch",scan when time allows,2023-10-02T20:47:41Z +issue,108277,Transformer performance drop due to slow PyTorch GEMMs,https://github.com/pytorch/pytorch/issues/108277,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2023-10-02T13:01:11Z +issue,98939,"torch.dist with minus norm returns tensor(0.), while with -inf can return result",https://github.com/pytorch/pytorch/issues/98939,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-10-02T07:44:31Z +issue,110238,PyTorch with non-shared build (building a single shared lib) is unsupported,https://github.com/pytorch/pytorch/issues/110238,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: static linking",scan when time allows,2023-09-30T10:36:12Z +issue,103570,File Missing When i build with C++,https://github.com/pytorch/pytorch/issues/103570,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-09-29T19:34:56Z +issue,110130,Create a new heuristic TD rule for failures coming from base commit of the pull requests,https://github.com/pytorch/pytorch/issues/110130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2023-09-29T16:33:39Z +issue,110194,cudaMallocAsync cause too much fragmentation.,https://github.com/pytorch/pytorch/issues/110194,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: memory usage",scan when time allows,2023-09-28T21:59:32Z +issue,47688,ARM Mac 16-core Neural Engine,https://github.com/pytorch/pytorch/issues/47688,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-09-28T20:03:39Z +issue,104188,Add memory managemenet information for Apple silicon mps backend,https://github.com/pytorch/pytorch/issues/104188,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2023-09-28T14:47:04Z +issue,110156,Add _worker_end_fn_t to the DataLoader,https://github.com/pytorch/pytorch/issues/110156,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2023-09-28T14:06:36Z +issue,109497,Very big differences in output of `torch.lobpcg` (values and run-time) compared to SciPy on a very ill-conditioned Laplacian matrix,https://github.com/pytorch/pytorch/issues/109497,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2023-09-28T09:53:54Z +issue,110116,Skip cuda kernel launch with torch.sum when dimension length is 0,https://github.com/pytorch/pytorch/issues/110116,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2023-09-28T00:41:20Z +issue,110084,scatter_add: Mixing 0-dim and 1-dim tensors,https://github.com/pytorch/pytorch/issues/110084,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: edge cases, module: python frontend",scan when time allows,2023-09-28T00:40:32Z +issue,106596,[discussion] move-semantics for tensors,https://github.com/pytorch/pytorch/issues/106596,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-09-27T18:37:59Z +issue,109462,Inconsistent behavior for in-place operations on coalesced sparse tensors,https://github.com/pytorch/pytorch/issues/109462,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-09-27T16:06:38Z +issue,20102,LSTM forget bias must be initialized properly,https://github.com/pytorch/pytorch/issues/20102,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: initialization, module: nn, module: rnn",scan when time allows,2023-09-27T12:41:15Z +issue,109383,cuda rng state for 2.0.1 cannot be used for 2.1.0,https://github.com/pytorch/pytorch/issues/109383,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: random",scan when time allows,2023-09-26T21:26:42Z +issue,108174,FlashAttentionV2 will OOM when building on ci/cd with default settings,https://github.com/pytorch/pytorch/issues/108174,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda",scan when time allows,2023-09-26T20:05:42Z +issue,110014,tan/tanh discrepancies with complex due to jiterator,https://github.com/pytorch/pytorch/issues/110014,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: jiterator",scan when time allows,2023-09-26T11:23:19Z +issue,110032,Race condition on shutdown involving PThreadPool and autograd,https://github.com/pytorch/pytorch/issues/110032,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: sanitizers",scan when time allows,2023-09-26T10:40:53Z +issue,42080,To have single cuda context across multiple processes,https://github.com/pytorch/pytorch/issues/42080,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-09-26T02:16:46Z +issue,109191,Gradients across different ranks are not synchronized when using DDP,https://github.com/pytorch/pytorch/issues/109191,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-09-25T16:17:43Z +issue,109938,Batching for is_in,https://github.com/pytorch/pytorch/issues/109938,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-09-25T11:12:11Z +issue,109958,How to compile torch 2.0.1 version from source?,https://github.com/pytorch/pytorch/issues/109958,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2023-09-25T11:01:11Z +issue,109946,Indexed batch matrix multiplication to support MoEs and FFFs,https://github.com/pytorch/pytorch/issues/109946,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-09-25T10:58:40Z +issue,109968,Dtype hard-coded in DataLoader (for python floats).,https://github.com/pytorch/pytorch/issues/109968,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2023-09-25T00:33:16Z +issue,40471,please add 'tensor.astype(dtype_string)' syntax for numpy interoperability,https://github.com/pytorch/pytorch/issues/40471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-09-24T19:36:59Z +issue,24870,Recommendations for Grid Sample/Affine Grid/Displacement Fields/Optical Flow,https://github.com/pytorch/pytorch/issues/24870,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2023-09-24T08:02:28Z +issue,108870,"Inconsistent, platform-dependent torch.ones_like behavior on metatensors",https://github.com/pytorch/pytorch/issues/108870,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2023-09-23T21:43:49Z +issue,84348,"RuntimeError: ""slow_conv2d_cpu"" not implemented for 'Half'",https://github.com/pytorch/pytorch/issues/84348,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2023-09-23T14:56:08Z +issue,40989,Difference between allocated and reserved CUDA memory,https://github.com/pytorch/pytorch/issues/40989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2023-09-23T09:29:47Z +issue,109870,Wrongly returns nan for vectorized complex numbers division on PPC/ZArch,https://github.com/pytorch/pytorch/issues/109870,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: power,scan when time allows,2023-09-23T01:19:26Z +issue,109923,Import order issue with torch and pybind11 Library Statically Linked to libstdc++,https://github.com/pytorch/pytorch/issues/109923,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: static linking",scan when time allows,2023-09-23T01:19:21Z +issue,109934,test test_2d_fsdp_integration_fsdp_nested_param_groups failed,https://github.com/pytorch/pytorch/issues/109934,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-09-23T01:16:24Z +issue,109863,[BUG?] Why Allocator use stream to manage Block?,https://github.com/pytorch/pytorch/issues/109863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-09-23T01:11:49Z +issue,103499,CUBLAS_WORKSPACE_CONFIG can not be parsed,https://github.com/pytorch/pytorch/issues/103499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2023-09-22T18:48:15Z +issue,106931,Other overloads of `_foreach_clamp`,https://github.com/pytorch/pytorch/issues/106931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer",scan when time allows,2023-09-22T18:02:33Z +issue,106951,stride of gradient is not same as the corresponding tensor,https://github.com/pytorch/pytorch/issues/106951,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: optimizer",scan when time allows,2023-09-22T18:02:25Z +issue,106989,`ray` multiprocessing interference by torch import,https://github.com/pytorch/pytorch/issues/106989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2023-09-21T22:55:25Z +issue,109724,assert_is_valid_input_type is too weak,https://github.com/pytorch/pytorch/issues/109724,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2023-09-21T13:28:06Z +issue,109747,[RFC][TorchElastic] topology info in training apps/ranks,https://github.com/pytorch/pytorch/issues/109747,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2023-09-21T13:27:38Z +issue,107389,caffe does not respect CUDNN_LIB_DIR when building from source (cmake),https://github.com/pytorch/pytorch/issues/107389,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: nvfuser",scan when time allows,2023-09-21T02:19:44Z +issue,109460,[BUG][pytree] treespec serialization for locally defined classes and namedtuple types,https://github.com/pytorch/pytorch/issues/109460,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pytree,scan when time allows,2023-09-20T15:08:18Z +issue,108483,Efficient and robust calculation of diag(sparse @ diag @ sparse),https://github.com/pytorch/pytorch/issues/108483,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-09-20T13:07:15Z +issue,103581,Passing dict in datapipe/dataset will have memory leak problem,https://github.com/pytorch/pytorch/issues/103581,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2023-09-20T04:22:19Z +issue,109586,Max pool with negative integer inputs and channels_last memory layout gives the wrong values,https://github.com/pytorch/pytorch/issues/109586,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-09-19T15:16:45Z +issue,108645,uninformative OOM error,https://github.com/pytorch/pytorch/issues/108645,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-09-19T07:10:05Z +issue,51803,torch.fx.symbolic_trace fails on torch.arange with input-dependent size,https://github.com/pytorch/pytorch/issues/51803,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-09-19T04:22:35Z +issue,109494,Performance degradation on AMD + A800 when computation is small,https://github.com/pytorch/pytorch/issues/109494,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2023-09-19T03:49:50Z +issue,109386,Make Fx Generating Incorrect Graph For GPTQ model,https://github.com/pytorch/pytorch/issues/109386,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: proxytensor",scan when time allows,2023-09-18T19:47:40Z +issue,73738,torch.ao.quantization.ReuseInputObserver should also reuse the dtype of the input,https://github.com/pytorch/pytorch/issues/73738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-09-18T17:00:32Z +issue,73941,"[Quant] Framework observes weight in convert, changing numerics",https://github.com/pytorch/pytorch/issues/73941,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-09-18T16:59:07Z +issue,74112,torch.jit.script does not custom state_dicts,https://github.com/pytorch/pytorch/issues/74112,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: quantization",scan when time allows,2023-09-18T16:58:39Z +issue,55945,Quantizable LSTMCell does not work correctly.,https://github.com/pytorch/pytorch/issues/55945,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-09-18T16:52:52Z +issue,99225,Libtorch consumes too much memory as 16225,https://github.com/pytorch/pytorch/issues/99225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: memory usage",scan when time allows,2023-09-18T09:13:43Z +issue,87468,Categorical fails simplex validation after its own normalisation on CUDA,https://github.com/pytorch/pytorch/issues/87468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-09-17T03:13:15Z +issue,109009,TypeError: mask must have dtype bool,https://github.com/pytorch/pytorch/issues/109009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2023-09-16T19:43:41Z +issue,8242,Different behavior of LSTM and LSTMCell implementation,https://github.com/pytorch/pytorch/issues/8242,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility, module: numerical-stability, module: rnn",scan when time allows,2023-09-16T09:00:40Z +issue,43949,bytes(byte_tensor) gives strange error + [feature request] support memoryview(tensor),https://github.com/pytorch/pytorch/issues/43949,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: error checking",scan when time allows,2023-09-13T18:26:10Z +issue,70926,Feature Request: Implement `torch.sparse.spdiags` and `torch.sparse.diags`,https://github.com/pytorch/pytorch/issues/70926,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-09-13T16:06:48Z +issue,72897,Hanging Validation,https://github.com/pytorch/pytorch/issues/72897,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: ddp, module: deadlock",scan when time allows,2023-09-13T04:53:49Z +issue,108976,About FSDP,https://github.com/pytorch/pytorch/issues/108976,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-09-13T04:37:03Z +issue,108983,ZeroTensor (and probably neg/conj) doesn't play well with wrapper tensor subclasses,https://github.com/pytorch/pytorch/issues/108983,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2023-09-12T23:59:02Z +issue,50034,Multinomial without replacement produces samples that have zero probability,https://github.com/pytorch/pytorch/issues/50034,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: numpy, module: random, module: ux",scan when time allows,2023-09-12T21:43:05Z +issue,107581,[LibTorch/iOS] Building with METAL support script is freezing,https://github.com/pytorch/pytorch/issues/107581,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",scan when time allows,2023-09-12T20:47:26Z +issue,107627,ModuleNotFoundError: No module named 'torchgen.code_template',https://github.com/pytorch/pytorch/issues/107627,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: build, oncall: mobile",scan when time allows,2023-09-12T20:39:08Z +issue,104606,"Failure in optimize_for_mobile when using conv1d(..., padding='same')",https://github.com/pytorch/pytorch/issues/104606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,scan when time allows,2023-09-12T20:32:16Z +issue,105494,workaround for using vmap when .item() is being used internally,https://github.com/pytorch/pytorch/issues/105494,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2023-09-12T19:21:14Z +issue,90551,Could not run 'aten::as_strided' with arguments from the 'Metal' backend.,https://github.com/pytorch/pytorch/issues/90551,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ios,scan when time allows,2023-09-12T07:44:05Z +issue,35636,Caffe2 generate_proposals_op_gpu_test crashes on Windows,https://github.com/pytorch/pytorch/issues/35636,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2023-09-12T01:08:29Z +issue,102966,"how to workaround the error ""don't have an op for vulkan_prepack::create_linear_context"" ?",https://github.com/pytorch/pytorch/issues/102966,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vulkan",scan when time allows,2023-09-12T00:19:52Z +issue,74739,GradScaler support FP16 gradients?,https://github.com/pytorch/pytorch/issues/74739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-09-12T00:11:37Z +issue,76609,AttributeError: '_thread._local' object has no attribute 'rel_tol' (cannot use TestCase.assertEqual from other threads),https://github.com/pytorch/pytorch/issues/76609,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2023-09-11T19:28:51Z +issue,108882,Inconsistent any( ) between cuda and cpu - Incorrect complex to bool conversion,https://github.com/pytorch/pytorch/issues/108882,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: correctness (silent)",scan when time allows,2023-09-11T18:41:03Z +issue,108862,"A100 runners down: apt-get install nvidia-docker2, Could not get lock /var/lib/dpkg/lock-frontend",https://github.com/pytorch/pytorch/issues/108862,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2023-09-11T18:27:27Z +issue,109010,The following will always fail on NixOS,https://github.com/pytorch/pytorch/issues/109010,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2023-09-11T18:25:18Z +issue,108934,PPC64le: vsx_helpers.h errors,https://github.com/pytorch/pytorch/issues/108934,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-09-11T18:22:30Z +issue,108888,RuntimeError: Unrecognized tensor type ID: ZeroTensor,https://github.com/pytorch/pytorch/issues/108888,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-09-11T17:38:31Z +issue,108152,`Tensor.uniform_` uses illegal argument name `from`,https://github.com/pytorch/pytorch/issues/108152,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: python frontend",scan when time allows,2023-09-11T17:14:21Z +issue,39421,torch::jit::script::Module::to(torch::kDouble) also casts buffers,https://github.com/pytorch/pytorch/issues/39421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-09-11T14:22:35Z +issue,46253,Add short guide for updating `torch.norm` calls to `torch.linalg.norm`,https://github.com/pytorch/pytorch/issues/46253,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-09-11T11:42:31Z +issue,106959,"`tensor.repeat` quirks: has no `torch.` variant, no `out=` variant, no inplace variant | `torch.tile` also does not have `out=` variant and uses `dims=` instead of `dim=`",https://github.com/pytorch/pytorch/issues/106959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: viewing and reshaping",scan when time allows,2023-09-10T17:29:07Z +issue,87358,Triangular solver for sparse matrices,https://github.com/pytorch/pytorch/issues/87358,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2023-09-10T15:27:24Z +issue,24185,[feature request] Subset of eigenvalues/eigenvectors,https://github.com/pytorch/pytorch/issues/24185,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2023-09-10T12:22:04Z +issue,28119,Provide function similar to cdist that returns dist_p^p,https://github.com/pytorch/pytorch/issues/28119,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distance functions,scan when time allows,2023-09-10T11:04:25Z +issue,106011,Misleading error message in multilabel_margin_loss when passing incompatible tensor dimensions,https://github.com/pytorch/pytorch/issues/106011,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-09-10T05:20:07Z +issue,90920,Support for Transformer Models on Android with Vulkan Backend,https://github.com/pytorch/pytorch/issues/90920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2023-09-10T02:56:13Z +issue,108900,[c10d] fix functional collective reduce op naming convention,https://github.com/pytorch/pytorch/issues/108900,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2023-09-08T20:44:39Z +issue,94471,[Functionalization] `index_reduce_` op tests with functionalization enabled,https://github.com/pytorch/pytorch/issues/94471,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: meta tensors",scan when time allows,2023-09-08T17:43:10Z +issue,97210,"INTERNAL ASSERT FAILED at ""../c10/cuda/CUDAGraphsC10Utils.h"":73, please report a bug to PyTorch. Unknown CUDA graph CaptureStatus32729",https://github.com/pytorch/pytorch/issues/97210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2023-09-08T16:13:53Z +issue,108859,Certain torch functions are not handled by torch func wrapper,https://github.com/pytorch/pytorch/issues/108859,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-09-08T14:57:49Z +issue,108642,torch.topk returned values and indices are reordered if sorted=False,https://github.com/pytorch/pytorch/issues/108642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2023-09-08T14:53:36Z +issue,108671,avg_pool3d_backward fails on meta with grad_input parameter,https://github.com/pytorch/pytorch/issues/108671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2023-09-08T14:49:49Z +issue,108627,autocast not consistent across different GPUs (A100 and RTX A6000),https://github.com/pytorch/pytorch/issues/108627,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-09-08T14:47:21Z +issue,108651,libtorch: runtime error when iterating batch of dataloader,https://github.com/pytorch/pytorch/issues/108651,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-09-08T14:47:04Z +issue,89336,"Conv2d error on M1 mac, RuntimeError: NNPACK SpatialConvolution_updateOutput failed",https://github.com/pytorch/pytorch/issues/89336,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: convolution, module: macos",scan when time allows,2023-09-08T14:07:52Z +issue,108851,about nccl not work,https://github.com/pytorch/pytorch/issues/108851,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: nccl",scan when time allows,2023-09-08T13:39:53Z +issue,108858,Ubuntu vs Windows: torch.cuda.OutOfMemoryError only happens on Ubuntu,https://github.com/pytorch/pytorch/issues/108858,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2023-09-08T13:38:26Z +issue,70243,Unable to compile PyTorch when libcudart_static.so is not available,https://github.com/pytorch/pytorch/issues/70243,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2023-09-08T06:44:58Z +issue,108407,torch.einsum() computes different results on cpu and cuda on A100 GPU.,https://github.com/pytorch/pytorch/issues/108407,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2023-09-08T04:54:48Z +issue,108047,DDP training can not accept subnet address in IPV6,https://github.com/pytorch/pytorch/issues/108047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-09-08T03:04:35Z +issue,81065,[Releng] Improve the tutorials release process,https://github.com/pytorch/pytorch/issues/81065,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2023-09-07T21:49:05Z +issue,81883,RuntimeError: CUDA error: no kernel image is available for execution on the device,https://github.com/pytorch/pytorch/issues/81883,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-09-07T21:36:09Z +issue,92838,Enable OnDemand for Open Source CI,https://github.com/pytorch/pytorch/issues/92838,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-09-07T17:40:41Z +issue,99918,[RFC] DebugMode,https://github.com/pytorch/pytorch/issues/99918,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-09-07T17:40:39Z +issue,34660,"Add ""strict"" flag to ignore missing parameters in Optimizer.load_state_dict",https://github.com/pytorch/pytorch/issues/34660,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-09-07T14:17:22Z +issue,34611,Eigen version for PyTorch ?,https://github.com/pytorch/pytorch/issues/34611,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-09-07T03:21:17Z +issue,81140,libprotobuf version compatibility,https://github.com/pytorch/pytorch/issues/81140,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2023-09-07T02:52:16Z +issue,108496,The CPU version of `torch.cummax` is slow,https://github.com/pytorch/pytorch/issues/108496,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2023-09-06T18:34:57Z +issue,51778,Assertion on the existence of RRef user-facing methods' docstrings,https://github.com/pytorch/pytorch/issues/51778,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2023-09-06T13:04:22Z +issue,108231,DistributedSampler class: Change total_size into num_samples,https://github.com/pytorch/pytorch/issues/108231,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-09-05T19:00:44Z +issue,108569,Call for a deterministic implementation of scatter_add_cuda_kernel,https://github.com/pytorch/pytorch/issues/108569,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2023-09-05T15:17:18Z +issue,108494,backend-friendly distributions,https://github.com/pytorch/pytorch/issues/108494,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: distributions",scan when time allows,2023-09-04T21:56:58Z +issue,75440,[quant] PackedLinearWeight::apply_dynamic_impl does not handle ReLUFused template argument,https://github.com/pytorch/pytorch/issues/75440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-09-01T23:15:09Z +issue,108102,_sampled_addmm_kernel cause 'misaligned address' with new triton pin,https://github.com/pytorch/pytorch/issues/108102,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-09-01T21:54:11Z +issue,108244,Pytorch versions without the abi3 flag,https://github.com/pytorch/pytorch/issues/108244,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: python frontend",scan when time allows,2023-08-31T21:13:41Z +issue,106124,[feature request] Better argument checks and error messaging for `tensor.repeat`,https://github.com/pytorch/pytorch/issues/106124,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2023-08-31T20:48:25Z +issue,28754,ABI backwards compatibility,https://github.com/pytorch/pytorch/issues/28754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: abi,scan when time allows,2023-08-31T18:20:12Z +issue,44989,"Support unfold for integral types (long, byte etc) tensors",https://github.com/pytorch/pytorch/issues/44989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-08-30T15:02:39Z +issue,108197,AdaptiveMaxPool documentation is not detailed,https://github.com/pytorch/pytorch/issues/108197,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: pooling",scan when time allows,2023-08-30T14:36:36Z +issue,45009,Median / quantile / mode / rank / percentile pooling,https://github.com/pytorch/pytorch/issues/45009,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: pooling, module: sorting and selection",scan when time allows,2023-08-30T11:13:01Z +issue,44991,"Unite/unify tensor.unfold with F.unfold and make them more performant (zero-copy or little-copy with stride tricks, as in NumPy)",https://github.com/pytorch/pytorch/issues/44991,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: ux",scan when time allows,2023-08-30T11:10:05Z +issue,108190,[FSDP] incorrect backward prefetch order when using BackwardPrefetch.BACKWARD_POST,https://github.com/pytorch/pytorch/issues/108190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2023-08-29T23:07:52Z +issue,108046,Enhanced Available Backend Discovery and Selection in PyTorch 2,https://github.com/pytorch/pytorch/issues/108046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-08-29T18:29:13Z +issue,70919,`linspace` should support an `endpoint` parameter,https://github.com/pytorch/pytorch/issues/70919,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: tensor creation",scan when time allows,2023-08-29T16:25:57Z +issue,21956,nn.init.orthogonal_ doesn't work with multiprocessing,https://github.com/pytorch/pytorch/issues/21956,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dependency bug, module: initialization, module: multiprocessing, module: nn",scan when time allows,2023-08-29T14:24:16Z +issue,31942,Support sparse inputs for torch.block_diag,https://github.com/pytorch/pytorch/issues/31942,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sparse, module: tensor creation",scan when time allows,2023-08-29T10:38:07Z +issue,107575,halo，I continue pretrain llama2-13B model ，but save state_dict is about 50GB file,https://github.com/pytorch/pytorch/issues/107575,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-08-29T08:56:28Z +issue,107700,where.self_out doesn't fail gracefully when inputs have different dtypes,https://github.com/pytorch/pytorch/issues/107700,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: edge cases, module: type promotion",scan when time allows,2023-08-28T17:33:28Z +issue,107693,Inconsistencies when casting to integral types,https://github.com/pytorch/pytorch/issues/107693,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: int overflow, module: type promotion",scan when time allows,2023-08-28T17:30:12Z +issue,104193,Conversion from strided to batched sparse compressed tensor with a non-constant number of zeros in batches fails,https://github.com/pytorch/pytorch/issues/104193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-08-28T16:23:59Z +issue,108016,Failure in Initiating Pyotch DDP-style code ( Multi-machine multi-card environment),https://github.com/pytorch/pytorch/issues/108016,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-08-28T14:59:02Z +issue,106265,RuntimeError: Expected a proper Tensor but got None (or an undefined Tensor in C++) for argument #0 'grad_y',https://github.com/pytorch/pytorch/issues/106265,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: mps",scan when time allows,2023-08-28T14:38:01Z +issue,38090,ImportError: libtorch_cpu.so: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/38090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-08-27T19:22:00Z +issue,104761,"Torch Filename Storage hangs on ""file_system"" sharing strategy after in-place fill",https://github.com/pytorch/pytorch/issues/104761,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: multiprocessing",scan when time allows,2023-08-26T15:09:27Z +issue,107961,"aten.lift throws error in dynamo backends -> RuntimeError: !at::functionalization::impl::isFunctionalTensor(self) INTERNAL ASSERT FAILED at ""../aten/src/ATen/FunctionalizeFallbackKernel.cpp"":167",https://github.com/pytorch/pytorch/issues/107961,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2023-08-25T22:25:49Z +issue,107821,`torch.distributions.Pareto.sample` sometimes gives `inf`,https://github.com/pytorch/pytorch/issues/107821,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: nans and infs",scan when time allows,2023-08-25T19:43:27Z +issue,107864,`C10_HOST_DEVICE` for `std::isnan(c10::complex)`?,https://github.com/pytorch/pytorch/issues/107864,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-08-24T23:04:19Z +issue,107855,About the multi-node example not working properly,https://github.com/pytorch/pytorch/issues/107855,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-08-24T23:02:05Z +issue,107820,`add_image_with_boxes` method from `torch.utils.tensorboard.writer.SummaryWriter` is broken,https://github.com/pytorch/pytorch/issues/107820,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",scan when time allows,2023-08-24T22:59:37Z +issue,34646,"Support creating a CPU tensor from ctypes pointer in Python / from_blob(ptr, shape, strides, dtype)",https://github.com/pytorch/pytorch/issues/34646,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: memory usage, module: numpy",scan when time allows,2023-08-24T20:00:55Z +issue,107701,arange.out produces incorrect output when out tensor has dtype long,https://github.com/pytorch/pytorch/issues/107701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-08-24T17:47:09Z +issue,107678,[Torch.fx] Torch fx failed to trace torch extension library,https://github.com/pytorch/pytorch/issues/107678,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-08-24T01:44:42Z +issue,107443,Determinism by using datapipes shuffle,https://github.com/pytorch/pytorch/issues/107443,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: determinism",scan when time allows,2023-08-23T22:37:07Z +issue,107680,torch.nn.functional.cross_entropy different loss when providing one_hot_target and class weights,https://github.com/pytorch/pytorch/issues/107680,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2023-08-23T20:31:24Z +issue,31423,Add option in LSTM layer to access all cell states of all time steps,https://github.com/pytorch/pytorch/issues/31423,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2023-08-23T14:42:50Z +issue,45897,Unify matrix multiplications operations,https://github.com/pytorch/pytorch/issues/45897,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2023-08-23T10:26:36Z +issue,84510,[NVFuser] RuntimeError: ref_id_it != replayed_concrete_ids_.vector().end() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/84510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: nvfuser",scan when time allows,2023-08-23T07:19:16Z +issue,70701,AdaptiveAvgPool1d - RuntimeError: shmem_size <= sharedMemPerBlockINTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/70701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-08-23T03:48:18Z +issue,107528,Inconsistent results when running torch.arctanh,https://github.com/pytorch/pytorch/issues/107528,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-08-23T00:01:59Z +issue,107555,Integer multiplication overflow when running torch.nn.MaxUnpool3d,https://github.com/pytorch/pytorch/issues/107555,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-23T00:01:45Z +issue,107751,conv cudnn support integers,https://github.com/pytorch/pytorch/issues/107751,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2023-08-23T00:01:26Z +issue,107703,"Hardtanh docs are inaccurate/incomplete, since hardtanh behaves like clamp",https://github.com/pytorch/pytorch/issues/107703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-08-22T23:58:20Z +issue,107702,Inconsistencies when handling scalars that are out of the range relative to the input tensor's dtype,https://github.com/pytorch/pytorch/issues/107702,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:58:08Z +issue,107695,New variables in torch._ops.py pollute the torch.ops namespace,https://github.com/pytorch/pytorch/issues/107695,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: library,scan when time allows,2023-08-22T23:55:50Z +issue,107590,Select on a coalesced COO tensor returns COO tensor with coalesce flag set to False.,https://github.com/pytorch/pytorch/issues/107590,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-08-22T23:46:27Z +issue,107556,Integer multiplication overflow when running torch.nn.AdaptiveAvgPool2d,https://github.com/pytorch/pytorch/issues/107556,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:43:40Z +issue,107554,Integer multiplication overflow when running torch.diagflat,https://github.com/pytorch/pytorch/issues/107554,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:43:32Z +issue,107552,Storage size calculation overflowed when running torch.nn.functional.interpolate,https://github.com/pytorch/pytorch/issues/107552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:43:13Z +issue,107550,Integer multiplication overflow when running torch.eye,https://github.com/pytorch/pytorch/issues/107550,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:42:14Z +issue,107548,Integer calculation overflow when running torch.nn.functional.adaptive_avg_pool2d,https://github.com/pytorch/pytorch/issues/107548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:42:05Z +issue,107546,Integer overflow when running torch.nn.functional.upsample_bilinear,https://github.com/pytorch/pytorch/issues/107546,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:41:55Z +issue,107545,Integer overflow when running torch.nn.functional.upsample,https://github.com/pytorch/pytorch/issues/107545,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:41:45Z +issue,107544,Integer overflow when running torch.nn.ReplicationPad3d,https://github.com/pytorch/pytorch/issues/107544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:38:54Z +issue,107543,Integer overflow when running torch.nn.AdaptiveAvgPool2d,https://github.com/pytorch/pytorch/issues/107543,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:38:46Z +issue,107541,Integer overflow when running torch.nn.MaxUnpool2d,https://github.com/pytorch/pytorch/issues/107541,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-22T23:38:34Z +issue,107536,Appending new logs to existing tbevent files when using tensorboard,https://github.com/pytorch/pytorch/issues/107536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-08-22T23:36:49Z +issue,107494,Mismatch in type of error raised when reducing along empty slice between eager and primtorch,https://github.com/pytorch/pytorch/issues/107494,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2023-08-22T23:31:20Z +issue,63188,ROCm build documentation lacks needed dependencies,https://github.com/pytorch/pytorch/issues/63188,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs, module: rocm",scan when time allows,2023-08-22T21:29:17Z +issue,77576,`stateless.functional_call` doesn't work with `nn.DataParallel`,https://github.com/pytorch/pytorch/issues/77576,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nn",scan when time allows,2023-08-22T19:20:41Z +issue,51351,Run time error,https://github.com/pytorch/pytorch/issues/51351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: dependency bug, module: wsl",scan when time allows,2023-08-22T13:39:50Z +issue,51210,"Complex->Real cast is a warning, calling `real` or `imag` on non-complex tensors is an Error.",https://github.com/pytorch/pytorch/issues/51210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: ux",scan when time allows,2023-08-22T02:00:56Z +issue,45833,Support torch.mean for BoolTensors and other integer tensor inputs (without manual upcasting and hopefully without hidden upcasting),https://github.com/pytorch/pytorch/issues/45833,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: reductions, module: type promotion",scan when time allows,2023-08-21T22:11:07Z +issue,107412,"max_pool1d, max_pool2d, max_pool3d Integers for cpu and cuda",https://github.com/pytorch/pytorch/issues/107412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2023-08-21T21:43:53Z +issue,107514,conv2d wrong results on 3090/3090ti,https://github.com/pytorch/pytorch/issues/107514,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2023-08-21T21:21:56Z +issue,103530,Deduplicate the operands passed into torch.cond after dynamo tracing.,https://github.com/pytorch/pytorch/issues/103530,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-08-21T20:08:47Z +issue,51280,"[ux] Proposal to have t() === transpose(-1, -2), since batches are very frequent",https://github.com/pytorch/pytorch/issues/51280,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: ux",scan when time allows,2023-08-21T19:12:21Z +issue,107539,Integer overflow when running torch.nn.functional.max_unpool2d,https://github.com/pytorch/pytorch/issues/107539,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: int overflow,scan when time allows,2023-08-21T16:34:48Z +issue,84052,Index out of bounds Error with PerChannel Quantization,https://github.com/pytorch/pytorch/issues/84052,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-21T15:26:24Z +issue,40759,Quantization support for F.Softmax,https://github.com/pytorch/pytorch/issues/40759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-21T14:27:27Z +issue,69900,Port MarginRankingLoss to TensorIterator,https://github.com/pytorch/pytorch/issues/69900,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn, module: performance",scan when time allows,2023-08-21T13:28:15Z +issue,70238,`TensorIterator`: refactor `build_ternary_op` to match binary versions,https://github.com/pytorch/pytorch/issues/70238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensoriterator,scan when time allows,2023-08-21T13:28:12Z +issue,72134,Add support for `complex` `mean` to the `normal` operator,https://github.com/pytorch/pytorch/issues/72134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: distributions",scan when time allows,2023-08-21T13:28:03Z +issue,74952,[feature request] Make `index_select` parallel.,https://github.com/pytorch/pytorch/issues/74952,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: performance",scan when time allows,2023-08-21T13:27:52Z +issue,105105,Remaining functions without meta registrations,https://github.com/pytorch/pytorch/issues/105105,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2023-08-21T13:27:28Z +issue,106622,FFT Samples Inputs with More than Three Dimensions,https://github.com/pytorch/pytorch/issues/106622,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: tests",scan when time allows,2023-08-21T10:50:37Z +issue,106584,Lacking commutativity of `tensor.expand` and `tensor.flatten`,https://github.com/pytorch/pytorch/issues/106584,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2023-08-20T21:13:48Z +issue,92818,Segfault when running torch.atan2,https://github.com/pytorch/pytorch/issues/92818,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2023-08-20T13:03:33Z +issue,87794,CUDA error: operation not permitted when stream is capturing (2 GPUs),https://github.com/pytorch/pytorch/issues/87794,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multi-gpu,scan when time allows,2023-08-20T12:11:45Z +issue,35872,torch.autograd.set_detect_anomaly(True) does not exist in C++?,https://github.com/pytorch/pytorch/issues/35872,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-08-20T05:23:59Z +issue,107392,Overly strict type hints for `torch.utils.data.random_split`,https://github.com/pytorch/pytorch/issues/107392,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: typing",scan when time allows,2023-08-20T04:59:21Z +issue,107374,DataParallel scatter method split tensor wrong,https://github.com/pytorch/pytorch/issues/107374,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: data parallel",scan when time allows,2023-08-18T12:54:01Z +issue,28244,ProcessGroupMPI reports incorrect world size,https://github.com/pytorch/pytorch/issues/28244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-08-18T02:42:40Z +issue,107387,Incorrect type hint for `torch.library.Library.define`,https://github.com/pytorch/pytorch/issues/107387,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: library, module: typing",scan when time allows,2023-08-17T22:16:29Z +issue,107321,"using Union[str, Tensor] as an argument to a torch.jit.script function",https://github.com/pytorch/pytorch/issues/107321,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-08-17T15:39:43Z +issue,65528,I cannot use x.to(GPU) or x.cuda(GPU),https://github.com/pytorch/pytorch/issues/65528,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs",scan when time allows,2023-08-17T14:40:37Z +issue,106467,nn.CrossEntropyLoss with invalid target generates corrups memory eventualy leading to CUDA error: an illegal memory access,https://github.com/pytorch/pytorch/issues/106467,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2023-08-17T14:30:16Z +issue,104506,TImeout in NCCL doesn't work,https://github.com/pytorch/pytorch/issues/104506,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-08-17T08:40:32Z +issue,107253,Conda configuration shouldn't pollute $PATH variable,https://github.com/pytorch/pytorch/issues/107253,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, oncall: releng",scan when time allows,2023-08-17T03:09:18Z +issue,107217,Documenting `IterableDataset`'s needing `StopIteration` for finite data,https://github.com/pytorch/pytorch/issues/107217,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2023-08-17T02:50:31Z +issue,104147,PyTorch2.0 ROCM LayerNorm HIP error: invalid configuration,https://github.com/pytorch/pytorch/issues/104147,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2023-08-16T16:01:38Z +issue,104776,torch.nn.CrossEntropyLoss: class weighting changes label_smoothing,https://github.com/pytorch/pytorch/issues/104776,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2023-08-16T15:40:18Z +issue,107125,combining `vmap` with NN containing `MaxPool2d' leads to discrepancies in output,https://github.com/pytorch/pytorch/issues/107125,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-08-16T14:39:50Z +issue,50276,torch.meshgrid is divergent from np.meshgrid,https://github.com/pytorch/pytorch/issues/50276,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy",scan when time allows,2023-08-16T07:18:06Z +issue,68429,istft gradcheck fails on ROCm,https://github.com/pytorch/pytorch/issues/68429,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: fft, module: rocm",scan when time allows,2023-08-15T21:51:03Z +issue,107201,The difference between channels last backward and channels first backward of AvgPool2d on CUDA is too large,https://github.com/pytorch/pytorch/issues/107201,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory format, module: nn",scan when time allows,2023-08-15T21:18:53Z +issue,103482,Document CI retry rules,https://github.com/pytorch/pytorch/issues/103482,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2023-08-15T19:44:31Z +issue,100158,`torch.sparse_csc_tensor` matrix multiplication produces MKL error SPARSE_STATUS_ALLOC_FAILED when density is too high,https://github.com/pytorch/pytorch/issues/100158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-08-15T14:29:41Z +issue,107102,Error when using sparse_coo tensor with optimizer,https://github.com/pytorch/pytorch/issues/107102,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-08-14T20:04:54Z +issue,46904,Authentication for RPC,https://github.com/pytorch/pytorch/issues/46904,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe, oncall: distributed",scan when time allows,2023-08-12T03:27:41Z +issue,81667,[Tracker] AO migration of quantization from `torch.nn` to `torch.ao.nn`,https://github.com/pytorch/pytorch/issues/81667,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-11T18:49:42Z +issue,107011,max_pool3d_with_indices_backward_cuda does not have a deterministic implementation,https://github.com/pytorch/pytorch/issues/107011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: pooling",scan when time allows,2023-08-11T13:47:33Z +issue,26889,Statically checked tensor shapes,https://github.com/pytorch/pytorch/issues/26889,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2023-08-11T13:16:22Z +issue,77053,Incorrect documentation in ``gumble_softmax`` function.,https://github.com/pytorch/pytorch/issues/77053,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2023-08-11T09:19:30Z +issue,79709,ONEDNN testing is not done properly in quantization codebase,https://github.com/pytorch/pytorch/issues/79709,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-10T20:08:56Z +issue,79307,`prepare_qat_fx` docstring doesn't run,https://github.com/pytorch/pytorch/issues/79307,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: quantization",scan when time allows,2023-08-10T20:02:56Z +issue,81287,Runtime error in Libtorch cpp project (Didn't find engine for operation quantized::conv2d_prepack NoQEngine),https://github.com/pytorch/pytorch/issues/81287,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-10T20:02:24Z +issue,79739,quantization: misleading backend config for linear_dynamic_fp16,https://github.com/pytorch/pytorch/issues/79739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-10T20:02:02Z +issue,81381,quantization: QConfigMapping should be easy to print,https://github.com/pytorch/pytorch/issues/81381,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-10T19:59:12Z +issue,83672,quantization: unexpected casting of tensor min and max to int in histogram observer,https://github.com/pytorch/pytorch/issues/83672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-10T19:57:37Z +issue,84148,[Quant] Reference get_default_qconfig_mapping in docs/tutorials,https://github.com/pytorch/pytorch/issues/84148,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-08-10T19:51:15Z +issue,34772,"Unsupported ONNX op (Upsample 3D, bicubic) contrary to documentation",https://github.com/pytorch/pytorch/issues/34772,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: vision",scan when time allows,2023-08-10T18:08:16Z +issue,83773,[Nested Tensor] view + inplace for autograd.,https://github.com/pytorch/pytorch/issues/83773,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nestedtensor",scan when time allows,2023-08-10T16:22:00Z +issue,70239,`pytorch` hangs during interaction with `ray` package,https://github.com/pytorch/pytorch/issues/70239,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: deadlock",scan when time allows,2023-08-10T12:21:34Z +issue,18095,torch.flip is inconsistent with np.flip and also uses `dims` arg instead of `dim`,https://github.com/pytorch/pytorch/issues/18095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-08-10T09:05:04Z +issue,22331,Integer division by Zero giving large number results instead of NaN/inf on Windows,https://github.com/pytorch/pytorch/issues/22331,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2023-08-09T19:41:13Z +issue,35881,Error when loading jit traced FasterRCNN model in C++,https://github.com/pytorch/pytorch/issues/35881,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-08-09T19:38:21Z +issue,27815,Improve the error message when trying to install in a 32-bit Python environment,https://github.com/pytorch/pytorch/issues/27815,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-08-09T19:36:57Z +issue,67955,Torch 1.9.1.post3 - Error in magma_getdevice_arch: MAGMA not initialized (call magma_init() first) or bad device,https://github.com/pytorch/pytorch/issues/67955,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-08-09T18:47:02Z +issue,99297,Slicing and indexing support negative steps,https://github.com/pytorch/pytorch/issues/99297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2023-08-09T16:51:12Z +issue,106828,Use expect tests for error inputs,https://github.com/pytorch/pytorch/issues/106828,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2023-08-09T03:21:20Z +issue,95946,"Build from source, Undefined symbol: c10::detail::maybe_wrap_dim_slow(long long, long long, bool)",https://github.com/pytorch/pytorch/issues/95946,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",scan when time allows,2023-08-08T23:49:19Z +issue,101999,[FSDP] Ensure full precision checkpoints,https://github.com/pytorch/pytorch/issues/101999,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2023-08-08T22:04:52Z +issue,106549,Can't build PyTorch 1.13.1 with Vulkan support,https://github.com/pytorch/pytorch/issues/106549,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2023-08-08T21:27:13Z +issue,106237,ghstack + mergebot race condition,https://github.com/pytorch/pytorch/issues/106237,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2023-08-08T20:27:22Z +issue,106073,Potential lack of CI testing on older NVIDIA GPU,https://github.com/pytorch/pytorch/issues/106073,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2023-08-08T20:20:33Z +issue,106427,`torch.nn.utils.clip_grad_norm_()` causes H2D sync with foreach ops.,https://github.com/pytorch/pytorch/issues/106427,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: nn",scan when time allows,2023-08-08T14:47:48Z +issue,54209,[docs] Incomplete examples for torch.Generator,https://github.com/pytorch/pytorch/issues/54209,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-08-08T14:37:26Z +issue,41681,TypeError: cannot create 'generator' instances,https://github.com/pytorch/pytorch/issues/41681,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2023-08-08T11:09:55Z +issue,106713,Dev Container Support for PyTorch,https://github.com/pytorch/pytorch/issues/106713,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-08-07T18:50:36Z +issue,106718,Add AMD image to the .devcontainer spec,https://github.com/pytorch/pytorch/issues/106718,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-08-07T18:50:13Z +issue,106717,Provide .devcontainer PyTorch - MPS environment,https://github.com/pytorch/pytorch/issues/106717,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-08-07T18:48:05Z +issue,66573,tools/amd_build/build_amd.py should fail if any file fails to write,https://github.com/pytorch/pytorch/issues/66573,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: rocm",scan when time allows,2023-08-07T15:53:30Z +issue,106662,Will torch.sparse.mm support multiplying two boolean matrices?,https://github.com/pytorch/pytorch/issues/106662,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: sparse",scan when time allows,2023-08-06T23:20:05Z +issue,106579,"Boolean valued images loaded from disk, when converted to torch int/float tensor, the True valued pixels gets converted to 255 instead of 1",https://github.com/pytorch/pytorch/issues/106579,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor,scan when time allows,2023-08-05T18:32:43Z +issue,91936,Fails to build on ppc64le with clang,https://github.com/pytorch/pytorch/issues/91936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: vectorization",scan when time allows,2023-08-05T14:18:26Z +issue,106520,Distributed torch.linalg.eigh (and other functions) on cuda using cuSOLVERMG,https://github.com/pytorch/pytorch/issues/106520,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2023-08-05T08:28:44Z +issue,106604,Relu6 not able to process nan values,https://github.com/pytorch/pytorch/issues/106604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,scan when time allows,2023-08-05T05:53:17Z +issue,106606,More Performant CachingHostAllocator for Pinned Memory Allocation,https://github.com/pytorch/pytorch/issues/106606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-08-05T03:53:21Z +issue,106563,NotImplementedError: Could not run 'aten::multinomial' with arguments from the 'Meta' backend.,https://github.com/pytorch/pytorch/issues/106563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2023-08-05T03:47:01Z +issue,106540,Dataset with Queue issue,https://github.com/pytorch/pytorch/issues/106540,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2023-08-05T03:41:51Z +issue,106533,CUDA device support does not register allocator to c10::GetAllocator(...),https://github.com/pytorch/pytorch/issues/106533,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cuda",scan when time allows,2023-08-05T03:40:33Z +issue,106545,`softmax` to handle dimensions comprised of `-inf`,https://github.com/pytorch/pytorch/issues/106545,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-08-04T20:07:06Z +issue,55457,[feature request] TupleSequential module or an option for Sequential for unpacking tuples (i.e. support varargs),https://github.com/pytorch/pytorch/issues/55457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-08-04T11:00:59Z +issue,33298,[feature request] torch.expand to not require unsqueeze and match -1 to existing dimensions if tensor.shape.count(-1) == tensor.ndim,https://github.com/pytorch/pytorch/issues/33298,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-08-04T00:44:51Z +issue,42959,[feature request] Faster specialized int16->float32 conversions to match speed with NumPy,https://github.com/pytorch/pytorch/issues/42959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2023-08-03T23:13:38Z +issue,82316,Add more Vulkan operations,https://github.com/pytorch/pytorch/issues/82316,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2023-08-03T22:14:29Z +issue,106566,DTensor Sharding prop cache stats,https://github.com/pytorch/pytorch/issues/106566,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dtensor, oncall: distributed",scan when time allows,2023-08-03T19:07:35Z +issue,54216,Add inplace variant for torch.minimum / torch.maximum,https://github.com/pytorch/pytorch/issues/54216,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-08-03T18:26:08Z +issue,103352,"[feature request] Native method for iterating Python items of tensors: `iteritems()` and a new `tensor.item(i, j, k, ...)` method",https://github.com/pytorch/pytorch/issues/103352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-08-03T14:55:46Z +issue,106469,Extreme slowdown of torch.mm for certain sizes and strides with bfloat16,https://github.com/pytorch/pytorch/issues/106469,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda",scan when time allows,2023-08-03T02:36:49Z +issue,5740,[feature request] scalar input to scatter_add_ like scatter_,https://github.com/pytorch/pytorch/issues/5740,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2023-08-03T00:31:00Z +issue,84044,"hipErrorNoBinaryForGpu, but reversed",https://github.com/pytorch/pytorch/issues/84044,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2023-08-02T20:54:49Z +issue,106382,Installing torchvision for CPU leads to unwanted upgrade of torch + pip would not install nightly as considers that release is the latest (?),https://github.com/pytorch/pytorch/issues/106382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-08-02T17:07:15Z +issue,106450,Check for output_padding <= stride/dilation in ConvTranspose1d,https://github.com/pytorch/pytorch/issues/106450,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: padding",scan when time allows,2023-08-02T13:59:47Z +issue,106173,`torch.ops.aten.split.Tensor._schema` return alias annotations are wrong,https://github.com/pytorch/pytorch/issues/106173,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2023-08-02T13:46:20Z +issue,106375,nll_loss reference shouldn't be registered as a decomposition.,https://github.com/pytorch/pytorch/issues/106375,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: decompositions, module: primtorch",scan when time allows,2023-08-02T13:45:58Z +issue,51011,DataLoader is slow in spawned processes,https://github.com/pytorch/pytorch/issues/51011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing, module: performance",scan when time allows,2023-08-02T07:57:46Z +issue,22541,make[2]: *** No rule to make target 'libtorch/lib/libc10.so',https://github.com/pytorch/pytorch/issues/22541,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2023-08-02T06:37:36Z +issue,59545,max-sum operation,https://github.com/pytorch/pytorch/issues/59545,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: reductions",scan when time allows,2023-08-01T23:36:09Z +issue,98481,Bring CudaPluggableAllocator to feature parity with the Native Allocator,https://github.com/pytorch/pytorch/issues/98481,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: internals",scan when time allows,2023-08-01T16:02:20Z +issue,106339,"ReduceLROnPlateau increases learning rate exponentially, causing training to diverge",https://github.com/pytorch/pytorch/issues/106339,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2023-08-01T15:54:57Z +issue,106359,backwards compatibility about class _LRScheduler,https://github.com/pytorch/pytorch/issues/106359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2023-08-01T15:54:08Z +issue,106297,Can't build with non-static protobuf,https://github.com/pytorch/pytorch/issues/106297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-08-01T07:12:03Z +issue,90289,quantization qconfig: can we set per-channel quant as default for qnnpack?,https://github.com/pytorch/pytorch/issues/90289,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-07-31T22:13:33Z +issue,104623,I propose a new overview section in the documentation,https://github.com/pytorch/pytorch/issues/104623,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-07-31T15:20:04Z +issue,98406,Contribute to the privateuse1 backend.,https://github.com/pytorch/pytorch/issues/98406,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: internals",scan when time allows,2023-07-31T14:34:34Z +issue,40373,[discussion] Expressing tensor dimension semantics / constraints through typing / constraints blocks. Constraints block could be scripted/traced and help for tracing/script execution and codegen,https://github.com/pytorch/pytorch/issues/40373,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2023-07-30T20:33:47Z +issue,105213,[DTensor] Module parallelized with ColwiseParallel should return a sharded tensor,https://github.com/pytorch/pytorch/issues/105213,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dtensor,scan when time allows,2023-07-29T00:20:44Z +issue,106206,Build failure due to C++ version mismatch,https://github.com/pytorch/pytorch/issues/106206,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-07-28T21:45:58Z +issue,58833,Foreach Functions Tracking Issue,https://github.com/pytorch/pytorch/issues/58833,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: performance",scan when time allows,2023-07-27T21:38:22Z +issue,105904,Libtorch linking Error:undefined reference,https://github.com/pytorch/pytorch/issues/105904,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-07-27T14:37:39Z +issue,105726,There is a big precision error between A100 and 3090 when using torch.matmul with fp16 precision,https://github.com/pytorch/pytorch/issues/105726,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half",scan when time allows,2023-07-27T07:57:27Z +issue,52256,[docs] Bad docs SEO,https://github.com/pytorch/pytorch/issues/52256,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2023-07-27T05:54:08Z +issue,105799,inconsistent signature for dataloader in docs/source/data.rst,https://github.com/pytorch/pytorch/issues/105799,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-07-26T23:33:31Z +issue,104479,FSDP Optimizer Overlap - follow ups,https://github.com/pytorch/pytorch/issues/104479,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2023-07-26T12:33:13Z +issue,104872,errors in CONTRIBUTING.md,https://github.com/pytorch/pytorch/issues/104872,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-07-26T04:43:38Z +issue,105916,Missing coalesced flag from `torch.autograd.Function.backward`,https://github.com/pytorch/pytorch/issues/105916,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-07-25T20:20:36Z +issue,99176,Add Debug builds for python with pydebug,https://github.com/pytorch/pytorch/issues/99176,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2023-07-25T20:20:16Z +issue,35641,"Dimension reducing variants of bitwise operations (bitwise_or, bitwise_and, bitwise_xor)",https://github.com/pytorch/pytorch/issues/35641,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2023-07-25T20:18:51Z +issue,105954,Ensure PRs are rebased on top of a recent commit (CI check),https://github.com/pytorch/pytorch/issues/105954,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: devx,scan when time allows,2023-07-25T20:16:03Z +issue,105846,Strange backward behavior with sparse tensors,https://github.com/pytorch/pytorch/issues/105846,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-07-25T11:01:34Z +issue,105860,Programmation error enabling unlegal memory access on gpu,https://github.com/pytorch/pytorch/issues/105860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: mps",scan when time allows,2023-07-24T21:30:27Z +issue,98947,Add `torch.cat` support for torch native sparse tensors. (Need for PyG),https://github.com/pytorch/pytorch/issues/98947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-07-24T20:13:02Z +issue,105682,`torch.autocast(bfloat16)` runs bwd matmuls in fp16,https://github.com/pytorch/pytorch/issues/105682,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-07-24T16:56:01Z +issue,104389,distributed hooks want to support custom device,https://github.com/pytorch/pytorch/issues/104389,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-07-22T04:11:52Z +issue,79299,PyTorch gets stuck when using an NVLink/A6000 and more than two GPUs,https://github.com/pytorch/pytorch/issues/79299,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2023-07-22T00:48:21Z +issue,105697,Support symmetry in einsum,https://github.com/pytorch/pytorch/issues/105697,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2023-07-21T08:53:43Z +issue,101370,SparseAdam: working with dense parameters but sparse gradients - usecase,https://github.com/pytorch/pytorch/issues/101370,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-07-20T21:53:38Z +issue,105629,F.pad will accept 0 and negative values as parameter,https://github.com/pytorch/pytorch/issues/105629,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn, module: padding",scan when time allows,2023-07-20T15:14:42Z +issue,105548,torch.sparse.sampled_addmm doesn't compute gradients for 3D tensors,https://github.com/pytorch/pytorch/issues/105548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-07-20T14:49:37Z +issue,61222,The syncbatchnorm can not be employed in the MODEL that the gradients including in the loss function,https://github.com/pytorch/pytorch/issues/61222,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: derivatives, module: norms and normalization, oncall: distributed",scan when time allows,2023-07-20T14:30:34Z +issue,105457,Top level Glossary for users (not contributers),https://github.com/pytorch/pytorch/issues/105457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-07-20T13:16:30Z +issue,105529,Efficient BMM for sparse-dense tensors,https://github.com/pytorch/pytorch/issues/105529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-07-20T12:17:31Z +issue,51199,[numpy] torch.ceil and torch.floor don't support integer inputs while Numpy does,https://github.com/pytorch/pytorch/issues/51199,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-07-20T07:27:19Z +issue,90760,Large slow down by not calling `torch.set_num_threads`,https://github.com/pytorch/pytorch/issues/90760,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: linear algebra, module: performance",scan when time allows,2023-07-20T03:49:57Z +issue,103829,RPC Framework support custom backend,https://github.com/pytorch/pytorch/issues/103829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-07-20T02:54:16Z +issue,103313,Fast kernels for low rank matrix multiplication,https://github.com/pytorch/pytorch/issues/103313,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2023-07-19T18:14:15Z +issue,105329,Softmax doesn't support sparse tensors with the CSR layout,https://github.com/pytorch/pytorch/issues/105329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-07-19T11:12:32Z +issue,91989,Torch's affinity setting lead to openvino using only one core.,https://github.com/pytorch/pytorch/issues/91989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: intel,scan when time allows,2023-07-19T03:42:58Z +issue,105445,Silent Error of torch.fx.symbolic_trace when forward hooks are registered,https://github.com/pytorch/pytorch/issues/105445,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-07-18T21:27:52Z +issue,105499,[FSDP] Revisit mixed-precision casting logic,https://github.com/pytorch/pytorch/issues/105499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2023-07-18T20:52:18Z +issue,105220,Significant time difference of calculating Jacobian matrix using jacrev and oracle functions,https://github.com/pytorch/pytorch/issues/105220,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2023-07-18T19:54:19Z +issue,105304,"Backward pass with sparse parameters results in error ""Sparse division requires a scalar or zero-dim dense tensor divisor""",https://github.com/pytorch/pytorch/issues/105304,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: optimizer, module: sparse",scan when time allows,2023-07-18T15:38:14Z +issue,104510,"Using the latest version of Torch, when the code executes tcpstore, there is no response",https://github.com/pytorch/pytorch/issues/104510,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-07-18T12:52:47Z +issue,104867,rfftn and irfftn operations in pt2 return different results compared to v1.12.1,https://github.com/pytorch/pytorch/issues/104867,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: fft, module: third_party",scan when time allows,2023-07-18T09:02:07Z +issue,89490,build: cmake: ability to disable -Werror* (-Werror considered harmful),https://github.com/pytorch/pytorch/issues/89490,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-07-18T04:54:32Z +issue,25104,[FR] Dropout modules/functions should take in generator=,https://github.com/pytorch/pytorch/issues/25104,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: random",scan when time allows,2023-07-17T22:01:00Z +issue,32305,[FR] histc should (optionally) return a long tensor,https://github.com/pytorch/pytorch/issues/32305,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2023-07-17T21:26:37Z +issue,105062,[feature request] make the input k in rot90 a list of int to rotate tensors individually in a batch,https://github.com/pytorch/pytorch/issues/105062,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-07-17T21:06:31Z +issue,105326,Can't vmap over torch.tensor constructor,https://github.com/pytorch/pytorch/issues/105326,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-07-17T15:47:54Z +issue,103422,Possible memory leak when using Torch and Torchvision in conjunction with XGBoost,https://github.com/pytorch/pytorch/issues/103422,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: vision",scan when time allows,2023-07-17T14:53:48Z +issue,102085,BatchNorm can't be symbolically traced with torch.fx as a standalone module,https://github.com/pytorch/pytorch/issues/102085,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-07-17T08:17:53Z +issue,31460,DataParallel doesn't properly handle kwargs,https://github.com/pytorch/pytorch/issues/31460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nn",scan when time allows,2023-07-17T01:51:08Z +issue,30711,Sorting in embedding_dense_backward_cuda takes very long time,https://github.com/pytorch/pytorch/issues/30711,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: embedding, module: performance",scan when time allows,2023-07-15T21:19:21Z +issue,104845,Implement `diag` method for sparse COO tensors,https://github.com/pytorch/pytorch/issues/104845,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-07-15T10:43:42Z +issue,104711,New Loss Function Add In Pytorch,https://github.com/pytorch/pytorch/issues/104711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2023-07-15T00:28:03Z +issue,104925,Unnecessary record_stream call for backend:cudaMallocAsync,https://github.com/pytorch/pytorch/issues/104925,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: logging",scan when time allows,2023-07-14T21:25:00Z +issue,49016,Discrete Cosine Transform,https://github.com/pytorch/pytorch/issues/49016,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-07-14T16:50:17Z +issue,104962,Add a diagram showing the code structure to CONTRIBUTING.md,https://github.com/pytorch/pytorch/issues/104962,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-07-14T15:24:55Z +issue,105060,extra information messages for mac in setup.py would help.,https://github.com/pytorch/pytorch/issues/105060,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2023-07-14T03:16:36Z +issue,104466,`torch.view_as_real(tensor)` should return `nn.identity(tensor)` if its not complex instead of raising an error,https://github.com/pytorch/pytorch/issues/104466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2023-07-11T17:59:21Z +issue,102740,parameterizations.orthogonal does not work as intended with nn.GRU or nn.LSTM,https://github.com/pytorch/pytorch/issues/102740,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn.utils.parametrize, module: rnn",scan when time allows,2023-07-11T12:57:11Z +issue,104823,Unrelated error messages with torch.nn.AdaptiveAvgPool3d,https://github.com/pytorch/pytorch/issues/104823,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2023-07-11T08:16:51Z +issue,104935,torch version compare,https://github.com/pytorch/pytorch/issues/104935,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-07-11T04:17:38Z +issue,104789,[BE] Evaluate and improve eager for-loop optimizer memory perf,https://github.com/pytorch/pytorch/issues/104789,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-07-11T01:18:08Z +issue,104602,"F.adaptive_avg_pool3d(input, 1) returns infinity in half precision",https://github.com/pytorch/pytorch/issues/104602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: nn, module: numerical-stability",scan when time allows,2023-07-10T23:35:43Z +issue,104154,Numbers bigger than the range should be inf while the implementation just keeps the original.,https://github.com/pytorch/pytorch/issues/104154,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nans and infs",scan when time allows,2023-07-10T23:35:33Z +issue,104875,torch/testing/_comparison.py: If you are a user and see this message during normal operation please file an issue,https://github.com/pytorch/pytorch/issues/104875,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2023-07-10T23:34:53Z +issue,104853,torch.norm inconsistency?,https://github.com/pytorch/pytorch/issues/104853,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization, module: numerical-stability",scan when time allows,2023-07-10T23:26:33Z +issue,52262,tolist called on torch scalar does not return a list => proposal to support new arg force = True to provide a work around,https://github.com/pytorch/pytorch/issues/52262,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy",scan when time allows,2023-07-10T21:37:02Z +issue,1362,Feature Request: noise contrastive estimation/negative sampling,https://github.com/pytorch/pytorch/issues/1362,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-07-10T17:01:21Z +issue,103879,Can ``torch.vmap`` add ``grad_fn``= SelectBackward when maping over some dimension of the inputs?,https://github.com/pytorch/pytorch/issues/103879,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-07-10T13:40:39Z +issue,87449,pytorch could not build from source with cudnn 8.0.5,https://github.com/pytorch/pytorch/issues/87449,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: cudnn",scan when time allows,2023-07-10T02:15:41Z +issue,96085,Add support for `__collate__` attrib on dataset elements in `default_collate`,https://github.com/pytorch/pytorch/issues/96085,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-07-09T10:32:37Z +issue,104808,PyTorch built with CuDNN-8.8.1 crashes if CuDNN-8.9.2 is installed on the system,https://github.com/pytorch/pytorch/issues/104808,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cudnn",scan when time allows,2023-07-07T22:24:16Z +issue,104788,Use `isinstance` instead of `type` when checking for `torch.nn.Parameter`,https://github.com/pytorch/pytorch/issues/104788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-07-07T20:32:09Z +issue,90394,FX graph mode quant: backendconfig configuration missing for torch.nn.GRU,https://github.com/pytorch/pytorch/issues/90394,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-07-07T16:22:14Z +issue,104768,Support for `eval` in functional_call,https://github.com/pytorch/pytorch/issues/104768,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functional ux, module: nn",scan when time allows,2023-07-07T16:07:50Z +issue,100051,Issue of HistogramObserver to handle abnormal value,https://github.com/pytorch/pytorch/issues/100051,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2023-07-07T16:02:34Z +issue,94586,`permute` for named tensors,https://github.com/pytorch/pytorch/issues/94586,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2023-07-07T12:30:07Z +issue,104704,generate_vmap_rule=True sometimes gives batched grad_output,https://github.com/pytorch/pytorch/issues/104704,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-07-06T22:30:35Z +issue,104620,`torch.distributed.rpc.backend_registry.register_backend` fails to update `BackendType` enum,https://github.com/pytorch/pytorch/issues/104620,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2023-07-06T20:25:53Z +issue,104698,Incorrect Error Message Ordering for nn.AdaptiveAvgPool2d with Incorrect output_size,https://github.com/pytorch/pytorch/issues/104698,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-07-06T18:06:52Z +issue,103258,"Warn / deprecate / remove ProcessGroupNCCL._group_start(), _group_end() APIs",https://github.com/pytorch/pytorch/issues/103258,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-07-05T20:42:26Z +issue,104593,Bug in Conv/BN fuser with torch.fx,https://github.com/pytorch/pytorch/issues/104593,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: fx",scan when time allows,2023-07-05T17:27:15Z +issue,94804,Build Error: no matching function for call to ‘dnnl::graph::stream::stream()’,https://github.com/pytorch/pytorch/issues/94804,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn",scan when time allows,2023-07-05T13:02:48Z +issue,94869,aten::cudnn_convolution chooses different conv implementation given the same inputs.,https://github.com/pytorch/pytorch/issues/94869,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn",scan when time allows,2023-07-04T20:30:34Z +issue,103397,LayerNorm freeze processes using torch multiprocessing,https://github.com/pytorch/pytorch/issues/103397,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2023-07-04T20:08:52Z +issue,104315,Torch randperm with device mps does not sample exactly uniformly from all possible permutations,https://github.com/pytorch/pytorch/issues/104315,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: random",scan when time allows,2023-07-04T15:41:04Z +issue,72737,Pattern Matching with Tensors,https://github.com/pytorch/pytorch/issues/72737,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-07-04T05:34:04Z +issue,102269,Import of torch breaks standard multiprocessing,https://github.com/pytorch/pytorch/issues/102269,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dependency bug, module: multiprocessing, module: openmp",scan when time allows,2023-07-03T08:18:25Z +issue,102772,Support In-place Triangular Matrix Multiplication,https://github.com/pytorch/pytorch/issues/102772,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2023-07-02T20:16:58Z +issue,79337,Conda install from pytorch-nightly channel does not install the expected version on macOS,https://github.com/pytorch/pytorch/issues/79337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: macos",scan when time allows,2023-07-01T09:37:40Z +issue,104472,Investigate numerical stability of forward-mode AD of some foreach functions,https://github.com/pytorch/pytorch/issues/104472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2023-06-30T19:56:14Z +issue,86694,Feature request: Deterministic test input generation,https://github.com/pytorch/pytorch/issues/86694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2023-06-30T17:14:22Z +issue,104107,Tensor to_sparse fails on large matrices,https://github.com/pytorch/pytorch/issues/104107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2023-06-30T14:15:57Z +issue,104322,Illegal Memory Access on H100 `TestSparseCompressedTritonKernelsCUDA.test_triton_sampled_addmm_block_size_16_cuda_bfloat16`,https://github.com/pytorch/pytorch/issues/104322,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2023-06-29T17:23:17Z +issue,104163,Nested Tensor with PyG dataset custom class,https://github.com/pytorch/pytorch/issues/104163,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-06-29T08:05:08Z +issue,58466,`set_per_process_memory_fraction()` does not ensure max used GPU memory below fraction,https://github.com/pytorch/pytorch/issues/58466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-06-28T21:15:01Z +issue,104282,Rename `topic: not user facing` to `release notes: not user facing`,https://github.com/pytorch/pytorch/issues/104282,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-06-28T16:50:05Z +issue,41839,"[Feature Request] Add `to`, `cpu`, and `cuda` method to optimizer",https://github.com/pytorch/pytorch/issues/41839,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-06-28T05:54:13Z +issue,104252,[RFC] Make `_HYBRID_SHARD_ZERO2` public as `HYBRID_SHARD_GRAD_OP`,https://github.com/pytorch/pytorch/issues/104252,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fsdp,scan when time allows,2023-06-27T18:11:14Z +issue,103589,`interpolate` with `antialias=True` on CUDA doesn't work if the difference of spatial size is large,https://github.com/pytorch/pytorch/issues/103589,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: interpolation",scan when time allows,2023-06-27T08:43:54Z +issue,3625,Proposal: combine requires_grad and retain_grad(),https://github.com/pytorch/pytorch/issues/3625,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-06-26T22:16:06Z +issue,103960,CODEOWNERS file has errors due to non existent people being referred to,https://github.com/pytorch/pytorch/issues/103960,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-06-26T18:24:55Z +issue,98495,Strided to batch BSR/BSC conversion fails when the number of zeros per block varies while the number of blocks per patch is constant,https://github.com/pytorch/pytorch/issues/98495,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-06-26T16:28:04Z +issue,103959,"Need the full ""Release Compatibility Matrix"" of torch",https://github.com/pytorch/pytorch/issues/103959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: releng",scan when time allows,2023-06-25T01:37:54Z +issue,99797,Automatic broadcasting for sparse csr tensors,https://github.com/pytorch/pytorch/issues/99797,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-06-23T21:35:02Z +issue,103820,Upgrading SpGEMM algorithm to resolve Cusparse SpGEMM insufficient resources problem,https://github.com/pytorch/pytorch/issues/103820,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2023-06-23T20:42:27Z +issue,101069,"torch.autograd.detect_anomaly should report the original forward trace as part of the error, rather than as out of band warning",https://github.com/pytorch/pytorch/issues/101069,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-06-23T16:16:56Z +issue,97827,Memory leak when saving an input tensor returned as-is if mark_dirty and running with dual tensors,https://github.com/pytorch/pytorch/issues/97827,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases, module: forward ad",scan when time allows,2023-06-23T16:16:00Z +issue,91996,autograd.functional.jacobian : Imaginary part is lost for functions with real input and complex output.,https://github.com/pytorch/pytorch/issues/91996,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-06-23T15:42:30Z +issue,94990,interactions between views + autograd.Function + AOTAutograd causes memory leak,https://github.com/pytorch/pytorch/issues/94990,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-06-23T15:29:58Z +issue,94111,`svd` triggers INTERNAL ASSERT FAILED when computing jacobian in forward mode,https://github.com/pytorch/pytorch/issues/94111,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex, module: forward ad, module: linear algebra",scan when time allows,2023-06-23T15:28:56Z +issue,96766,Follow-ups to do after adding nested checkpoint,https://github.com/pytorch/pytorch/issues/96766,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-06-23T15:21:58Z +issue,95132,Differentiate with regard a subset of the input,https://github.com/pytorch/pytorch/issues/95132,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-06-23T15:21:19Z +issue,97575,Multi-output derivative formulas can save unnecessary tensors,https://github.com/pytorch/pytorch/issues/97575,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nestedtensor",scan when time allows,2023-06-23T15:14:21Z +issue,103802,Eager PTDQ Performs Worse Than Non-Quantized Linear Layer on CPU(in Terms of Speed),https://github.com/pytorch/pytorch/issues/103802,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-06-23T09:19:28Z +issue,103803,Support `Sequence` type in JIT,https://github.com/pytorch/pytorch/issues/103803,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: typing, oncall: jit",scan when time allows,2023-06-23T03:25:16Z +issue,60401,Need workaround to support multiprocess CUDA tensor sharing on Jetson Platforms,https://github.com/pytorch/pytorch/issues/60401,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2023-06-22T20:45:04Z +issue,104029,SummaryWriter.add_embedding not working for RGBA images,https://github.com/pytorch/pytorch/issues/104029,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-06-22T20:42:50Z +issue,103895,Issue with loading similar checkpoints in a distributed fashion,https://github.com/pytorch/pytorch/issues/103895,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-06-22T20:31:36Z +issue,104033,Reproducibility documentation should be updated,https://github.com/pytorch/pytorch/issues/104033,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numerical-reproducibility",scan when time allows,2023-06-22T17:23:55Z +issue,46739,torch.utils.tensorboard.SummaryWriter.add_embedding fails for some label_img sizes,https://github.com/pytorch/pytorch/issues/46739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-06-22T15:17:06Z +issue,104040,Improved error checking for custom Function when saving intermediates,https://github.com/pytorch/pytorch/issues/104040,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: double backwards",scan when time allows,2023-06-22T13:35:41Z +issue,14489,"Batch matmul with sparse matrix, dense vector",https://github.com/pytorch/pytorch/issues/14489,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-06-22T09:59:55Z +issue,41128,Memory bug for backward on torch.sparse.mm?,https://github.com/pytorch/pytorch/issues/41128,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-06-22T09:51:09Z +issue,101934,Parameter gradient is not moved parameter is moved across devices,https://github.com/pytorch/pytorch/issues/101934,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: docs, module: nn",scan when time allows,2023-06-22T01:52:48Z +issue,103860,"Long PR description leads to ""Argument list too long"" error from docker",https://github.com/pytorch/pytorch/issues/103860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: devx",scan when time allows,2023-06-21T23:24:06Z +issue,103370,torchgen/gen_backend_stubs.py compatibility with DispatchStubs,https://github.com/pytorch/pytorch/issues/103370,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: dispatch, module: structured kernels",scan when time allows,2023-06-21T20:16:08Z +issue,103868,"row.device().is_cpu() INTERNAL ASSERT FAILED at ""csrc/cpu/diag_cpu.cpp"":7",https://github.com/pytorch/pytorch/issues/103868,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: macos,scan when time allows,2023-06-21T20:08:02Z +issue,103847,Some parameters are missing type descriptions,https://github.com/pytorch/pytorch/issues/103847,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-06-21T16:37:58Z +issue,103846,"The document style is inconsistent with other documents, and the parameter type is not clearly highlight",https://github.com/pytorch/pytorch/issues/103846,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-06-21T16:37:28Z +issue,103837,type conflict,https://github.com/pytorch/pytorch/issues/103837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-06-21T16:31:44Z +issue,103832,Torch 1.13 for GPU breaks if libcublas is already present.,https://github.com/pytorch/pytorch/issues/103832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2023-06-21T16:07:45Z +issue,103798,Type misalignments in `nn.functional` (also JIT related),https://github.com/pytorch/pytorch/issues/103798,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: typing, oncall: jit",scan when time allows,2023-06-21T16:00:56Z +issue,103787,Generate complete annotations for `torch._C._nn`,https://github.com/pytorch/pytorch/issues/103787,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2023-06-21T15:58:39Z +issue,54389,"torch.lerp and torch._foreach_lerp should support uint8 inputs (for vision), int16 (for audio), int32/int64 (for generality) without (up)casting to float for and/or dtype argument",https://github.com/pytorch/pytorch/issues/54389,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2023-06-21T08:16:18Z +issue,78487,torch.fx: symbolic_trace: ones() received an invalid combination of arguments,https://github.com/pytorch/pytorch/issues/78487,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-06-21T06:54:03Z +issue,103643,"""addmm_out_sparse_csr_impl_mkl"" not implemented for 'Byte'",https://github.com/pytorch/pytorch/issues/103643,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-06-21T00:59:19Z +issue,103640,Disclose C++ ATen ops type promotion rules under OpOverload in Python,https://github.com/pytorch/pytorch/issues/103640,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-06-20T21:45:34Z +issue,55366,bool_tensor.sum(dtype=torch.int32) creates int32-copy of the original int8 tensor,https://github.com/pytorch/pytorch/issues/55366,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: reductions",scan when time allows,2023-06-20T21:29:27Z +issue,103161,Calling jacrev with LSTM and functional_call gives error,https://github.com/pytorch/pytorch/issues/103161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-06-20T20:18:59Z +issue,78210,Performance with MPS on AMD GPUs are worse than CPU,https://github.com/pytorch/pytorch/issues/78210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mps, module: performance",scan when time allows,2023-06-20T15:47:34Z +issue,53191,RuntimeError: cuDNN error: CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/53191,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2023-06-20T06:01:01Z +issue,86076,Figure out the future of Metal backend given the existence of MPS,https://github.com/pytorch/pytorch/issues/86076,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: mps",scan when time allows,2023-06-18T12:26:36Z +issue,103761,Merge type stubs for `torch.nn.functional`,https://github.com/pytorch/pytorch/issues/103761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2023-06-17T16:31:44Z +issue,74034,[torch.onnx] ONNX export failed on adaptive_avg_pool2d because input size not accessible not supported,https://github.com/pytorch/pytorch/issues/74034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2023-06-17T13:27:37Z +issue,72283,Add Sparsemax function to libtorch,https://github.com/pytorch/pytorch/issues/72283,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-06-16T10:24:07Z +issue,103608,Installing Torch on AMD Platform Leads to Huge Docker Image,https://github.com/pytorch/pytorch/issues/103608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2023-06-15T17:34:52Z +issue,103671,"""Y.getIntrusivePtr()->set_storage(X.getIntrusivePtr()->storage()); "" in C++ is not supported",https://github.com/pytorch/pytorch/issues/103671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-06-15T17:30:21Z +issue,103518,`gradcheck` produces false positives with sparse inputs when `masked=False`.,https://github.com/pytorch/pytorch/issues/103518,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-06-14T21:40:24Z +issue,103241,Image Processing with Pytorch,https://github.com/pytorch/pytorch/issues/103241,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-06-14T16:50:31Z +issue,101369,Theme update,https://github.com/pytorch/pytorch/issues/101369,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2023-06-14T16:34:21Z +issue,103498,ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 6047) of binary: /home/win10-ubuntu/anaconda3/envs/vicuna-7b/bin/python,https://github.com/pytorch/pytorch/issues/103498,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-06-14T03:44:01Z +issue,103372,ImportError: cannot import name 'Store' from 'torch.distributed',https://github.com/pytorch/pytorch/issues/103372,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-06-14T03:43:13Z +issue,102953,TypeError: (): incompatible function arguments,https://github.com/pytorch/pytorch/issues/102953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-06-14T03:35:20Z +issue,103382,NotImplementedError Could not run 'c10d::alltoall_' with arguments from the 'Meta' backend.,https://github.com/pytorch/pytorch/issues/103382,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-06-13T17:10:35Z +issue,20686,In-source build causes repeating filename annotations (Windows doesn't support out-of-source build),https://github.com/pytorch/pytorch/issues/20686,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: windows",scan when time allows,2023-06-13T13:26:38Z +issue,102423,Add default device function interface for device-aware apis,https://github.com/pytorch/pytorch/issues/102423,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2023-06-13T11:17:45Z +issue,100054,Add compile option -Werror=return-type compile error,https://github.com/pytorch/pytorch/issues/100054,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-06-13T08:49:22Z +issue,101110,Tensorboard graph tracing with torch fx API,https://github.com/pytorch/pytorch/issues/101110,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: tensorboard",scan when time allows,2023-06-13T07:20:37Z +issue,103393,Typing missing on arithmetic ops on `Tensor`,https://github.com/pytorch/pytorch/issues/103393,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2023-06-12T21:02:23Z +issue,103375,Inplace binary ops on tensor subclasses can cause mypy error,https://github.com/pytorch/pytorch/issues/103375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2023-06-12T20:54:52Z +issue,99806,`cat` gradgrad tests failing,https://github.com/pytorch/pytorch/issues/99806,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",scan when time allows,2023-06-12T17:48:43Z +issue,30386,Exponentiated gradient descent,https://github.com/pytorch/pytorch/issues/30386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-06-12T13:37:17Z +issue,102437,discuss.pytorch.org signup issue,https://github.com/pytorch/pytorch/issues/102437,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-06-11T13:20:30Z +issue,99035,Issue on building from source: Remove -mfpu=neon option on MacOS with Apple silicon,https://github.com/pytorch/pytorch/issues/99035,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-06-11T07:43:51Z +issue,103272,oneDNN kernel fails to compile,https://github.com/pytorch/pytorch/issues/103272,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mkldnn,scan when time allows,2023-06-10T04:05:35Z +issue,103316,binary_cross_entropy (loss) seems to be giving incorrect values for very negative logits,https://github.com/pytorch/pytorch/issues/103316,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-06-09T23:48:17Z +issue,103212,PyTorch can not be compiled with MKLDNN if system compiler is clang,https://github.com/pytorch/pytorch/issues/103212,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn",scan when time allows,2023-06-09T23:36:26Z +issue,102977,raise `RuntimeError` faster when loading an object with a torch CUDA tensor on a CPU-only machine,https://github.com/pytorch/pytorch/issues/102977,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization",scan when time allows,2023-06-09T19:22:14Z +issue,102971,Discussion and Design for Masked Loss Functions which can be used with PackedSequence training (but not exclusively),https://github.com/pytorch/pytorch/issues/102971,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: masked operators",scan when time allows,2023-06-09T16:48:09Z +issue,103060,Symbolic trace error about torch.nn.functional.pad,https://github.com/pytorch/pytorch/issues/103060,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-06-09T16:01:07Z +issue,103306,DTensor uneven sharding corner cases.,https://github.com/pytorch/pytorch/issues/103306,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-06-09T11:26:35Z +issue,102162,NotImplementedError in backprop on on dense-sparse matrices,https://github.com/pytorch/pytorch/issues/102162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-06-09T11:07:05Z +issue,103160,Allow overriding __repr__ to call dataclass_repr (infinite recursion right now),https://github.com/pytorch/pytorch/issues/103160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2023-06-08T15:14:48Z +issue,103131,"error: ‘aligned_alloc’ was not declared in this scope static_cast(aligned_alloc(FLATBUFFERS_MAX_ALIGNMENT, size)), free);",https://github.com/pytorch/pytorch/issues/103131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-06-08T15:01:04Z +issue,72948,Feature: a consistent Python and C++ logging facility that handles different classes of warnings,https://github.com/pytorch/pytorch/issues/72948,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2023-06-08T08:12:58Z +issue,35529,Support bool input tensors for argmax / argmin / sort / topk and other functions,https://github.com/pytorch/pytorch/issues/35529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2023-06-07T00:01:58Z +issue,16797,"Should be a way to unpickle an object with a torch cuda tensor on a CPU-only machine when using plain ""pickle""",https://github.com/pytorch/pytorch/issues/16797,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-06-06T14:33:47Z +issue,71117,Rollup: forward-mode AD operator coverage,https://github.com/pytorch/pytorch/issues/71117,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad",scan when time allows,2023-06-06T00:26:29Z +issue,102741,Unlocking PyTorch's Power: README.md in Multiple Languages!,https://github.com/pytorch/pytorch/issues/102741,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-06-05T21:54:15Z +issue,7313,[proposal] [discussion] Refactor pruning/weight_norm using new Reparametrization functionality + actually deprecate old impl of SpectralNorm,https://github.com/pytorch/pytorch/issues/7313,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-06-05T19:55:20Z +issue,102921,Unbox expectedml,https://github.com/pytorch/pytorch/issues/102921,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-06-05T16:20:20Z +issue,91395,PyObject preservation and resurrection for `StorageImpl`,https://github.com/pytorch/pytorch/issues/91395,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2023-06-05T13:40:31Z +issue,102830,Unknow error when using `make_graphed_callables`,https://github.com/pytorch/pytorch/issues/102830,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2023-06-05T13:16:44Z +issue,74940,Feature Request: Add J0 J1 J2 H0 H1 H2 Bessel functions,https://github.com/pytorch/pytorch/issues/74940,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2023-06-03T08:01:17Z +issue,101291,fused torch.optim.AdamW isn't faster than the unfused version,https://github.com/pytorch/pytorch/issues/101291,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-06-03T03:02:47Z +issue,102811,`torch.poisson(torch.tensor([torch.inf))` returns 0,https://github.com/pytorch/pytorch/issues/102811,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: edge cases, module: error checking",scan when time allows,2023-06-02T22:17:08Z +issue,102753,Error: no matching constructor for initialization of 'at::OptionalIntArrayRef',https://github.com/pytorch/pytorch/issues/102753,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2023-06-02T17:32:10Z +issue,102761,Pytorch Build images for RISCV64 Devices in the nightly builds,https://github.com/pytorch/pytorch/issues/102761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-06-02T16:10:40Z +issue,102812,[DTensor] Error in distribute_module with module._apply,https://github.com/pytorch/pytorch/issues/102812,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-06-02T13:02:21Z +issue,97757,oneDNN 3.0+ support,https://github.com/pytorch/pytorch/issues/97757,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: intel,scan when time allows,2023-06-02T07:48:47Z +issue,102732,Building NCCL with `make -l $MAX_JOBS` slows down builds,https://github.com/pytorch/pytorch/issues/102732,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-06-01T17:20:05Z +issue,67586,SequentialLR have a question and why it use `step(epoch)`,https://github.com/pytorch/pytorch/issues/67586,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2023-06-01T16:54:59Z +issue,102609,[prims] torch.ops.aten.le decomposition confuses scalars and tensors,https://github.com/pytorch/pytorch/issues/102609,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2023-06-01T16:09:49Z +issue,53758,ZeroDivisionError: float division by zero in Adam (bias_correction1 is zero),https://github.com/pytorch/pytorch/issues/53758,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: optimizer",scan when time allows,2023-06-01T14:19:03Z +issue,99432,[DTensor] parallelize_module failed with nn.Transformer and the PairwiseParallel plan,https://github.com/pytorch/pytorch/issues/99432,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-05-31T21:58:47Z +issue,102610,Cannot invoke prims.sum with output_dtype,https://github.com/pytorch/pytorch/issues/102610,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2023-05-31T13:23:50Z +issue,102515,pytorch java api documentation is not clear and does not cover example,https://github.com/pytorch/pytorch/issues/102515,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: java, oncall: mobile",scan when time allows,2023-05-31T04:05:22Z +issue,102556,test_functional_autograd_benchmark.py::TestFunctionalAutogradBenchmark::test_fast_tasks passes with all NaNs,https://github.com/pytorch/pytorch/issues/102556,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2023-05-31T00:04:34Z +issue,102501,[Utils][tensorboard]Enhancement: Include 'max_outputs' parameter in torch.utils.tensorboard.summary's 'image' method,https://github.com/pytorch/pytorch/issues/102501,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-05-31T00:01:58Z +issue,102533,Segfault when running vulkan program linked against libtorch,https://github.com/pytorch/pytorch/issues/102533,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2023-05-30T23:50:31Z +issue,102511,Faster BatchSampler with big batch size,https://github.com/pytorch/pytorch/issues/102511,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-05-30T23:48:47Z +issue,102459,Matrix multiplication performance regression in case of an additional dimension of size 1,https://github.com/pytorch/pytorch/issues/102459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: dependency bug",scan when time allows,2023-05-30T23:43:52Z +issue,102435,torch.distributed.all_reduce() has inconsistent behavior,https://github.com/pytorch/pytorch/issues/102435,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-05-30T21:16:41Z +issue,96111,Static size boolean masking,https://github.com/pytorch/pytorch/issues/96111,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2023-05-30T19:41:57Z +issue,102272,"after add /path_to_libtorch/libtorch/lib to LD_LIBRARY_PATH, I can't import torch_scatter.",https://github.com/pytorch/pytorch/issues/102272,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: build, module: cpp",scan when time allows,2023-05-27T00:41:42Z +issue,102368,Issue with ShufflerIterDataPipe in torch 1.13.1,https://github.com/pytorch/pytorch/issues/102368,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2023-05-26T19:54:39Z +issue,102081,CPU Fallback does not convert Tensor?[],https://github.com/pytorch/pytorch/issues/102081,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2023-05-25T22:56:18Z +issue,101653,Unexpected behavior comparing uint8 tensor to value greater than 255,https://github.com/pytorch/pytorch/issues/101653,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: type promotion",scan when time allows,2023-05-25T14:26:21Z +issue,101890,Can't reproduce/non-deterministic results with CUDA,https://github.com/pytorch/pytorch/issues/101890,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism",scan when time allows,2023-05-24T20:15:18Z +issue,95121,cuda 12 support request.,https://github.com/pytorch/pytorch/issues/95121,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-05-24T16:29:41Z +issue,14701,CI: Flaky download from download.pytorch.org,https://github.com/pytorch/pytorch/issues/14701,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: flaky-tests",scan when time allows,2023-05-23T19:36:59Z +issue,55607,pytorch inference lead to memory leak in cpu,https://github.com/pytorch/pytorch/issues/55607,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2023-05-23T13:47:37Z +issue,100062,add github check that diffs generated code,https://github.com/pytorch/pytorch/issues/100062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: infra",scan when time allows,2023-05-23T12:17:00Z +issue,101968,CrossEntropyLoss output difference on Windows vs. Linux,https://github.com/pytorch/pytorch/issues/101968,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2023-05-23T08:18:39Z +issue,100785,compile torch2.0 in debug mode,https://github.com/pytorch/pytorch/issues/100785,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-05-23T00:43:15Z +issue,99311,Functorch pytrees with custom iterables,https://github.com/pytorch/pytorch/issues/99311,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: pytree",scan when time allows,2023-05-22T17:50:36Z +issue,36217,Rectify docs for MultiLabelSoftMarginLoss,https://github.com/pytorch/pytorch/issues/36217,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2023-05-21T23:47:53Z +issue,38271,"PyTorch's flip returns a new tensor, but NumPy's flip returns a view",https://github.com/pytorch/pytorch/issues/38271,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-05-20T10:44:28Z +issue,101871,Can group convolution support other grouping methods?,https://github.com/pytorch/pytorch/issues/101871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2023-05-19T20:29:23Z +issue,101428,2D inputs to linear layers run up to 25% slower than 4D ones on some Nvidia GPUs,https://github.com/pytorch/pytorch/issues/101428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2023-05-19T08:50:21Z +issue,10615,Request for better memory management,https://github.com/pytorch/pytorch/issues/10615,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2023-05-18T19:31:56Z +issue,101781,Add support for bfloat16 in torch.from_numpy(),https://github.com/pytorch/pytorch/issues/101781,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: numpy",scan when time allows,2023-05-18T17:24:02Z +issue,101536,"torch.cuda.set_device cannot use to set cpu device, but give an ambiguity hint",https://github.com/pytorch/pytorch/issues/101536,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-05-18T07:08:24Z +issue,101716,"[TorchScript] aten::__and__ with argument types: Tensor, bool not supported",https://github.com/pytorch/pytorch/issues/101716,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-05-17T23:26:21Z +issue,70166,[feature request] Autocast module and function wrappers,https://github.com/pytorch/pytorch/issues/70166,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-05-17T20:35:13Z +issue,101669,[refs] inplace references resize the input to match the broadcasted input shape,https://github.com/pytorch/pytorch/issues/101669,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2023-05-17T17:59:15Z +issue,101586,"/Users/davidlaxer/pytorch/third_party/tensorpipe/third_party/libuv/src/unix/getaddrinfo.c:165:10: error: implicit declaration of function 'uv__idna_toascii' [-Werror,-Wimplicit-function-declaration] rc = uv__idna_toascii(hostname,",https://github.com/pytorch/pytorch/issues/101586,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-05-17T17:47:54Z +issue,101564,not yet implemented the batching rule for torchaudio::_lfilter,https://github.com/pytorch/pytorch/issues/101564,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: functorch",scan when time allows,2023-05-17T17:45:28Z +issue,101506,Support pipeline parallelism with PyG,https://github.com/pytorch/pytorch/issues/101506,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-05-17T17:40:18Z +issue,101470,[bazel] add inductor to bazel build,https://github.com/pytorch/pytorch/issues/101470,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: build",scan when time allows,2023-05-17T17:39:08Z +issue,101319,Deprecated File bug,https://github.com/pytorch/pytorch/issues/101319,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-05-17T17:36:45Z +issue,101415,import functorch.dim monkeypatches torch,https://github.com/pytorch/pytorch/issues/101415,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-05-17T17:36:25Z +issue,96743,Pruning under channels_last format,https://github.com/pytorch/pytorch/issues/96743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: pruning",scan when time allows,2023-05-17T14:09:51Z +issue,101505,Investigate random sequence number broadcast initially incorrect,https://github.com/pytorch/pytorch/issues/101505,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-05-16T16:59:06Z +issue,87159,"torch/csrc/utils/python_arg_parser.h:424:94: error: format ‘%ld’ expects argument of type ‘long int’, but argument 7 has type ‘int’",https://github.com/pytorch/pytorch/issues/87159,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",scan when time allows,2023-05-16T16:55:28Z +issue,79967,[feature request] LazyTensor that provides/loads/computes its contents only upon request to be returned from torch.load,https://github.com/pytorch/pytorch/issues/79967,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2023-05-15T23:35:58Z +issue,93206,Build from Source Issues on MacOS Ventura 13.2,https://github.com/pytorch/pytorch/issues/93206,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2023-05-15T18:02:34Z +issue,101251,Mac m2 MPSNDArray.mm:78: failed assertion `[MPSNDArrayDescriptor sliceDimension:withSubrange:] error: dimension index (2) not within number of dimensions (2) Dimension indices are 0-based',https://github.com/pytorch/pytorch/issues/101251,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2023-05-15T16:21:01Z +issue,101359,RuntimeError in Scaled Dot Product Attention Tutorial Code,https://github.com/pytorch/pytorch/issues/101359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2023-05-15T13:48:59Z +issue,101334,Functions for Calculating Skewness and Kurtosis,https://github.com/pytorch/pytorch/issues/101334,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2023-05-15T10:25:05Z +issue,94816,grid_sample with relative grid,https://github.com/pytorch/pytorch/issues/94816,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn",scan when time allows,2023-05-15T07:04:10Z +issue,51429,"dist.init_process_group('nccl', ...) hangs in some combinations of pytorch+python+cuda version",https://github.com/pytorch/pytorch/issues/51429,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-05-14T23:20:20Z +issue,101177,version 4.26.1 to 4.29.0 has two bugs,https://github.com/pytorch/pytorch/issues/101177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-05-14T03:58:48Z +issue,101210,GLCM implementation in pytorch C++ api and cuda,https://github.com/pytorch/pytorch/issues/101210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-05-12T19:53:58Z +issue,80296,Sort tensors inplace,https://github.com/pytorch/pytorch/issues/80296,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2023-05-12T18:26:17Z +issue,101246,Tool for identifying where in eager model an operation is nondeterministic,https://github.com/pytorch/pytorch/issues/101246,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2023-05-12T14:21:45Z +issue,81749,BatchNorm for complex tensor,https://github.com/pytorch/pytorch/issues/81749,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: primtorch",scan when time allows,2023-05-12T13:47:37Z +issue,51623,[docs] FractionalMaxPool2d : _random_samples arg is present in signature but undocumented.,https://github.com/pytorch/pytorch/issues/51623,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: pooling",scan when time allows,2023-05-12T08:49:18Z +issue,100957,PyTorch installs the file mkldnn.cmake that looks for the package MKLDNN that doesn't exist,https://github.com/pytorch/pytorch/issues/100957,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn",scan when time allows,2023-05-12T08:10:55Z +issue,101189,`dense -> sparse compressed` to work with empty batches.,https://github.com/pytorch/pytorch/issues/101189,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-05-12T01:05:34Z +issue,101159,Inconsistency between GPU memory usage in torch.cuda.memory_summary and nvidia-smi,https://github.com/pytorch/pytorch/issues/101159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-05-12T01:03:24Z +issue,101080,Sequence annotation in type hints is wrong,https://github.com/pytorch/pytorch/issues/101080,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2023-05-11T23:09:31Z +issue,100123,Error building Pytorch from source,https://github.com/pytorch/pytorch/issues/100123,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: rocm",scan when time allows,2023-05-11T20:05:29Z +issue,100617,GPU VRAM usage significantly higher for Lenet5 models when compared to other frameworks,https://github.com/pytorch/pytorch/issues/100617,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-05-11T14:18:28Z +issue,37092,nn.Bilinear cannot be used inside nn.Sequential,https://github.com/pytorch/pytorch/issues/37092,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-05-11T07:48:29Z +issue,101070,Lazily format C++ stack trace if it is not used,https://github.com/pytorch/pytorch/issues/101070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2023-05-11T04:00:21Z +issue,101106,"addmv doesn't do type promotion correctly,",https://github.com/pytorch/pytorch/issues/101106,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: type promotion",scan when time allows,2023-05-11T03:37:34Z +issue,101073,Windows fatal exception: stack overflow while using pytorch for computing,https://github.com/pytorch/pytorch/issues/101073,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-05-11T03:23:07Z +issue,100850,[BUG] Poor torch.bmm performance on H100,https://github.com/pytorch/pytorch/issues/100850,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: performance",scan when time allows,2023-05-10T08:04:02Z +issue,100932,torch.concat fails with float16 input in autocast(device_type=cpu) context,https://github.com/pytorch/pytorch/issues/100932,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: bfloat16, module: cpu",scan when time allows,2023-05-10T06:27:04Z +issue,100705,torch.cuda.amp.GradScaler initialization,https://github.com/pytorch/pytorch/issues/100705,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-05-10T05:39:57Z +issue,67680,OOM with a lot of GPU memory left,https://github.com/pytorch/pytorch/issues/67680,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2023-05-10T02:19:29Z +issue,96637,Not allow force merge when lint fails and not because of broken trunk,https://github.com/pytorch/pytorch/issues/96637,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2023-05-09T19:01:13Z +issue,95645,Create a new Docker image with all inductor benchmarks and pre-trained models downloaded,https://github.com/pytorch/pytorch/issues/95645,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: devx",scan when time allows,2023-05-09T18:55:20Z +issue,98948,behaviour of `torch.tensor()` changes after editing `Tensor.__getitem__`,https://github.com/pytorch/pytorch/issues/98948,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-05-09T18:32:36Z +issue,99544,Cross compile Pytorch for ARM in Bazel,https://github.com/pytorch/pytorch/issues/99544,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",scan when time allows,2023-05-09T17:53:55Z +issue,30574,`index_select` with multidimensional `index`,https://github.com/pytorch/pytorch/issues/30574,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2023-05-09T15:07:38Z +issue,90261,Why torch.mode return different value between CPU and GPU,https://github.com/pytorch/pytorch/issues/90261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-05-09T09:16:14Z +issue,100801,There is a performance drop because we have not yet implemented the batching rule for aten::native_dropout_backward,https://github.com/pytorch/pytorch/issues/100801,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-05-08T22:18:12Z +issue,100795,Quickstart notebook fails to train properly with ROCm,https://github.com/pytorch/pytorch/issues/100795,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2023-05-08T22:16:50Z +issue,100904,Revise glossary,https://github.com/pytorch/pytorch/issues/100904,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-05-08T22:10:19Z +issue,35633,Integration of Large Model Support in PyTorch,https://github.com/pytorch/pytorch/issues/35633,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: memory usage",scan when time allows,2023-05-08T19:22:36Z +issue,50402,ValueError: optimizer got an empty parameter list,https://github.com/pytorch/pytorch/issues/50402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-05-08T17:35:02Z +issue,100807,[torch.compile] returns output with WRONG SHAPE after `cat_slice_cat`,https://github.com/pytorch/pytorch/issues/100807,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-05-07T11:49:17Z +issue,100584,[doc] torch.scalar_tensor doc is missing,https://github.com/pytorch/pytorch/issues/100584,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-05-05T13:35:36Z +issue,84972,Add unit tests for test decorators,https://github.com/pytorch/pytorch/issues/84972,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2023-05-05T02:06:17Z +issue,66787,loading large model not finished after 16 hours,https://github.com/pytorch/pytorch/issues/66787,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: serialization",scan when time allows,2023-05-04T05:20:59Z +issue,99640,Ban GradScaler scale from being less than 1,https://github.com/pytorch/pytorch/issues/99640,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-05-03T17:23:22Z +issue,100425,Higher GPU consumption for Lenet-5 and LSTM models when compared to other frameworks,https://github.com/pytorch/pytorch/issues/100425,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2023-05-03T15:28:52Z +issue,99630,Parameterisation of MultivariateNormal distribution using Cholesky decomposition of precision matrix,https://github.com/pytorch/pytorch/issues/99630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-05-03T15:26:18Z +issue,99287,Add `TORCH_ASSERT_ONLY_METHOD_OPERATORS` to functorch codebase,https://github.com/pytorch/pytorch/issues/99287,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-05-03T14:16:02Z +issue,99558,Batching rule not implemented for aten::narrow.Tensor,https://github.com/pytorch/pytorch/issues/99558,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-05-02T21:32:06Z +issue,18220,Context Manager that disables training mode with in a nn.Module.,https://github.com/pytorch/pytorch/issues/18220,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-05-01T19:19:22Z +issue,100244,torch.utils._content_store will deduplicate storage with identical contents; may be problematic for mutation,https://github.com/pytorch/pytorch/issues/100244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-05-01T15:32:00Z +issue,87433,index_select() applied in sparse tensor can't backprop,https://github.com/pytorch/pytorch/issues/87433,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-05-01T14:33:11Z +issue,100145,This flag not work : torch.backends.cudnn.allow_tf32 = False,https://github.com/pytorch/pytorch/issues/100145,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-04-28T09:05:32Z +issue,94115,`cat` fails the gradient computation in forward mode with empty tensors when used with legacy vmap,https://github.com/pytorch/pytorch/issues/94115,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: forward ad",scan when time allows,2023-04-27T15:33:03Z +issue,66357,OpInfos disabled for batched forward grad computation,https://github.com/pytorch/pytorch/issues/66357,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, module: tests, module: vmap",scan when time allows,2023-04-27T02:59:50Z +issue,99299,There has implmenet bug in LTC IrBuilder's MakeSizeMul method.,https://github.com/pytorch/pytorch/issues/99299,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2023-04-26T19:33:01Z +issue,37246,`max_norm` parameter on nn.Embedding will fail inside nn.DataParallel,https://github.com/pytorch/pytorch/issues/37246,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nn",scan when time allows,2023-04-26T19:08:53Z +issue,93884,Importing tensorflow (2.12) before torch (2.0) hangs at import torch,https://github.com/pytorch/pytorch/issues/93884,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-04-26T16:49:46Z +issue,100012,Dataloader multiprocess loading with num_worker > 0 calls __main__ file to run,https://github.com/pytorch/pytorch/issues/100012,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-04-26T14:04:29Z +issue,98073,Add PrivateUse1 folder in aten/src/ATen,https://github.com/pytorch/pytorch/issues/98073,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2023-04-26T14:01:11Z +issue,98414,"PyTorch 1.12, high failure rate for test_optim/test_nadam",https://github.com/pytorch/pytorch/issues/98414,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-04-25T05:15:21Z +issue,99790,Cannot compile torch 1.10 in CentOS 7.3,https://github.com/pytorch/pytorch/issues/99790,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-04-25T00:45:43Z +issue,99770,Deformable Convolution export to onnx,https://github.com/pytorch/pytorch/issues/99770,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2023-04-24T23:55:21Z +issue,99812,cuda.is_available() error,https://github.com/pytorch/pytorch/issues/99812,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-04-24T23:54:52Z +issue,99836,NTK notebook calculates wrong object - wrong output dimensions,https://github.com/pytorch/pytorch/issues/99836,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-04-24T23:52:23Z +issue,99681,`torch.ops.aten.empty` is not discoverable from `dir(torch.ops.aten)` until explicitly calling getattr,https://github.com/pytorch/pytorch/issues/99681,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: library,scan when time allows,2023-04-24T20:53:16Z +issue,98924,Integrate open device privateuse1 customized method registration,https://github.com/pytorch/pytorch/issues/98924,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2023-04-24T17:50:11Z +issue,99710,pca_lowrank and svd_lowrank broken under automatic mixed precision.,https://github.com/pytorch/pytorch/issues/99710,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: half, module: linear algebra",scan when time allows,2023-04-24T11:31:13Z +issue,99515,Support polyphase channelizer,https://github.com/pytorch/pytorch/issues/99515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2023-04-24T07:40:55Z +issue,99690,"gpu training work well, but cpu training not work",https://github.com/pytorch/pytorch/issues/99690,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: fft",scan when time allows,2023-04-23T22:30:34Z +issue,96738,Many padding Module fail memory_format tests,https://github.com/pytorch/pytorch/issues/96738,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: intel, module: memory format, module: nn",scan when time allows,2023-04-23T07:14:28Z +issue,95945,CPU time performance is unstable,https://github.com/pytorch/pytorch/issues/95945,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2023-04-23T05:02:53Z +issue,76751,Default repr of __get__ methods in __torch_function__ is bad,https://github.com/pytorch/pytorch/issues/76751,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2023-04-22T15:58:55Z +issue,29637,Compiled functions can't take variable number of arguments,https://github.com/pytorch/pytorch/issues/29637,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-04-21T21:56:22Z +issue,99684,In torchelastic support running worker rank 0 on agent rank 0 consistently,https://github.com/pytorch/pytorch/issues/99684,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2023-04-21T20:11:09Z +issue,98200,torch.nn.init functions with `generator` argument,https://github.com/pytorch/pytorch/issues/98200,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: random",scan when time allows,2023-04-21T19:26:37Z +issue,96855,Performance Drop for linalg_ldl_factor and ldl_solve,https://github.com/pytorch/pytorch/issues/96855,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: vmap",scan when time allows,2023-04-20T20:39:33Z +issue,48486,[complex] torch.abs: does not match numpy,https://github.com/pytorch/pytorch/issues/48486,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs, module: numpy",scan when time allows,2023-04-20T10:50:14Z +issue,99138,torch.cuda.is_available() return False,https://github.com/pytorch/pytorch/issues/99138,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2023-04-20T08:42:12Z +issue,67684,[feature request] torch.clamp on BoolTensors,https://github.com/pytorch/pytorch/issues/67684,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: numpy",scan when time allows,2023-04-19T21:20:34Z +issue,71552,[docs] Tensor.uniform_ docs are not clear about whether from/to boundary values are included in sampling or not,https://github.com/pytorch/pytorch/issues/71552,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: random",scan when time allows,2023-04-19T21:12:18Z +issue,99410,"torch.nn.functional.multilabel_margin_loss cuda lacks checking of ""out of bound""",https://github.com/pytorch/pytorch/issues/99410,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking, module: nn",scan when time allows,2023-04-19T20:20:41Z +issue,99351,we should make semantically meaningless positional arguments positional only in our operator API,https://github.com/pytorch/pytorch/issues/99351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: python frontend",scan when time allows,2023-04-19T12:41:36Z +issue,27030,Tensorboard logging image WITH LABEL,https://github.com/pytorch/pytorch/issues/27030,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-04-19T09:31:27Z +issue,52495,link failed when using custom build pytorch on Android,https://github.com/pytorch/pytorch/issues/52495,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: arm, module: undefined reference",scan when time allows,2023-04-19T06:57:09Z +issue,31815,ImportError: /home/xx/anaconda3/lib/python3.7/site-packages/torch/lib/libtorch_python.so: undefined symbol: _ZNK5torch3jit5Graph8toStringE,https://github.com/pytorch/pytorch/issues/31815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-04-19T05:22:03Z +issue,53440,Add complex support for torch.unique,https://github.com/pytorch/pytorch/issues/53440,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy, module: python array api",scan when time allows,2023-04-19T05:01:54Z +issue,99407,Torch.fx.symbolic_trace removes some of the keys from module state_dict,https://github.com/pytorch/pytorch/issues/99407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-04-19T02:22:26Z +issue,99414,The meta implementation of `index_put` does not do any check,https://github.com/pytorch/pytorch/issues/99414,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2023-04-19T02:05:13Z +issue,98792,DataLoader doesn't accept non-cpu device for loading.,https://github.com/pytorch/pytorch/issues/98792,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-04-18T15:04:02Z +issue,99359,[c++17] Replace lock_guard with scoped_lock,https://github.com/pytorch/pytorch/issues/99359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2023-04-18T14:36:10Z +issue,99316,torch.linalg.lstsq doc arguments error,https://github.com/pytorch/pytorch/issues/99316,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: linear algebra",scan when time allows,2023-04-18T12:31:42Z +issue,25092,No way to correctly reset weights of a model with spectral norm,https://github.com/pytorch/pytorch/issues/25092,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-04-18T09:07:45Z +issue,99265,the error message of torch.addcmul is wrong,https://github.com/pytorch/pytorch/issues/99265,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2023-04-17T17:13:30Z +issue,99310,Torch func Documentation for trees,https://github.com/pytorch/pytorch/issues/99310,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: functorch",scan when time allows,2023-04-17T16:00:03Z +issue,99248,tools PYTHONPATH trick in run_test.py does not work reliably,https://github.com/pytorch/pytorch/issues/99248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2023-04-17T15:26:48Z +issue,99147,`torch.sparse.sum` backward fails when reducing over dense dimensions.,https://github.com/pytorch/pytorch/issues/99147,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-04-17T08:22:47Z +issue,99160,"Run ChatRWKV on MBP(intel CPU)+eGPU[rx6800 16G], returna a very big number -9223372036854775808, looks like overflow",https://github.com/pytorch/pytorch/issues/99160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2023-04-14T20:21:05Z +issue,99042,Memory allocation issues in distributions.multivariate_normal.MultivariateNormal,https://github.com/pytorch/pytorch/issues/99042,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-04-14T17:19:14Z +issue,99142,"More Nested Tensor Functionality (layer_norm, cross_entropy / log_softmax&nll_loss)",https://github.com/pytorch/pytorch/issues/99142,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-04-14T16:29:52Z +issue,99023,Dtype changes while going from FX graph -> Torchscript,https://github.com/pytorch/pytorch/issues/99023,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-04-14T15:49:23Z +issue,36041,[DISCUSSION] Better user experience for debugging on Windows,https://github.com/pytorch/pytorch/issues/36041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: windows",scan when time allows,2023-04-13T22:34:03Z +issue,98917,Change module to module_ in torch/csrc/api/include/torch/python.h,https://github.com/pytorch/pytorch/issues/98917,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-04-13T21:47:57Z +issue,98977,[BE] Dedup the functorch skipOps mechanism and the common_method_invocations one,https://github.com/pytorch/pytorch/issues/98977,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2023-04-13T19:58:54Z +issue,98942,[torch.fx] Upgrade on node info,https://github.com/pytorch/pytorch/issues/98942,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-04-13T07:55:40Z +issue,98863,Problem with instalation torch2 on a100+cu12.1,https://github.com/pytorch/pytorch/issues/98863,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2023-04-12T21:06:08Z +issue,1794,Feature request: reverse_padded_sequence,https://github.com/pytorch/pytorch/issues/1794,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2023-04-12T16:04:14Z +issue,98907,Move template code to header,https://github.com/pytorch/pytorch/issues/98907,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-04-12T04:02:19Z +issue,98805,Discrepancy of supported Python versions between Get Started page and index of pre-built binaries for PIP installation,https://github.com/pytorch/pytorch/issues/98805,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-04-11T23:40:12Z +issue,98673,[ux] Non-blocking tensor constructors,https://github.com/pytorch/pytorch/issues/98673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2023-04-11T21:05:27Z +issue,98871,Document the user-facing API for the component-level logging system,https://github.com/pytorch/pytorch/issues/98871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-04-11T20:54:47Z +issue,98587,Cannot use AT_CUDA_DRIVER_CHECK from user code,https://github.com/pytorch/pytorch/issues/98587,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions, module: cuda, module: internals",scan when time allows,2023-04-11T19:19:40Z +issue,98557,torch.jit.script codegen warning with cuda and vmap,https://github.com/pytorch/pytorch/issues/98557,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, oncall: jit",scan when time allows,2023-04-11T12:42:15Z +issue,98808,[FSDP] move up the first all gather,https://github.com/pytorch/pytorch/issues/98808,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2023-04-11T12:37:52Z +issue,91334,Pytorch 1.13 conda package with cuda requires too many unneccessary packages,https://github.com/pytorch/pytorch/issues/91334,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2023-04-11T07:33:33Z +issue,98617,Add test/distributed/test_c10d_mpi.py,https://github.com/pytorch/pytorch/issues/98617,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-04-11T05:39:48Z +issue,98286,"When I use the DDP model, I use a custom loss function, when the batch size changes during training, the process will be stuck.",https://github.com/pytorch/pytorch/issues/98286,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-04-11T05:28:29Z +issue,98515,torch.cond should work with expressions involving SymInt,https://github.com/pytorch/pytorch/issues/98515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-04-10T17:04:39Z +issue,98541,Memory corruption using torch.ops.* to access re-registered operator,https://github.com/pytorch/pytorch/issues/98541,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2023-04-10T16:56:57Z +issue,98499,`torch.nn.utils.rnn.unpad_sequence` modifies arguments in-place,https://github.com/pytorch/pytorch/issues/98499,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rnn",scan when time allows,2023-04-10T16:56:18Z +issue,89730,torch.addbmm throws different exception differences on CPU and GPU.,https://github.com/pytorch/pytorch/issues/89730,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",scan when time allows,2023-04-10T13:13:07Z +issue,88144,AvgPool2D output shapes are inconsistent when ceil_mode=True,https://github.com/pytorch/pytorch/issues/88144,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: pooling",scan when time allows,2023-04-10T05:42:31Z +issue,41164,Pytorch.js would be nice!,https://github.com/pytorch/pytorch/issues/41164,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-04-09T17:45:08Z +issue,89393,Edge case: CPU bool abs is not supported,https://github.com/pytorch/pytorch/issues/89393,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: numpy",scan when time allows,2023-04-09T12:14:18Z +issue,42095,[RFC] Add Windows support to torch.distributed package,https://github.com/pytorch/pytorch/issues/42095,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: distributed",scan when time allows,2023-04-09T09:16:45Z +issue,37487,Log-linear version of cumsum and cumprod,https://github.com/pytorch/pytorch/issues/37487,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2023-04-08T06:29:07Z +issue,91608,SSL: CERTIFICATE_VERIFY_FAILED while trying to download pretrained model within a company that transforms SSL certificates for security purposes,https://github.com/pytorch/pytorch/issues/91608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2023-04-08T01:45:46Z +issue,97498,Function Registry for extending collate_fn,https://github.com/pytorch/pytorch/issues/97498,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-04-07T22:32:49Z +issue,98498,"Higher order derivatives not working when setting compute device to `torch.device(""mps"")`",https://github.com/pytorch/pytorch/issues/98498,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: mps",scan when time allows,2023-04-07T14:33:31Z +issue,24468,[Tensorboard] Write summaries to S3 or GCS bucket,https://github.com/pytorch/pytorch/issues/24468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-04-07T08:05:03Z +issue,28594,[BUG] Can't Deepcopy module with weightnorm,https://github.com/pytorch/pytorch/issues/28594,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-04-07T07:08:50Z +issue,94912,PR #88607 breaks build for POWER9 CPU,https://github.com/pytorch/pytorch/issues/94912,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2023-04-06T15:38:31Z +issue,98484,Improvements to FSDP debugability,https://github.com/pytorch/pytorch/issues/98484,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fsdp, oncall: distributed",scan when time allows,2023-04-06T11:30:35Z +issue,88464,MaxPool1D output shapes can be negative when ceil_mode=True,https://github.com/pytorch/pytorch/issues/88464,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2023-04-06T07:37:40Z +issue,92128,Forward arguments are not updated in DDP,https://github.com/pytorch/pytorch/issues/92128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-04-06T06:55:35Z +issue,98169,The operator 'aten::_weight_norm_interface' is not currently implemented for the MPS device.,https://github.com/pytorch/pytorch/issues/98169,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2023-04-05T19:28:37Z +issue,98409,`torch.Tensor.layout` is not documented,https://github.com/pytorch/pytorch/issues/98409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: python frontend",scan when time allows,2023-04-05T17:17:39Z +issue,98075,Unexpected results with torch.nn.functional.layer_norm,https://github.com/pytorch/pytorch/issues/98075,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization, module: numerical-stability",scan when time allows,2023-04-05T15:05:12Z +issue,92251,"`torch.load(..., map_location=""cuda:0"")` allocates memory on both CPU and GPU",https://github.com/pytorch/pytorch/issues/92251,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: serialization",scan when time allows,2023-04-05T14:56:37Z +issue,98338,[functorch] [vmap] tests fail when `_set_vmap_fallback_enabled(False)`.,https://github.com/pytorch/pytorch/issues/98338,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-04-05T14:18:19Z +issue,95369,build failed when strictly following the guidelines,https://github.com/pytorch/pytorch/issues/95369,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-04-05T10:01:02Z +issue,73554,Build failure using GCC 11.2.0,https://github.com/pytorch/pytorch/issues/73554,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn, module: third_party",scan when time allows,2023-04-05T09:12:04Z +issue,95024,cuDNN doesn't support convolutions with more than `INT_MAX` elements and native kernel uses too much memory,https://github.com/pytorch/pytorch/issues/95024,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudacachingallocator, module: cudnn",scan when time allows,2023-04-05T03:48:48Z +issue,98070,Request custom backend device memory Allocator.,https://github.com/pytorch/pytorch/issues/98070,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: cudacachingallocator, module: memory usage",scan when time allows,2023-04-05T00:44:08Z +issue,98004,matmul with CSR matrix in inference mode throws an exception,https://github.com/pytorch/pytorch/issues/98004,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-04-05T00:40:45Z +issue,97902,Multiple model init using OpenMP in c++ does not speed up,https://github.com/pytorch/pytorch/issues/97902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2023-04-05T00:40:43Z +issue,97638,Regression in jit for f-strings with new lines,https://github.com/pytorch/pytorch/issues/97638,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-04-05T00:40:41Z +issue,97595,can get_submodule be called within a ScriptFunction ?,https://github.com/pytorch/pytorch/issues/97595,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-04-05T00:40:39Z +issue,73638,addcdiv is failing the ASAN test for zero divisors,https://github.com/pytorch/pytorch/issues/73638,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: tests",scan when time allows,2023-04-05T00:40:38Z +issue,89420,Edge case: torch.baddbmm supports double x int8 x int8 inputs on CPU but not CUDA,https://github.com/pytorch/pytorch/issues/89420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: linear algebra, module: type promotion",scan when time allows,2023-04-05T00:39:00Z +issue,89419,torch.equal can still run successfully when the parameter types are different.,https://github.com/pytorch/pytorch/issues/89419,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: type promotion",scan when time allows,2023-04-05T00:26:21Z +issue,97236,Non-deterministic results when training a model on GPU with MPS backend,https://github.com/pytorch/pytorch/issues/97236,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: mps",scan when time allows,2023-04-04T21:55:03Z +issue,89418,"torch.floor_divide: The dividend of torch.floor_divide is set to 0, but it can still run on the GPU.",https://github.com/pytorch/pytorch/issues/89418,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2023-04-04T15:52:18Z +issue,89381,torch.nn.ReplicationPad1d:The description of the exception information thrown is not accurate,https://github.com/pytorch/pytorch/issues/89381,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: padding",scan when time allows,2023-04-04T15:30:42Z +issue,98260,Broken mypy check in test_type_hints.py::TestTypeHints::test_doc_examples,https://github.com/pytorch/pytorch/issues/98260,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2023-04-04T14:06:34Z +issue,98193,"RuntimeError: CUDA error: an illegal memory access was encountered, torch/cuda/streams.py"", line 94, in synchronize",https://github.com/pytorch/pytorch/issues/98193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-04-04T13:47:35Z +issue,70073,Feature request: [STFT] Add warning message if signal length is not a multiple of hop_length in torch.stft,https://github.com/pytorch/pytorch/issues/70073,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: fft",scan when time allows,2023-04-04T12:32:58Z +issue,1512,[feature request] Support tensors of different sizes as batch elements in DataLoader,https://github.com/pytorch/pytorch/issues/1512,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: nestedtensor",scan when time allows,2023-04-04T10:43:01Z +issue,89369,RuntimeError: CUDA error: device-side assert triggered,https://github.com/pytorch/pytorch/issues/89369,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2023-04-04T01:30:25Z +issue,89342,torch.mm: Exceptions thrown on the CPU and GPU are inconsistent,https://github.com/pytorch/pytorch/issues/89342,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",scan when time allows,2023-04-04T01:06:10Z +issue,97849,TORCH_LIBRARIES variable leads to undefined reference function error in compiling while using libtorch in c++,https://github.com/pytorch/pytorch/issues/97849,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2023-04-03T17:09:55Z +issue,18993,ninja: build stopped: subcommand failed.,https://github.com/pytorch/pytorch/issues/18993,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-04-03T01:59:00Z +issue,77253,lintrunner not working,https://github.com/pytorch/pytorch/issues/77253,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2023-04-02T18:21:39Z +issue,55399,[FX][testing] Test symbolic tracing of detectron2,https://github.com/pytorch/pytorch/issues/55399,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-04-02T02:40:06Z +issue,97961,[CI/Infra] Record keeping: runner shutdown spike,https://github.com/pytorch/pytorch/issues/97961,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-04-02T01:27:30Z +issue,93947,[RFC] Support Huge Model Init Without mallocs for Compile/Distributed Use Cases,https://github.com/pytorch/pytorch/issues/93947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-04-01T00:17:01Z +issue,97030,DDP static graph fails for static model,https://github.com/pytorch/pytorch/issues/97030,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2023-03-31T19:27:25Z +issue,83244,torch.nn.Upsample's error message is inconsistent with the documentation,https://github.com/pytorch/pytorch/issues/83244,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2023-03-31T15:35:28Z +issue,97966,torch.randn signature is missing generator,https://github.com/pytorch/pytorch/issues/97966,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: python frontend, module: random",scan when time allows,2023-03-31T14:44:53Z +issue,97676,Pytorch 2 compile + fsdp + transformers crash,https://github.com/pytorch/pytorch/issues/97676,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xla,scan when time allows,2023-03-30T21:20:19Z +issue,97991,functional collective should respect the whole mesh,https://github.com/pytorch/pytorch/issues/97991,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-03-30T16:40:04Z +issue,7617,"checkpoint(function, *args) should have the same requires_grad as function(*args)",https://github.com/pytorch/pytorch/issues/7617,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd",scan when time allows,2023-03-30T14:16:59Z +issue,92285,InstanceNorm operator support for Vulkan devices,https://github.com/pytorch/pytorch/issues/92285,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2023-03-30T09:34:19Z +issue,95934,arange bug,https://github.com/pytorch/pytorch/issues/95934,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2023-03-30T03:26:42Z +issue,97915,Investigate Lazy{*}Norm{*}d modules no batch dim support,https://github.com/pytorch/pytorch/issues/97915,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-03-29T21:54:36Z +issue,97823,Using `param in param_list` can trigger `non-singleton dimension` error?,https://github.com/pytorch/pytorch/issues/97823,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-03-29T18:34:16Z +issue,97852,Functionalize crashes on train_step GraphModule,https://github.com/pytorch/pytorch/issues/97852,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2023-03-29T18:09:21Z +issue,97500,.set_ operation on a view (detach()) of the view tensor changes grad_fn of the original view tensor from ViewBackward0 to AsStridedBackward0,https://github.com/pytorch/pytorch/issues/97500,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-03-29T15:59:40Z +issue,81412,num_worker and prefetch_factor in DataLoader do not scale,https://github.com/pytorch/pytorch/issues/81412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2023-03-29T02:05:18Z +issue,97802,Some c++ library docstrings incorrectly linked/repeated,https://github.com/pytorch/pytorch/issues/97802,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs, module: mps",scan when time allows,2023-03-28T18:53:12Z +issue,97761,torch.testing.assert_close: allow check to fail on part on the input,https://github.com/pytorch/pytorch/issues/97761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2023-03-28T14:42:44Z +issue,90284,Public API definition is not compatible with `torch.testing`,https://github.com/pytorch/pytorch/issues/90284,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: testing",scan when time allows,2023-03-28T14:27:55Z +issue,97503,torch.cppExtension won't work with wsl2,https://github.com/pytorch/pytorch/issues/97503,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: wsl,scan when time allows,2023-03-28T07:55:00Z +issue,97659,How do I get the original object wrapped by the torch.fx.proxy class？,https://github.com/pytorch/pytorch/issues/97659,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: fx,scan when time allows,2023-03-28T02:10:58Z +issue,97656,[bug] Internal assert failed when using pyro,https://github.com/pytorch/pytorch/issues/97656,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: linear algebra, module: macos, module: python frontend",scan when time allows,2023-03-27T20:58:01Z +issue,97653,transposed 2d copy bfloat16 support,https://github.com/pytorch/pytorch/issues/97653,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-27T18:17:08Z +issue,97631,torch.zeros_like on a zero-sized BSR/BSC tensor results invalid tensor,https://github.com/pytorch/pytorch/issues/97631,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: sparse",scan when time allows,2023-03-27T18:07:10Z +issue,97597,Insufficient MPS Documentation,https://github.com/pytorch/pytorch/issues/97597,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: mps",scan when time allows,2023-03-27T17:37:24Z +issue,97189,torch.randint range for torch.int64 dtype seems wrong,https://github.com/pytorch/pytorch/issues/97189,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-03-27T17:28:45Z +issue,97504,"Burn benchmark suites into CI docker image. Not only this saves test time, but also it will get rid of occasional model installation failures. (@weiwangmeta )",https://github.com/pytorch/pytorch/issues/97504,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-27T16:45:42Z +issue,65153,Unification of model initialization methods / naming across domain libraries + support of skip_init,https://github.com/pytorch/pytorch/issues/65153,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: initialization,scan when time allows,2023-03-27T15:48:57Z +issue,97484,[DTensor] Add a unittest to cover default PG condition for DeviceMesh,https://github.com/pytorch/pytorch/issues/97484,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-03-27T14:49:13Z +issue,97114,[Feature Proposal: New Distributed Training Algorithms] LSGD and EASGD,https://github.com/pytorch/pytorch/issues/97114,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-03-27T14:40:19Z +issue,97670,[FSDP] test model.eval() + keep_low_precision_grads,https://github.com/pytorch/pytorch/issues/97670,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-03-27T14:14:24Z +issue,24904,tensorboard add_graph error,https://github.com/pytorch/pytorch/issues/24904,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-03-27T10:00:19Z +issue,95921,ROCm distributed flaky on test_distributed_spawn,https://github.com/pytorch/pytorch/issues/95921,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2023-03-24T22:23:53Z +issue,97329,"torch.ops.aten.pow(2.0, 3) return unexpected value with complex type",https://github.com/pytorch/pytorch/issues/97329,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: internals",scan when time allows,2023-03-24T17:31:32Z +issue,76244,Initial integration of ZenDNN as backend into PyTorch,https://github.com/pytorch/pytorch/issues/76244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2023-03-24T17:08:43Z +issue,97118,nn.Conv function to compute conv formula,https://github.com/pytorch/pytorch/issues/97118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2023-03-24T16:54:10Z +issue,80017,RPC init fails and crashes when world_size is greater than 18,https://github.com/pytorch/pytorch/issues/80017,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2023-03-24T10:38:15Z +issue,97397,Import fails when both `USE_TENSORPIPE=OFF` and `USE_DISTRIBUTED=ON`.,https://github.com/pytorch/pytorch/issues/97397,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: tensorpipe",scan when time allows,2023-03-24T05:42:55Z +issue,97408,Further memcopy improvement at FX body level,https://github.com/pytorch/pytorch/issues/97408,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-24T05:41:53Z +issue,96613,The output of torch.histc is incorrect on both CPU and CUDA,https://github.com/pytorch/pytorch/issues/96613,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-03-24T01:29:21Z +issue,97421,"After the release of pytorch 2.0.0, the compilation of ACLs is problematic.",https://github.com/pytorch/pytorch/issues/97421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-03-23T20:22:52Z +issue,97283,`jacfwd` fails when computing the gradient for `channels_last` tensor,https://github.com/pytorch/pytorch/issues/97283,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: memory format",scan when time allows,2023-03-23T08:04:27Z +issue,40087,Make Scaling in BatchNorm optional,https://github.com/pytorch/pytorch/issues/40087,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-23T04:20:07Z +issue,97068,[RFC] CPU float16 performance optimization on eager mode.,https://github.com/pytorch/pytorch/issues/97068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: half",scan when time allows,2023-03-23T02:52:42Z +issue,40967,Inconsistent behaviour when parameter appears multiple times in parameter list,https://github.com/pytorch/pytorch/issues/40967,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2023-03-22T16:24:55Z +issue,33248,PyTorch 1.4.0 CUDA initialization error with CPU-only (multiprocessing) on Python 3.7.5,https://github.com/pytorch/pytorch/issues/33248,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda",scan when time allows,2023-03-22T11:23:31Z +issue,97286,`jacrev` and `jacfwd` raise an error that `Sparse CSR tensors do not have strides`,https://github.com/pytorch/pytorch/issues/97286,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: sparse",scan when time allows,2023-03-21T22:33:29Z +issue,81750,Modernize logging tensor in torch.testing._internal,https://github.com/pytorch/pytorch/issues/81750,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: logging",scan when time allows,2023-03-21T19:46:30Z +issue,67913,Tracker: ModuleInfo-based testing,https://github.com/pytorch/pytorch/issues/67913,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: tests",scan when time allows,2023-03-21T17:14:21Z +issue,97196,Sequential/Partial unpickling and loading of models,https://github.com/pytorch/pytorch/issues/97196,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pickle,scan when time allows,2023-03-21T09:58:33Z +issue,97004,Small learning rate with `capturable=True` causes Adam optimizer to blow up model parameters.,https://github.com/pytorch/pytorch/issues/97004,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2023-03-21T09:04:43Z +issue,97088,A Segment Fault can be triggered in torch.embedding,https://github.com/pytorch/pytorch/issues/97088,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-03-20T23:56:22Z +issue,97087,A Segment Fault can be triggered in torch.adjoint,https://github.com/pytorch/pytorch/issues/97087,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-03-20T23:56:07Z +issue,97026,How to get list of all valid devices?,https://github.com/pytorch/pytorch/issues/97026,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-03-20T23:49:13Z +issue,97014,deprecate integral and boolean dtype support torch.logit and torch.special.logit,https://github.com/pytorch/pytorch/issues/97014,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: special",scan when time allows,2023-03-20T15:55:20Z +issue,81692,Failed to static link latest cuDNN while compiling,https://github.com/pytorch/pytorch/issues/81692,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cudnn",scan when time allows,2023-03-18T12:32:46Z +issue,96908,Doing inplace on a inplace view of tensor that retains_grad triggers internal assert,https://github.com/pytorch/pytorch/issues/96908,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-03-17T21:03:44Z +issue,38624,Wheels not manylinux1 compliant,https://github.com/pytorch/pytorch/issues/38624,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2023-03-17T13:37:15Z +issue,94333,cpu log1p for bfloat16 gives wrong result.,https://github.com/pytorch/pytorch/issues/94333,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cpu",scan when time allows,2023-03-17T02:45:40Z +issue,96769,`sparse.mm` triggers INTERNAL ASSERT FAILED when backwarding,https://github.com/pytorch/pytorch/issues/96769,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-03-16T16:12:24Z +issue,70555,[feature request] Exponential moving average (EMA) of a tensor across a dimension,https://github.com/pytorch/pytorch/issues/70555,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-03-16T09:02:24Z +issue,96789,`cumprod` triggers INTERNAL ASSERT FAILED when `out` is a tensor on cuda but input is on cpu,https://github.com/pytorch/pytorch/issues/96789,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: error checking",scan when time allows,2023-03-16T01:01:37Z +issue,95776,`torch.Tensor.is_set_to` raises `NotImplementedError` when inputs contain sparse tensor,https://github.com/pytorch/pytorch/issues/95776,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-03-15T16:12:34Z +issue,76689,Performance bad on ARM AArch64 for PyTorch C++,https://github.com/pytorch/pytorch/issues/76689,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",scan when time allows,2023-03-15T13:15:04Z +issue,91760,Inconsistent rank among torch.distributed primitives,https://github.com/pytorch/pytorch/issues/91760,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-03-15T10:28:08Z +issue,95622,"Testing InvokeAI 2.3.1.post1, using mps, with PyTorch nightly dev20230226 yields RuntimeError cross-device copies are not allowed!)",https://github.com/pytorch/pytorch/issues/95622,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2023-03-15T02:15:43Z +issue,95756,`torch.nanmedian` return a negative value when input is empty,https://github.com/pytorch/pytorch/issues/95756,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2023-03-15T00:55:43Z +issue,94779,Better Numpy API (interoperability between ML frameworks),https://github.com/pytorch/pytorch/issues/94779,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-03-14T21:57:09Z +issue,96704,`logical_xx` operations trigger INTERNAL ASSERT FAIL when `input` is complex tensor on cuda and `other` is on cpu,https://github.com/pytorch/pytorch/issues/96704,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2023-03-14T19:18:38Z +issue,96585,Proposal: Disable GC in test suite; GC after every test case,https://github.com/pytorch/pytorch/issues/96585,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: devx, module: flaky-tests, module: infra, module: testing",scan when time allows,2023-03-14T19:12:52Z +issue,96305,"Reuse autograd.grad graph for rapid, repeated gradient calculation",https://github.com/pytorch/pytorch/issues/96305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda graphs",scan when time allows,2023-03-14T13:17:17Z +issue,31461,Make RRef.to_here() non-blocking,https://github.com/pytorch/pytorch/issues/31461,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2023-03-14T10:16:58Z +issue,90998,"Crash in `index_select` with singleton `self`, non-singleton `index`",https://github.com/pytorch/pytorch/issues/90998,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-14T08:11:00Z +issue,96036,Questions and Possible Features: Pytorch RPC 'future.wait()' will not release GIL which will block other thread's execution when using multithreading.,https://github.com/pytorch/pytorch/issues/96036,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2023-03-14T04:31:54Z +issue,96615,Tensor Permutation Along Given Axis,https://github.com/pytorch/pytorch/issues/96615,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-03-13T22:37:18Z +issue,96629,Dataloader should kill & restart workers when timeout is hit,https://github.com/pytorch/pytorch/issues/96629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-03-13T22:35:10Z +issue,83509,Build and Run QNNPACK on X86,https://github.com/pytorch/pytorch/issues/83509,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-03-13T14:56:55Z +issue,95572,Support datatype argument for torch.distributed.all_gather() (And the whole distributed module),https://github.com/pytorch/pytorch/issues/95572,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: nccl",scan when time allows,2023-03-13T10:55:47Z +issue,94441,`get_debug_state` a script function causes INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/94441,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2023-03-13T05:03:39Z +issue,96523,"MPS Backend Doc, model = YourFavoriteNet() not defined",https://github.com/pytorch/pytorch/issues/96523,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: mps",scan when time allows,2023-03-12T12:43:43Z +issue,74901,"torch.fx.wrap will not work, when encapsulate the code",https://github.com/pytorch/pytorch/issues/74901,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-03-11T07:25:59Z +issue,93121,[RFC] Flop counters in PyTorch,https://github.com/pytorch/pytorch/issues/93121,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-03-10T22:59:28Z +issue,57776,[rfc] Build a debug tool to help users find which parameters are unused during DDP training,https://github.com/pytorch/pytorch/issues/57776,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-03-10T21:51:41Z +issue,96494,The sign of torch.distributions.transforms.PowerTransform seems to be incorrect,https://github.com/pytorch/pytorch/issues/96494,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-03-10T21:04:58Z +issue,69520,race condition of agent's TCP store on exit_barrier(),https://github.com/pytorch/pytorch/issues/69520,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: r2p,scan when time allows,2023-03-10T16:28:03Z +issue,96518,fft should ignore dims with shape 1,https://github.com/pytorch/pytorch/issues/96518,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2023-03-10T16:18:21Z +issue,96447,Information about CPU in `collect_env` is too verbose,https://github.com/pytorch/pytorch/issues/96447,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: collect_env.py,scan when time allows,2023-03-10T16:13:19Z +issue,96428,Compressed sparse constructor allows mixed `int32/int64` indices which leads to dtype promotion/demotion in conversions.,https://github.com/pytorch/pytorch/issues/96428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-03-10T16:12:00Z +issue,96412,Proxy Options for Pytorch Hub,https://github.com/pytorch/pytorch/issues/96412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2023-03-10T16:11:55Z +issue,95953,The torch.sparse document's typo error,https://github.com/pytorch/pytorch/issues/95953,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sparse",scan when time allows,2023-03-09T17:40:29Z +issue,94675,"`UnsupportedOperatorError`, `OnnxExporterError` and `SymbolicValueError` related to MultiheadAttention export to onnx with torch.jit.script",https://github.com/pytorch/pytorch/issues/94675,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2023-03-09T17:37:24Z +issue,58812,"Transferring tensor to the gpu and converting dtype in a single call to .to() is slower than first transferring, then converting.",https://github.com/pytorch/pytorch/issues/58812,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2023-03-09T13:47:04Z +issue,96236,nn.interpolate scale_factor floors output size with floating,https://github.com/pytorch/pytorch/issues/96236,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nn",scan when time allows,2023-03-09T12:02:19Z +issue,96318,Support managed memory backed dlpack with torch.from_dlpack,https://github.com/pytorch/pytorch/issues/96318,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dlpack,scan when time allows,2023-03-09T12:01:35Z +issue,95815,Let Nested Tensor Metadata be cached on GPU,https://github.com/pytorch/pytorch/issues/95815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-03-09T00:05:45Z +issue,49370,"PyEval_SaveThread: the function must be called with the GIL held, but the GIL is released",https://github.com/pytorch/pytorch/issues/49370,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2023-03-08T22:19:18Z +issue,94395,`func.jacrev()` should be implemented as `func.jacfwd().mT.contiguous()`,https://github.com/pytorch/pytorch/issues/94395,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: functorch",scan when time allows,2023-03-08T20:47:41Z +issue,96337,pytreeify decorators,https://github.com/pytorch/pytorch/issues/96337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: pytree",scan when time allows,2023-03-08T20:41:54Z +issue,94593,Memory corruptions can be triggered in torch._remove_batch_dim,https://github.com/pytorch/pytorch/issues/94593,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vmap,scan when time allows,2023-03-08T20:39:29Z +issue,95872,corrupted size vs prev size error,https://github.com/pytorch/pytorch/issues/95872,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ddp,scan when time allows,2023-03-08T18:25:35Z +issue,96140,mkldnn matmul kernel may be slower than openblas kernel for very small tensor shapes,https://github.com/pytorch/pytorch/issues/96140,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, module: performance",scan when time allows,2023-03-08T17:26:27Z +issue,89675,[ONNX] torch.onnx.export snapshots the grads as constants in onnx when op is in cuda device,https://github.com/pytorch/pytorch/issues/89675,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2023-03-08T13:54:11Z +issue,89634,[ONNX] torch.onnx.export can not export the grad of conv when the op is in CPU,https://github.com/pytorch/pytorch/issues/89634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2023-03-08T13:54:04Z +issue,44551,PyTorch 1.6 DataParallel causes CUDNN_STATUS_BAD_PARAM in backward pass,https://github.com/pytorch/pytorch/issues/44551,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: data parallel",scan when time allows,2023-03-08T11:24:00Z +issue,95207,"new backend privateuseone with ""to"" op",https://github.com/pytorch/pytorch/issues/95207,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2023-03-07T18:47:19Z +issue,78248,"USE_NATIVE_ARCH flag causes nvcc build failure due to ""'arch=native': expected a number""",https://github.com/pytorch/pytorch/issues/78248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-03-07T18:35:02Z +issue,96187,`gradgradcheck` does not work with sparse inputs.,https://github.com/pytorch/pytorch/issues/96187,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2023-03-07T18:13:06Z +issue,70920,`unique` should be split into four partial functions,https://github.com/pytorch/pytorch/issues/70920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2023-03-07T09:32:58Z +issue,70914,support setting `step` in `arange` without setting `end`,https://github.com/pytorch/pytorch/issues/70914,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: tensor creation",scan when time allows,2023-03-07T09:32:01Z +issue,61649,Move `tensordot` into `torch.linalg`,https://github.com/pytorch/pytorch/issues/61649,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: python array api",scan when time allows,2023-03-07T09:30:58Z +issue,58742,Python Array API New Operators Tracker,https://github.com/pytorch/pytorch/issues/58742,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2023-03-07T09:28:34Z +issue,70910,`eye` should support other diagonals than the main one,https://github.com/pytorch/pytorch/issues/70910,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: tensor creation",scan when time allows,2023-03-07T09:20:17Z +issue,63935,Inconsistent behaviour of index put when assigning from other type,https://github.com/pytorch/pytorch/issues/63935,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: type promotion",scan when time allows,2023-03-07T08:50:49Z +issue,55454,Include the cuda headers in the libtorch pre-built packages,https://github.com/pytorch/pytorch/issues/55454,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build, module: cuda",scan when time allows,2023-03-07T02:48:56Z +issue,96136,`torch.utils.checkpoint` should avoid updating BatchNorm statistics twice,https://github.com/pytorch/pytorch/issues/96136,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2023-03-07T02:10:27Z +issue,71683,EMA optimizer: class-form and function-form (using new foreach_lerp) - can be used for explicit robust updates of BatchNorm stats,https://github.com/pytorch/pytorch/issues/71683,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer",scan when time allows,2023-03-06T22:16:27Z +issue,95944,training hangs at line torch.cuda.synchronize(),https://github.com/pytorch/pytorch/issues/95944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: deadlock",scan when time allows,2023-03-06T15:50:19Z +issue,95956,[vulkan] missing aten::reflection_pad1d.out operator,https://github.com/pytorch/pytorch/issues/95956,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2023-03-06T15:45:46Z +issue,90827,Support for Pylint,https://github.com/pytorch/pytorch/issues/90827,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2023-03-06T15:40:13Z +issue,88893,RuntimeError: derivative for aten::mps_max_pool2d_backward is not implemented,https://github.com/pytorch/pytorch/issues/88893,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2023-03-06T03:43:14Z +issue,79851,[bazel] [ci] `//:module_test` CUDA error: CUDA driver version is insufficient for CUDA runtime version,https://github.com/pytorch/pytorch/issues/79851,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: cuda",scan when time allows,2023-03-03T18:12:06Z +issue,70870,Add flag for functional.Jacobian to return output as well,https://github.com/pytorch/pytorch/issues/70870,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-03-03T14:09:31Z +issue,76522,`torch.clamp` does not distribute gradients as element-wise`min/max` do,https://github.com/pytorch/pytorch/issues/76522,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-03-03T11:16:22Z +issue,95718,functorch.compile.memory_efficient_fusion errors with: RuntimeError: forward() Expected a value of type 'Tensor (inferred)' for argument 'primals_356' but instead found type 'int'.,https://github.com/pytorch/pytorch/issues/95718,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-03-03T07:10:43Z +issue,95917,torchvision Caltech101 collate_fn error,https://github.com/pytorch/pytorch/issues/95917,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-03T00:00:40Z +issue,27522,[feature request] Reduction (torch.add / torch.logaddexp / torch.max / torch.min / torch.mean) of several tensors without extra copies/allocations / memory accesses } TensorList inputs support,https://github.com/pytorch/pytorch/issues/27522,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-01T18:29:00Z +issue,95786,Static asserts on accessor templates,https://github.com/pytorch/pytorch/issues/95786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2023-03-01T16:38:16Z +issue,95768,Inconsistent behaviour of torch.all(),https://github.com/pytorch/pytorch/issues/95768,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-03-01T16:34:39Z +issue,95696,Can only import torch after Tensorflow accessed its gpu device,https://github.com/pytorch/pytorch/issues/95696,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-03-01T16:33:51Z +issue,95648,torch needs to SHOW that it support sm_89 even if functionally the same as sm_86,https://github.com/pytorch/pytorch/issues/95648,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-03-01T16:32:13Z +issue,95779,SymInt'ify _gather_sparse_backward,https://github.com/pytorch/pytorch/issues/95779,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-03-01T13:14:37Z +issue,50201,Excessive memory usage caused by Samplers storing lists of indices,https://github.com/pytorch/pytorch/issues/50201,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: memory usage",scan when time allows,2023-02-28T21:50:08Z +issue,95244,Make this ridiculously long error message more user friendly,https://github.com/pytorch/pytorch/issues/95244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2023-02-28T19:42:55Z +issue,94806,ImportError: cannot import name 'Backend' from 'torch._C._distributed_c10d' (unknown location),https://github.com/pytorch/pytorch/issues/94806,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-02-28T19:34:23Z +issue,95309,`torch.distributed.Store` triggers INTERNAL ASSER FAILED when seting,https://github.com/pytorch/pytorch/issues/95309,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-02-28T19:31:13Z +issue,95434,It seems that `torch.Tensor.addmv` and `torch.Tensor.addr` will check some inputs' dtype if and only if in `backward()`,https://github.com/pytorch/pytorch/issues/95434,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex, module: linear algebra, module: type promotion",scan when time allows,2023-02-28T17:09:19Z +issue,95485,Enrich shape operations with nested tensors,https://github.com/pytorch/pytorch/issues/95485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-02-28T03:19:36Z +issue,94704,`where` triggers INTERNAL ASSERT FAILED when `out` is a long tensor due to mixed types,https://github.com/pytorch/pytorch/issues/94704,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: type promotion",scan when time allows,2023-02-27T18:15:01Z +issue,95590,Confusing error messages from `torch.nn.LazyLinear` in different versions.,https://github.com/pytorch/pytorch/issues/95590,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: lazy",scan when time allows,2023-02-27T16:34:18Z +issue,95548,"torch.distributions.kumaraswamy.Kumaraswamy generates samples outside its support (0,1)",https://github.com/pytorch/pytorch/issues/95548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: nans and infs",scan when time allows,2023-02-27T16:27:44Z +issue,95463,`add/add_` for CSC: errors when trying to access non-existent `crow_indices`.,https://github.com/pytorch/pytorch/issues/95463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-02-27T16:25:57Z +issue,95462,Extend docs - Fixing out of memory with python garbage collection,https://github.com/pytorch/pytorch/issues/95462,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: memory usage",scan when time allows,2023-02-27T16:23:16Z +issue,95403,test_ddp_apply_optim_in_backward in distributed_test.py fails for gloo backend,https://github.com/pytorch/pytorch/issues/95403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2023-02-27T16:18:14Z +issue,95290,Continuous dropout layer,https://github.com/pytorch/pytorch/issues/95290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-02-27T12:19:26Z +issue,91653,Stochastic Illegal Memory Access error mid-epoch on AWS p4d instances,https://github.com/pytorch/pytorch/issues/91653,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2023-02-26T22:18:46Z +issue,85621,Gloo DDP SocketTimeout error on Windows,https://github.com/pytorch/pytorch/issues/85621,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2023-02-26T12:36:06Z +issue,51112,"torch.tensor(x) fails inconsistently, assumes 0 index exists in 'Series'",https://github.com/pytorch/pytorch/issues/51112,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2023-02-26T05:16:33Z +issue,95562,test_layer_norm_backward and test_layer_norm_backward_5d run OOM in slow gradcheck,https://github.com/pytorch/pytorch/issues/95562,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-02-26T01:33:46Z +issue,62350,NotImplementedError: Could not run 'torchvision::nms' with arguments from the 'CUDA' backend.,https://github.com/pytorch/pytorch/issues/62350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: vision",scan when time allows,2023-02-24T00:41:05Z +issue,95276,tabulate is used by `torch.fx.graph_module.GraphModule.print_tabular` but is not installed when installing pytorch,https://github.com/pytorch/pytorch/issues/95276,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-02-23T22:49:48Z +issue,89421,fx.wrap is ignored with make_fx proxy tensor tracer,https://github.com/pytorch/pytorch/issues/89421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-02-23T14:22:04Z +issue,95129,Default value of `validate_args` is set to `True` when passed as `None` in `Multinomial`,https://github.com/pytorch/pytorch/issues/95129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-02-23T01:57:40Z +issue,95058,Upsampling ResBlock GPU memory spike,https://github.com/pytorch/pytorch/issues/95058,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2023-02-22T23:03:08Z +issue,95122,CosineAnnealingWarmRestarts but restarts are becoming more frequent,https://github.com/pytorch/pytorch/issues/95122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2023-02-22T23:01:45Z +issue,95124,"`INTERNAL ASSERT FAILED` -When using the PyTorch docker environment released by pytorch, a Vulcan support issue occurs",https://github.com/pytorch/pytorch/issues/95124,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docker",scan when time allows,2023-02-22T22:50:17Z +issue,95135,Add local version identifier to wheel file names,https://github.com/pytorch/pytorch/issues/95135,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-02-22T22:48:24Z +issue,95146,Whether to consider native support for intel gpu？,https://github.com/pytorch/pytorch/issues/95146,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: intel,scan when time allows,2023-02-22T22:45:19Z +issue,95237,test_foreach failing cuda memory leak check,https://github.com/pytorch/pytorch/issues/95237,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: mta",scan when time allows,2023-02-22T22:34:38Z +issue,95077,Implement a `torch.cuda.visible_device_indexes` function.,https://github.com/pytorch/pytorch/issues/95077,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-02-22T22:34:14Z +issue,95304,`torch.cartesian_prod` returns inconsistent dimensions with only one input,https://github.com/pytorch/pytorch/issues/95304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2023-02-22T22:33:46Z +issue,95169,COO @ COO tries to allocate way too much memory on CUDA,https://github.com/pytorch/pytorch/issues/95169,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2023-02-22T22:33:46Z +issue,32293,Truncated normal distribution,https://github.com/pytorch/pytorch/issues/32293,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-02-21T20:22:31Z +issue,79091,Error with Named Tensors and multiple threads,https://github.com/pytorch/pytorch/issues/79091,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2023-02-21T16:56:35Z +issue,71156,Slow backward for matrix multiplication of two sparse COO tensors on CPU,https://github.com/pytorch/pytorch/issues/71156,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-02-20T16:10:14Z +issue,94801,Compiling PyTorch from Source on Xavier,https://github.com/pytorch/pytorch/issues/94801,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vectorization",scan when time allows,2023-02-19T23:59:39Z +issue,50940,SequentialSampler getting slower as time passing by,https://github.com/pytorch/pytorch/issues/50940,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-02-16T19:35:01Z +issue,52625,Add slicing to distributions,https://github.com/pytorch/pytorch/issues/52625,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-02-16T17:34:56Z +issue,81899,Dependency header directory is not properly expanded in the utils.cpp_extention in ninja mode,https://github.com/pytorch/pytorch/issues/81899,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2023-02-16T12:34:25Z +issue,93275,Bug in torch.linalg.svd,https://github.com/pytorch/pytorch/issues/93275,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mkl,scan when time allows,2023-02-16T01:04:43Z +issue,94457,LSTM on CPU is significantly slower on PyTorch compared to other frameworks,https://github.com/pytorch/pytorch/issues/94457,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2023-02-15T23:02:12Z +issue,18630,documentation for C++ / libtorch autograd profiler,https://github.com/pytorch/pytorch/issues/18630,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2023-02-15T22:19:40Z +issue,94827,Linking error with Libtorch,https://github.com/pytorch/pytorch/issues/94827,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: mkl",scan when time allows,2023-02-15T16:15:01Z +issue,91536,The speed of matrix inversion is relatively slow for many small matrices,https://github.com/pytorch/pytorch/issues/91536,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",scan when time allows,2023-02-15T00:59:38Z +issue,87353,Speed of torch.istft,https://github.com/pytorch/pytorch/issues/87353,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: performance",scan when time allows,2023-02-14T03:45:04Z +issue,94654,Missing FX documents for some modules,https://github.com/pytorch/pytorch/issues/94654,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-02-14T03:19:57Z +issue,94705,Split getitem OpInfo into dynamic and non-dynamic inputs,https://github.com/pytorch/pytorch/issues/94705,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2023-02-14T03:06:53Z +issue,94739,Update PyTorch's default C standard to C17 from C11,https://github.com/pytorch/pytorch/issues/94739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-02-14T02:59:25Z +issue,94591,Issue with `upsample_nearest2d` decomposition,https://github.com/pytorch/pytorch/issues/94591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: decompositions,scan when time allows,2023-02-14T02:58:51Z +issue,44931,Issue while writing scalars o tensorboard using writer.add_scalars(....),https://github.com/pytorch/pytorch/issues/44931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-02-13T19:30:05Z +issue,94590,A Segment Fault can be triggered in torch.affine_grid_generator,https://github.com/pytorch/pytorch/issues/94590,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-02-13T17:46:59Z +issue,94614,Jetson CI needs Updates,https://github.com/pytorch/pytorch/issues/94614,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: jetson,scan when time allows,2023-02-13T17:43:58Z +issue,94131,faster WeightedRandomSampler implementation based on alias method,https://github.com/pytorch/pytorch/issues/94131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-02-13T15:43:33Z +issue,94388,Inconsistent results when using torch.Tensor.bernoulli with float instead of Tensor probabilities,https://github.com/pytorch/pytorch/issues/94388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: distributions, module: random",scan when time allows,2023-02-10T14:42:31Z +issue,68323,sparse.mm: CUDA error: internal error when calling `cusparseSpGEMM_workEstimation [...]`,https://github.com/pytorch/pytorch/issues/68323,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2023-02-10T10:44:28Z +issue,71678,torch.bmm backward with sparse input,https://github.com/pytorch/pytorch/issues/71678,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-02-10T09:23:01Z +issue,27588,RuntimeError: cuDNN error: CUDNN_STATUS_MAPPING_ERROR,https://github.com/pytorch/pytorch/issues/27588,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2023-02-10T05:07:09Z +issue,62255,Support `torch.linalg.trace`,https://github.com/pytorch/pytorch/issues/62255,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2023-02-09T22:55:35Z +issue,52668,Independent axis support for torch.trace,https://github.com/pytorch/pytorch/issues/52668,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-02-09T22:52:19Z +issue,94451,"`jacrev` raise ""Cannot access storage of TensorWrapper"" error when computing the grad of `storage`",https://github.com/pytorch/pytorch/issues/94451,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: functorch",scan when time allows,2023-02-09T17:09:49Z +issue,94474,pybind11 SymNode binding is a footgun py::cast,https://github.com/pytorch/pytorch/issues/94474,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2023-02-09T15:22:03Z +issue,94504,[mypy] skipping mypy for a few torch/fx and torch/_subclass files,https://github.com/pytorch/pytorch/issues/94504,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2023-02-09T15:11:30Z +issue,94454,Document and promise reproducibility torch.randn / torch.rand / torch.randint family behavior on CPU devices,https://github.com/pytorch/pytorch/issues/94454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2023-02-09T14:48:35Z +issue,94443,A better error msg for `cuda.jiterator` when input is on `cpu`,https://github.com/pytorch/pytorch/issues/94443,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: jiterator,scan when time allows,2023-02-09T13:06:01Z +issue,94208,`zeros_like` + `fill_` makes the gradient computation in forward mode fail,https://github.com/pytorch/pytorch/issues/94208,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2023-02-09T13:04:15Z +issue,94161,JIT: Dropout fails codegen on the third forward passes,https://github.com/pytorch/pytorch/issues/94161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2023-02-09T13:02:28Z +issue,94132,Asking for a LAZYMODULEMIXIN warning,https://github.com/pytorch/pytorch/issues/94132,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lazy, module: molly-guard, module: nn",scan when time allows,2023-02-09T13:01:51Z +issue,94085,`unfold` fails in forward mode when unfolding a scalar tensor,https://github.com/pytorch/pytorch/issues/94085,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2023-02-09T12:59:32Z +issue,49095,NCCL Error when training with 2x 3090s.,https://github.com/pytorch/pytorch/issues/49095,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multi-gpu",scan when time allows,2023-02-09T03:35:52Z +issue,70669,CUSOLVER_STATUS_EXECUTION_FAILED when using the torch.logdet(),https://github.com/pytorch/pytorch/issues/70669,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2023-02-09T01:55:06Z +issue,62630,Instructions for CMAKE_PREFIX_PATH seem to be broken,https://github.com/pytorch/pytorch/issues/62630,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2023-02-09T01:13:58Z +issue,68041,[RFC] APEX style fused optimizers in PyTorch,https://github.com/pytorch/pytorch/issues/68041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mta, module: optimizer, module: performance",scan when time allows,2023-02-08T23:03:03Z +issue,92260,distributions.Beta returning incorrect results at 0 and 1,https://github.com/pytorch/pytorch/issues/92260,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-02-08T22:08:41Z +issue,94238,`PyTorchFileWriter` should drop the GIL while writing files,https://github.com/pytorch/pytorch/issues/94238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-02-08T19:47:11Z +issue,94304,"RuntimeError: p.block != nullptr && p.block->ptr != nullptr INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":1275, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/94304,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: cudacachingallocator",scan when time allows,2023-02-08T19:46:07Z +issue,94374,[fx] const_fold.split_const_subgraphs leads to UserWarning,https://github.com/pytorch/pytorch/issues/94374,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-02-08T19:25:44Z +issue,49825,Ensure inplace views are properly handled,https://github.com/pytorch/pytorch/issues/49825,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2023-02-08T16:26:51Z +issue,72408,Matrix multiplication is 30 times slower for integers than floats on CPU,https://github.com/pytorch/pytorch/issues/72408,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: linear algebra, module: performance",scan when time allows,2023-02-08T09:17:13Z +issue,69063,make: Makefile: No such file or directory,https://github.com/pytorch/pytorch/issues/69063,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-02-08T02:10:25Z +issue,61281,DataLoader with IterativeDataset throws an error when providing a BatchSampler,https://github.com/pytorch/pytorch/issues/61281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2023-02-07T23:41:12Z +issue,93240,USE_CUDNN=1 doesn't force cmake to fail if cudnn is not found,https://github.com/pytorch/pytorch/issues/93240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: build warnings, module: cuda",scan when time allows,2023-02-07T22:44:19Z +issue,94003,test_nccl_warn_not_in_group_debug_detail is flaky,https://github.com/pytorch/pytorch/issues/94003,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, oncall: distributed",scan when time allows,2023-02-07T17:17:25Z +issue,85157,NestedTensor 2.0 issue tracking,https://github.com/pytorch/pytorch/issues/85157,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2023-02-07T16:12:57Z +issue,94185,Addition of CSC/BSR/BSC tensors raises RuntimeError exceptions,https://github.com/pytorch/pytorch/issues/94185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-02-07T15:03:35Z +issue,50036,channels_last format convolution is slower than normal NCHW,https://github.com/pytorch/pytorch/issues/50036,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory format, module: performance",scan when time allows,2023-02-07T14:56:51Z +issue,17745,"distributed data parallel, gloo backend works, but nccl deadlock",https://github.com/pytorch/pytorch/issues/17745,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: nccl, oncall: distributed",scan when time allows,2023-02-07T06:22:45Z +issue,94021,Set AVX2 is minimum supported instruction set for Linux X86,https://github.com/pytorch/pytorch/issues/94021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: intel,scan when time allows,2023-02-06T18:24:26Z +issue,93944,error: no member named 'residual_with_sum_zero_point' in 'ideep::attr_t,https://github.com/pytorch/pytorch/issues/93944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2023-02-06T15:00:57Z +issue,93347,"when I want to use a new backend, how to deal with the op with 'device' argument?",https://github.com/pytorch/pytorch/issues/93347,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2023-02-06T14:44:50Z +issue,90547,Unable to link LibTorch against CUDA and CUDNN statically,https://github.com/pytorch/pytorch/issues/90547,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: static linking,scan when time allows,2023-02-06T09:43:32Z +issue,93955,"Enable Link Time Optimization in PyTorch 2.0 Release Binaries - Smaller, Faster, Better Binaries",https://github.com/pytorch/pytorch/issues/93955,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build, module: performance, oncall: releng",scan when time allows,2023-02-05T20:26:57Z +issue,93523,[functorch] [vmap] [SymInt][fake tensor],https://github.com/pytorch/pytorch/issues/93523,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-02-03T17:29:20Z +issue,38487,expected scalar type Half but found Float with torch.cuda.amp and torch.nn.DataParallel,https://github.com/pytorch/pytorch/issues/38487,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: data parallel",scan when time allows,2023-02-03T12:35:00Z +issue,92884,"[RFC] Make more operations inplace (GELU, BatchNorm, LayerNorm)",https://github.com/pytorch/pytorch/issues/92884,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-02-02T23:52:57Z +issue,62395,[FX] Does PyTorch want to support for dynamic control flow for torch.fx?,https://github.com/pytorch/pytorch/issues/62395,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-02-02T22:50:34Z +issue,93319,Minifier should not use pickle to save state into minifier launcher,https://github.com/pytorch/pytorch/issues/93319,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: fx",scan when time allows,2023-02-01T15:09:19Z +issue,93227,squeezenet1_1 fails accuracy with AMP (but not on CI and dashboard); minifier does not work (when not using cuDNN?),https://github.com/pytorch/pytorch/issues/93227,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2023-02-01T02:40:32Z +issue,93579,torchdynamo.export doesn't work with data-dependent control flow,https://github.com/pytorch/pytorch/issues/93579,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-02-01T01:06:34Z +issue,93458,minified code can not produce fp64_ref result,https://github.com/pytorch/pytorch/issues/93458,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-02-01T01:02:09Z +issue,93446,[Inductor] [CPU] Vectorization not supporting python pass-in scalar double in speech_transformer,https://github.com/pytorch/pytorch/issues/93446,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-02-01T01:01:55Z +issue,93444,[Inductor] [CPU] Maxpooling is not vectorized in shufflenet_v2_x1_0,https://github.com/pytorch/pytorch/issues/93444,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-02-01T01:01:53Z +issue,62168,Error installing from source,https://github.com/pytorch/pytorch/issues/62168,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-01-31T22:06:02Z +issue,93231,"Minifier produces minifier script that doesn't fail accuracy on Background_Matting (dynamic shapes, inductor, inference)",https://github.com/pytorch/pytorch/issues/93231,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: fx",scan when time allows,2023-01-31T21:39:24Z +issue,93317,Minifier doesn't save/load functorch config,https://github.com/pytorch/pytorch/issues/93317,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functorch, module: fx",scan when time allows,2023-01-31T21:16:10Z +issue,62998,Add nn.Module full pre-backward hooks,https://github.com/pytorch/pytorch/issues/62998,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-01-31T19:10:57Z +issue,89597,Caching a model's weights and state_dict to disk to save RAM,https://github.com/pytorch/pytorch/issues/89597,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-01-31T16:38:15Z +issue,93362,Make torch.testing functions overrideable with torch_function?,https://github.com/pytorch/pytorch/issues/93362,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: testing",scan when time allows,2023-01-31T16:25:39Z +issue,72880,Bazel fails in an obscure way if submodules are not initialized,https://github.com/pytorch/pytorch/issues/72880,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: third_party",scan when time allows,2023-01-31T11:14:42Z +issue,91553,Segmentation fault when running torch.nn.AdaptiveMaxPool2d,https://github.com/pytorch/pytorch/issues/91553,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-01-30T21:14:00Z +issue,91552,Overflow when running torch.nn.AdaptiveMaxPool2d,https://github.com/pytorch/pytorch/issues/91552,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: pooling",scan when time allows,2023-01-30T21:13:38Z +issue,91940,torchrun elastic always “address already in use” error,https://github.com/pytorch/pytorch/issues/91940,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: r2p",scan when time allows,2023-01-30T02:41:25Z +issue,91697,Build from source fails: undefined reference to caffe2::DeviceQuery,https://github.com/pytorch/pytorch/issues/91697,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-01-29T10:20:21Z +issue,91943,elastic job failed when scale down,https://github.com/pytorch/pytorch/issues/91943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: r2p",scan when time allows,2023-01-28T06:40:30Z +issue,90793,nn.CrossEntropyLoss error out when the sample size is large,https://github.com/pytorch/pytorch/issues/90793,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: edge cases, module: memory usage, module: nn",scan when time allows,2023-01-27T17:29:46Z +issue,92778,Process get killed when running torch.normal,https://github.com/pytorch/pytorch/issues/92778,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-01-27T07:02:32Z +issue,93017,numpy v1.24 does not work with `writer.add_histogram`,https://github.com/pytorch/pytorch/issues/93017,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-01-26T23:48:17Z +issue,91325,Support for saving multiple storages/tensors that view same data as different dtypes,https://github.com/pytorch/pytorch/issues/91325,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2023-01-26T19:26:11Z +issue,24081,Multi-gpu example freeze and is not killable,https://github.com/pytorch/pytorch/issues/24081,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: data parallel, module: deadlock, module: dependency bug",scan when time allows,2023-01-26T17:16:55Z +issue,53937,Only tensor can be proxied in builtin operators with `__torch_function__`,https://github.com/pytorch/pytorch/issues/53937,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: fx",scan when time allows,2023-01-26T13:23:47Z +issue,92967,Tracking issue for segfaults and floating point exceptions on 1.12.0,https://github.com/pytorch/pytorch/issues/92967,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-26T06:52:07Z +issue,84309,Unable to run a single convolutional layer in different CUDA-contexts,https://github.com/pytorch/pytorch/issues/84309,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-01-25T21:26:07Z +issue,66482,BAR1 memory of GPU is not released when main process is killed.,https://github.com/pytorch/pytorch/issues/66482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader, module: memory usage",scan when time allows,2023-01-25T17:19:39Z +issue,83388,"zero-numel tensor has ""RuntimeError: strides[cur - 1] == sizes[cur] * strides[cur] INTERNAL ASSERT FAILED"" in multi-thread.",https://github.com/pytorch/pytorch/issues/83388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, oncall: jit",scan when time allows,2023-01-25T16:13:34Z +issue,92907,Unable to find an engine to execute when using pip to install but not with conda,https://github.com/pytorch/pytorch/issues/92907,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: convolution, module: cudnn",scan when time allows,2023-01-25T16:06:09Z +issue,83376,Complex-Valued Gaussian distributions,https://github.com/pytorch/pytorch/issues/83376,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: distributions, module: random",scan when time allows,2023-01-25T11:52:16Z +issue,92920,Add Stride Argument For Constructors,https://github.com/pytorch/pytorch/issues/92920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2023-01-25T07:30:37Z +issue,92781,Floating point exception when running torch.nn.AdaptiveMaxPool3d,https://github.com/pytorch/pytorch/issues/92781,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-25T07:24:21Z +issue,57423,internal compiler error compiling PyTorch master on Fedora 33,https://github.com/pytorch/pytorch/issues/57423,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-01-25T06:56:33Z +issue,92835,Double free when running torch.linalg.ldl_solve,https://github.com/pytorch/pytorch/issues/92835,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-01-25T06:27:01Z +issue,92828,segfault when running torch.igamma,https://github.com/pytorch/pytorch/issues/92828,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-25T06:14:40Z +issue,35071,Options for printing the shape with print(tensor),https://github.com/pytorch/pytorch/issues/35071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: printing,scan when time allows,2023-01-25T01:56:59Z +issue,63855,Include tensor shape in its default print formatting,https://github.com/pytorch/pytorch/issues/63855,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: printing,scan when time allows,2023-01-25T01:54:57Z +issue,71210,axis to dim remapping is not working for flip and roll,https://github.com/pytorch/pytorch/issues/71210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2023-01-24T21:36:09Z +issue,92801,"[BE] move _apply_to_tensors from FSDP to torch.distributed.utils, use in _recursive_to",https://github.com/pytorch/pytorch/issues/92801,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2023-01-24T18:09:45Z +issue,78880,fatal_signal_asan_no_sig_test in current master hang.,https://github.com/pytorch/pytorch/issues/78880,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: deadlock, module: testing",scan when time allows,2023-01-24T14:06:50Z +issue,72393,Some system-installed headers are mistakenly used.,https://github.com/pytorch/pytorch/issues/72393,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-01-24T13:58:26Z +issue,92252,"`model.to(""cuda:0"")` does not release all CPU memory",https://github.com/pytorch/pytorch/issues/92252,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2023-01-23T21:58:35Z +issue,91556,Overflow when running torch.nn.AdaptiveMaxPool3d on torch 1.12.0 and 1.13.1,https://github.com/pytorch/pytorch/issues/91556,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-01-23T17:17:51Z +issue,92811,[complex] Jacobian of a non-holomorphic complex valued function,https://github.com/pytorch/pytorch/issues/92811,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-23T15:49:17Z +issue,28733,CPU MaxPool2d is very slow,https://github.com/pytorch/pytorch/issues/28733,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: cpu, module: performance, module: pooling",scan when time allows,2023-01-23T02:08:41Z +issue,92683,Torchscript troubles with complex values. RuntimeError: isInt() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/92683,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, oncall: jit",scan when time allows,2023-01-21T03:39:03Z +issue,92542,Pytorch AMP performance issue.,https://github.com/pytorch/pytorch/issues/92542,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: memory format",scan when time allows,2023-01-20T01:50:37Z +issue,92594,INTERNAL ASSERT FAILED when mixed dtypes for `addcmul_`,https://github.com/pytorch/pytorch/issues/92594,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: type promotion",scan when time allows,2023-01-20T01:32:14Z +issue,92173,"""multi device"" tests get skipped in standard CI",https://github.com/pytorch/pytorch/issues/92173,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",scan when time allows,2023-01-19T22:32:32Z +issue,89160,Got many TestDTensorOpsCUDA.test_dtensor_op_db_X test failures,https://github.com/pytorch/pytorch/issues/89160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-01-19T16:22:43Z +issue,49038,Make the forward function of a nn.Module and/or a certain function to work with only float32 when used with autocast,https://github.com/pytorch/pytorch/issues/49038,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: nn",scan when time allows,2023-01-19T01:14:44Z +issue,92223,Improve make_fx tracing speed,https://github.com/pytorch/pytorch/issues/92223,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: proxytensor",scan when time allows,2023-01-18T15:11:13Z +issue,92217,"false INTERNAL ASSERT FAILED at ""../c10/cuda/CUDAGraphsC10Utils.h"":73, please report a bug to PyTorch. Unknown CUDA graph CaptureStatus32680",https://github.com/pytorch/pytorch/issues/92217,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2023-01-18T13:49:37Z +issue,92350,TypeError: no implementation found for 'torch._ops.aten.max.default' on types that implement __torch_dispatch__: [],https://github.com/pytorch/pytorch/issues/92350,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2023-01-18T12:48:39Z +issue,92259,[discussion] Fused MLPs,https://github.com/pytorch/pytorch/issues/92259,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: inductor, oncall: pt2",scan when time allows,2023-01-18T12:48:20Z +issue,71301,Why AMP make backward speed more slow?,https://github.com/pytorch/pytorch/issues/71301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2023-01-18T12:47:26Z +issue,64535,Memory Leak in MKL OpenMP on AVX2 machine,https://github.com/pytorch/pytorch/issues/64535,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: intel, module: memory usage, module: mkl",scan when time allows,2023-01-18T08:12:33Z +issue,91810,[Bug/functorch] Cannot use `tensor.detach().numpy()` for `GradTrackingTensor`: Cannot access data pointer of Tensor that doesn't have storage,https://github.com/pytorch/pytorch/issues/91810,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-01-17T20:23:32Z +issue,91856,torchrun --help is too slow,https://github.com/pytorch/pytorch/issues/91856,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: r2p",scan when time allows,2023-01-17T18:15:49Z +issue,91855,torchrun default value of command line options,https://github.com/pytorch/pytorch/issues/91855,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: distributed, oncall: r2p",scan when time allows,2023-01-17T18:15:30Z +issue,91889,ToTensor deadlock in subprocess,https://github.com/pytorch/pytorch/issues/91889,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2023-01-16T10:36:11Z +issue,92086,Error while building pytorch mobile binaries from source,https://github.com/pytorch/pytorch/issues/92086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,scan when time allows,2023-01-16T09:37:45Z +issue,46374,Complex Number support for torch.nn modules,https://github.com/pytorch/pytorch/issues/46374,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nn",scan when time allows,2023-01-15T20:01:33Z +issue,85656,"[functorch] conv.{1, 2, 3}d should raise errors",https://github.com/pytorch/pytorch/issues/85656,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2023-01-15T08:28:01Z +issue,89105,Bad string in GLSL shader,https://github.com/pytorch/pytorch/issues/89105,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2023-01-15T04:06:03Z +issue,92148,Occassional OverflowError with mps running yolov7,https://github.com/pytorch/pytorch/issues/92148,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2023-01-13T18:34:59Z +issue,90147,[Feature Proposal] Extend torch hub to better support cloud serving and edge deployment,https://github.com/pytorch/pytorch/issues/90147,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2023-01-13T11:33:50Z +issue,91716,linspace (and arange) behaves differently on GPU and CPU,https://github.com/pytorch/pytorch/issues/91716,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-13T08:11:00Z +issue,53103,ModuleList not recognised as reversible by mypy,https://github.com/pytorch/pytorch/issues/53103,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: typing",scan when time allows,2023-01-12T23:03:08Z +issue,92075,"Hessian produces wrong results, but works if I add a perturbation",https://github.com/pytorch/pytorch/issues/92075,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-12T20:31:20Z +issue,92072,Implement forward AD with grid_sampler_2d,https://github.com/pytorch/pytorch/issues/92072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2023-01-12T19:33:23Z +issue,73563,Pytorch Installation from source fails,https://github.com/pytorch/pytorch/issues/73563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2023-01-12T01:44:07Z +issue,56862,Non-symbolic FX tracer,https://github.com/pytorch/pytorch/issues/56862,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2023-01-11T14:26:01Z +issue,58736,type promotion with 0d-tensors diverges from array API specification,https://github.com/pytorch/pytorch/issues/58736,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: type promotion",scan when time allows,2023-01-11T08:28:12Z +issue,91958,CUDA error `CUBLAS_STATUS_NOT_INITIALIZED`,https://github.com/pytorch/pytorch/issues/91958,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2023-01-11T06:42:18Z +issue,91809,Better API for `torch.cov` (and `Tensor.cov`),https://github.com/pytorch/pytorch/issues/91809,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2023-01-10T14:20:16Z +issue,91842,Adding a page for subfolder/subfile overview/descriptions in the developer wiki,https://github.com/pytorch/pytorch/issues/91842,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2023-01-10T14:10:14Z +issue,78489,[1.9.1] [collect_env] collect_env does not collect actual runtime-loaded cudnn version,https://github.com/pytorch/pytorch/issues/78489,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: collect_env.py, module: cudnn",scan when time allows,2023-01-09T20:42:42Z +issue,91887,Only the first logged trace in a given log dir is visible in tensorboard.,https://github.com/pytorch/pytorch/issues/91887,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2023-01-09T18:59:13Z +issue,91300,[🚀 Feature Request] pdf and sampling from Alpha-stable distribution,https://github.com/pytorch/pytorch/issues/91300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-01-09T18:18:54Z +issue,82076,Autocast documentation examples would break,https://github.com/pytorch/pytorch/issues/82076,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: docs",scan when time allows,2023-01-08T17:11:35Z +issue,91759,"Error while building pytorch from source on windows - Ninja Build Stopped, Subcommand Failed",https://github.com/pytorch/pytorch/issues/91759,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-06T04:23:04Z +issue,91678,Wrong in building torch from source,https://github.com/pytorch/pytorch/issues/91678,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-06T04:20:44Z +issue,91624,trainer,https://github.com/pytorch/pytorch/issues/91624,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2023-01-06T04:18:27Z +issue,91751,[bazel] replace //c10:headers dependency by //c10 dependency,https://github.com/pytorch/pytorch/issues/91751,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bazel,scan when time allows,2023-01-06T03:29:36Z +issue,87861,RuntimeError: unable to mmap 29764 bytes from file : Cannot allocate memory (12),https://github.com/pytorch/pytorch/issues/87861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2023-01-05T13:22:31Z +issue,91615,"sparse.mm(coo, dense) produces wrong results on T4/V100 GPUs",https://github.com/pytorch/pytorch/issues/91615,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-01-04T18:19:10Z +issue,91593,Division by zero error when running torch.nn.functional.lp_pool1d,https://github.com/pytorch/pytorch/issues/91593,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-01-04T16:14:11Z +issue,91618,There is no developer documentation about getting started with MPS native debugging,https://github.com/pytorch/pytorch/issues/91618,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: mps",scan when time allows,2023-01-04T15:33:48Z +issue,91599,"`mul(CSC, CSC)` fails with layout mismatch between the inputs and the output.",https://github.com/pytorch/pytorch/issues/91599,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2023-01-04T15:28:14Z +issue,62672,Add Convolution support for lazy tensor,https://github.com/pytorch/pytorch/issues/62672,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: lazy",scan when time allows,2023-01-03T22:20:34Z +issue,68058,Some lr-schedulers docs seems to have typos/missing information,https://github.com/pytorch/pytorch/issues/68058,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: lrscheduler, module: optimizer",scan when time allows,2023-01-03T18:52:59Z +issue,91574,Add BlockWise Distribution Support to the torch.distributions Package,https://github.com/pytorch/pytorch/issues/91574,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2023-01-03T15:22:26Z +issue,91437,using Tensor subclass between vmap layers,https://github.com/pytorch/pytorch/issues/91437,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: functorch",scan when time allows,2023-01-03T15:21:18Z +issue,91447,"Degenerate ranges are allowed in NumPy, but not in PyTorch.",https://github.com/pytorch/pytorch/issues/91447,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2023-01-03T10:11:33Z +issue,91582,Softmax function slows down for data with large range,https://github.com/pytorch/pytorch/issues/91582,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2023-01-03T08:48:26Z +issue,91557,Segmentation fault when running torch.nn.AdaptiveMaxPool3d,https://github.com/pytorch/pytorch/issues/91557,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2023-01-03T02:04:16Z +issue,73175,[feature request] Support tensor count vector argument in torch.split,https://github.com/pytorch/pytorch/issues/73175,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2023-01-02T22:34:29Z +issue,71329,"[RFC] Implement array methods (extend, insert, pop, etc) for container classes",https://github.com/pytorch/pytorch/issues/71329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2023-01-02T06:10:27Z +issue,41327,torch.nn.parallel.scatter_gather.gather can't gather outputs that are dataclasses,https://github.com/pytorch/pytorch/issues/41327,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2022-12-31T01:24:15Z +issue,70394,Intel MKL FATAL ERROR: This system does not meet the minimum requirements for use of the Intel(R) Math Kernel Library.,https://github.com/pytorch/pytorch/issues/70394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: dependency bug, module: intel, module: macos",scan when time allows,2022-12-30T12:56:39Z +issue,91509,[functorch] make batch norm docs point to UX limitations,https://github.com/pytorch/pytorch/issues/91509,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2022-12-29T16:31:52Z +issue,91471,Clean up nt impl duplicates where one can,https://github.com/pytorch/pytorch/issues/91471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-12-28T22:22:34Z +issue,19117,[jit] Can't `torch.jit.script` a lambda,https://github.com/pytorch/pytorch/issues/19117,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-12-28T20:14:59Z +issue,90412,Importing numpy makes Tensor min max crash,https://github.com/pytorch/pytorch/issues/90412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2022-12-28T19:42:23Z +issue,91417,[bazel] error: use of undeclared identifier 'cudaGraphDebugDotPrint',https://github.com/pytorch/pytorch/issues/91417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2022-12-28T14:00:21Z +issue,89344,AdaptiveAvgPool1d failed in the lower version,https://github.com/pytorch/pytorch/issues/89344,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2022-12-28T13:29:14Z +issue,90481,Internal assert when ctx.saved_tensors fails when saving results of an intermediate view tensor with torch.utils.checkpoint and use_reentrant=False,https://github.com/pytorch/pytorch/issues/90481,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd",scan when time allows,2022-12-28T10:17:37Z +issue,91406,Pytorch clang-tidy header-filter is still broken,https://github.com/pytorch/pytorch/issues/91406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-12-28T05:02:28Z +issue,91374,[RFC] `quantile` should work for `float16`/`half` on the GPU,https://github.com/pytorch/pytorch/issues/91374,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half",scan when time allows,2022-12-27T23:27:32Z +issue,91375,Internal Assert failed,https://github.com/pytorch/pytorch/issues/91375,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cuda graphs",scan when time allows,2022-12-27T23:26:59Z +issue,91039,Simple deleting from the sys cache fails on reimport,https://github.com/pytorch/pytorch/issues/91039,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-12-26T20:12:58Z +issue,55437,Error building docker image: No module named 'typing_extensions',https://github.com/pytorch/pytorch/issues/55437,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docker",scan when time allows,2022-12-25T14:14:10Z +issue,52295,Printing a Tensor that is being vmap'ed over in C++ raises an error,https://github.com/pytorch/pytorch/issues/52295,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vmap,scan when time allows,2022-12-24T10:07:37Z +issue,38536,"build libtorch problem: Configuring incomplete, errors occurred! _mm256_abs_epi16",https://github.com/pytorch/pytorch/issues/38536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-12-24T08:59:38Z +issue,78475,`with torch.backends.cudnn.flags(deterministic=True)` doesn't give an exception for ctc_loss backward on CUDA,https://github.com/pytorch/pytorch/issues/78475,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: determinism",scan when time allows,2022-12-23T03:36:58Z +issue,91156,`quantile` fails for `float16`/`half` inputs,https://github.com/pytorch/pytorch/issues/91156,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: half",scan when time allows,2022-12-23T00:45:43Z +issue,91280,custom Function that supports functorch jvp doesn't work with in-place,https://github.com/pytorch/pytorch/issues/91280,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: forward ad, module: functorch",scan when time allows,2022-12-22T21:45:50Z +issue,89907,nn.CrossEntropy/nn.NLLLoss : Request for option to specify invalid ignore_index for perf. optimization,https://github.com/pytorch/pytorch/issues/89907,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2022-12-22T17:30:57Z +issue,91199,functorch.functionalize doesn't work with torch.autograd.grad,https://github.com/pytorch/pytorch/issues/91199,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2022-12-22T03:18:22Z +issue,46409,Custom ops get stuck in multiprocess data loader under certain environments,https://github.com/pytorch/pytorch/issues/46409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: dataloader, module: deadlock",scan when time allows,2022-12-21T23:01:01Z +issue,91244,first class dims leak memeory,https://github.com/pytorch/pytorch/issues/91244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2022-12-21T13:38:21Z +issue,70505,`UninitializedParameter.to(device='meta')` creates a zero-sized meta tensor instead of remaining uninitialized,https://github.com/pytorch/pytorch/issues/70505,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lazy, module: meta tensors",scan when time allows,2022-12-21T05:59:23Z +issue,91210,Odd/hand-wavy mathematical notation for Conv2D,https://github.com/pytorch/pytorch/issues/91210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-12-21T04:57:09Z +issue,89068,view_copy out= does not reshape zero element tensors,https://github.com/pytorch/pytorch/issues/89068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: viewing and reshaping",scan when time allows,2022-12-20T23:16:33Z +issue,90999,overflow (?) on cuda tensor after matrix multiplication,https://github.com/pytorch/pytorch/issues/90999,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability",scan when time allows,2022-12-20T18:01:39Z +issue,90526,valgrind failure `Conditional jump or move depends on uninitialised value(s)`,https://github.com/pytorch/pytorch/issues/90526,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,scan when time allows,2022-12-20T08:56:47Z +issue,90905,Retrieve Tensor from Tensor.data_ptr(),https://github.com/pytorch/pytorch/issues/90905,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-12-20T08:54:10Z +issue,90663,[Feature Request] An alternative sampling routine for Dirichlet to fix Dirichlet and Beta sampling bugs,https://github.com/pytorch/pytorch/issues/90663,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-12-17T12:15:18Z +issue,90992,as_strided_scatter : INTERNAL_ASSERT_FAILED for requires_grad=True and non-config input,https://github.com/pytorch/pytorch/issues/90992,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2022-12-16T15:31:12Z +issue,90555,[RFC] Add torch.backends.tbb.is_available(),https://github.com/pytorch/pytorch/issues/90555,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tbb,scan when time allows,2022-12-15T23:49:10Z +issue,90895,ModuleNotFoundError: No module named 'torch._C._distributed_c10d'; 'torch._C' is not a package,https://github.com/pytorch/pytorch/issues/90895,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-12-15T22:21:04Z +issue,90832,Inference Mode docs,https://github.com/pytorch/pytorch/issues/90832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-12-15T04:01:05Z +issue,68377,Misleading Error Message from nn.RNN when Passing Incorrect Data Type,https://github.com/pytorch/pytorch/issues/68377,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2022-12-15T00:40:10Z +issue,90708,PixelShuffle/Unshuffle Channels Last Support on GPU,https://github.com/pytorch/pytorch/issues/90708,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2022-12-15T00:37:49Z +issue,90751,pytorch prune in libtorch,https://github.com/pytorch/pytorch/issues/90751,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-12-13T16:10:16Z +issue,33354,Long torchscript warmup times can be problematic for production serving,https://github.com/pytorch/pytorch/issues/33354,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-12-13T10:23:19Z +issue,87448,Semantics of sparse operations clarification - Sparsity of the gradient with respect to a sparse tensor input,https://github.com/pytorch/pytorch/issues/87448,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs, module: masked operators, module: sparse",scan when time allows,2022-12-13T05:41:07Z +issue,87956,Autograd doesn't stop executing backward graph early enough in situations involving set_,https://github.com/pytorch/pytorch/issues/87956,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-12-13T05:38:55Z +issue,87755,Autograd precision for CONV + BN between pytorch version 1.11.0 and 1.12.1,https://github.com/pytorch/pytorch/issues/87755,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: numerical-reproducibility, module: numerical-stability",scan when time allows,2022-12-13T05:31:32Z +issue,90742,Adopt full_backward_pre_hook in DDP,https://github.com/pytorch/pytorch/issues/90742,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-12-13T03:12:00Z +issue,52040,torch.distributed.TCPStore doesn't work with dual IPv4/IPv6 network interface,https://github.com/pytorch/pytorch/issues/52040,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-12-13T00:08:11Z +issue,87992,torch.rand(...) is not consistent for large shape dimensions across GPUs (with the same random seed),https://github.com/pytorch/pytorch/issues/87992,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: random",scan when time allows,2022-12-12T11:02:00Z +issue,43250,Support for Multi-Categorical in torch.distributions,https://github.com/pytorch/pytorch/issues/43250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-12-11T12:08:02Z +issue,36706,Reduction for `torch.int8` is super slow on CUDA,https://github.com/pytorch/pytorch/issues/36706,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance, module: tensoriterator",scan when time allows,2022-12-10T15:36:16Z +issue,90398,IValue(c10::List) constructor is confusing and undocumented,https://github.com/pytorch/pytorch/issues/90398,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-12-09T19:35:39Z +issue,90447,Different behavior for complex numbers operations with numpy,https://github.com/pytorch/pytorch/issues/90447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs, module: numpy",scan when time allows,2022-12-09T16:22:48Z +issue,90485,Tensor indexing and slicing documentation should explicitly state that indexing follows numpy semantics and link to the numpy indexing documentation.,https://github.com/pytorch/pytorch/issues/90485,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy",scan when time allows,2022-12-09T16:09:19Z +issue,90057,[GradScaler] Inconsistent scale values across different GPUs caused by uneven inputs for AMP DDP training,https://github.com/pytorch/pytorch/issues/90057,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-12-09T12:35:04Z +issue,88265,build: failure when upgrade oneTBB to 2021.7.0,https://github.com/pytorch/pytorch/issues/88265,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: tbb",scan when time allows,2022-12-09T03:01:59Z +issue,90065,forward-mode AD formula for torch.add (and possibly others) accidentally upcasts float32 to float64,https://github.com/pytorch/pytorch/issues/90065,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, module: functorch, module: performance",scan when time allows,2022-12-08T21:34:01Z +issue,89408,torch.nn.MultiLabelMarginLoss has different performance on CPU and GPU.,https://github.com/pytorch/pytorch/issues/89408,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2022-12-08T13:21:26Z +issue,51720,Support for 64-bit (ILP64) LAPACK,https://github.com/pytorch/pytorch/issues/51720,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: linear algebra",scan when time allows,2022-12-08T13:07:12Z +issue,90395,"Cannot add target-level dependencies to non-existent target ""gloo_cuda"".",https://github.com/pytorch/pytorch/issues/90395,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-12-08T01:13:49Z +issue,90256,LibTorch static build from source missing libshm.so,https://github.com/pytorch/pytorch/issues/90256,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-12-07T02:47:52Z +issue,19349,Different behavior when trace model.,https://github.com/pytorch/pytorch/issues/19349,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-12-07T01:12:16Z +issue,71806,Migrate master to main: https://github.com/pytorch/pytorch,https://github.com/pytorch/pytorch/issues/71806,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-12-06T19:20:05Z +issue,62540,[Meta] Change default branch name to `main` for repos in `pytorch` project,https://github.com/pytorch/pytorch/issues/62540,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: infra",scan when time allows,2022-12-06T19:19:53Z +issue,90126,torch._dynamo.exc.BackendCompilerFailed: compile_fx raised TypeError: tqdm.__init__() got an unexpected keyword argument 'desc',https://github.com/pytorch/pytorch/issues/90126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2022-12-06T18:11:09Z +issue,90114,documentation need to be as per python version,https://github.com/pytorch/pytorch/issues/90114,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-12-06T18:00:03Z +issue,87090,MPS memory usage significantly higher than on CPU,https://github.com/pytorch/pytorch/issues/87090,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: mps",scan when time allows,2022-12-06T12:04:52Z +issue,68916,Crash loading fairseq based model,https://github.com/pytorch/pytorch/issues/68916,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2022-12-05T23:37:07Z +issue,84944,test_warp_softmax_64bit_indexing_cuda_float16 takes ~147GB of CPU memory and is very slow,https://github.com/pytorch/pytorch/issues/84944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: testing",scan when time allows,2022-12-05T20:02:34Z +issue,47908,The speed of pytorch with cudatoolkit 11.0 is slower than cudatoolkit 10.2,https://github.com/pytorch/pytorch/issues/47908,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2022-12-05T17:21:30Z +issue,76105,Failed to build on Ubuntu 18.04 due to bad MPI linker flags,https://github.com/pytorch/pytorch/issues/76105,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-12-05T15:48:10Z +issue,89942,Random K compression hook in PyTorch DDP,https://github.com/pytorch/pytorch/issues/89942,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ddp,scan when time allows,2022-12-05T04:32:41Z +issue,76069,`init_process_group` hanging on HPC multi-node system w GPU,https://github.com/pytorch/pytorch/issues/76069,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-12-04T16:47:37Z +issue,62661,Rewrite PyTorch CUDA backend in Triton,https://github.com/pytorch/pytorch/issues/62661,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-12-02T21:11:17Z +issue,85485,"`max_pool2d_with_indices(self, ...)` shouldn't need to save `self` for backward",https://github.com/pytorch/pytorch/issues/85485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-12-02T15:47:39Z +issue,61167,Torch .pth model export to ONNX returned empty graph,https://github.com/pytorch/pytorch/issues/61167,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2022-12-02T11:13:15Z +issue,55967,"Cannot init, destroy, and then re-init process groups",https://github.com/pytorch/pytorch/issues/55967,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-12-02T00:01:18Z +issue,90000,"Can not use x=torch.tensor(b), to create a Tensor out of a List[List[Tensor]] (A List of Lists of Tensors)",https://github.com/pytorch/pytorch/issues/90000,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-12-01T22:28:59Z +issue,89609,NVFuser failing masked.{amax|amin|sum} extremal and correctness tests,https://github.com/pytorch/pytorch/issues/89609,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-12-01T08:37:16Z +issue,72138,[RFC] PyTorch Sharder for distributed training,https://github.com/pytorch/pytorch/issues/72138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-11-30T17:36:03Z +issue,26781,Matrix corresponding to convolution by a 2D kernel (convmtx2),https://github.com/pytorch/pytorch/issues/26781,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-11-30T07:19:24Z +issue,89686,MultiProcess tests fail when run on nodes with 1 GPU,https://github.com/pytorch/pytorch/issues/89686,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-11-29T16:46:49Z +issue,89054,Add a `device` keyword argument to `torch.manual_seed`,https://github.com/pytorch/pytorch/issues/89054,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2022-11-29T00:30:23Z +issue,85300,torchrun substitutes host names for IP addresses,https://github.com/pytorch/pytorch/issues/85300,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: distributed",scan when time allows,2022-11-28T22:42:34Z +issue,82479,Cloning conjugate tensor in torch_dispatch context produces non equality.,https://github.com/pytorch/pytorch/issues/82479,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: complex",scan when time allows,2022-11-28T18:52:21Z +issue,88942,`torch.svd_lowrank` Trigger RuntimeError under UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88942,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sanitizers, module: sparse",scan when time allows,2022-11-28T18:11:44Z +issue,88941,`torch.linalg.lstsq` Trigger RuntimeError under UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88941,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: linear algebra, module: sanitizers",scan when time allows,2022-11-28T18:10:02Z +issue,82534,Use NestedTensor in RNN models,https://github.com/pytorch/pytorch/issues/82534,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-11-28T18:01:03Z +issue,50718,dependency_links is deprecated,https://github.com/pytorch/pytorch/issues/50718,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2022-11-28T16:11:08Z +issue,89684,`positive_semidefinite` constraint fails on CUDA 11.7,https://github.com/pytorch/pytorch/issues/89684,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: linear algebra",scan when time allows,2022-11-28T15:17:13Z +issue,69202,CPU only mode with cudatoolkit 11.3,https://github.com/pytorch/pytorch/issues/69202,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-11-28T12:27:01Z +issue,73050,aten::batch_norm schema does not mention mutation of running_mean/running_var,https://github.com/pytorch/pytorch/issues/73050,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-11-27T21:51:54Z +issue,68405,"Windows multi machine multi card training, card initialization in GLOO communication.",https://github.com/pytorch/pytorch/issues/68405,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: windows, oncall: distributed",scan when time allows,2022-11-25T15:59:35Z +issue,50471,linker error when trying to use Metal backend in PyTorch Mobile,https://github.com/pytorch/pytorch/issues/50471,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",scan when time allows,2022-11-24T07:04:16Z +issue,88880,Add alphatensor support for faster matrix multiplication?,https://github.com/pytorch/pytorch/issues/88880,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-11-24T04:35:39Z +issue,89601,Building PyTorch with Vulkan backend fails (1.13 and master),https://github.com/pytorch/pytorch/issues/89601,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2022-11-24T00:08:37Z +issue,89574,Finish deprecation of autograd decorator over class objects,https://github.com/pytorch/pytorch/issues/89574,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-11-23T16:51:04Z +issue,88805,[feature request] Get/set fastmath CPU bit (and some other FPU flags?),https://github.com/pytorch/pytorch/issues/88805,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: numerical-stability",scan when time allows,2022-11-23T05:19:21Z +issue,10604,"Bilinear interpolation behavior inconsistent with TF, CoreML and Caffe",https://github.com/pytorch/pytorch/issues/10604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2022-11-22T08:47:18Z +issue,89416,"When the torch.masked_select operator passes in the same parameters, it behaves differently on CPU and GPU.",https://github.com/pytorch/pytorch/issues/89416,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: masked operators,scan when time allows,2022-11-22T08:32:09Z +issue,89275,"torch.addcdiv: input, tensor1, and tensor2 parameters should be of the same type",https://github.com/pytorch/pytorch/issues/89275,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: docs",scan when time allows,2022-11-22T01:57:50Z +issue,89125,binary_cross_entropy/bce_with_logits (+ other loss functions) for nested_tensor,https://github.com/pytorch/pytorch/issues/89125,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nestedtensor",scan when time allows,2022-11-21T18:02:22Z +issue,89320,`masked_fill` with `FloatTensor` mask will never mask but fails silently.,https://github.com/pytorch/pytorch/issues/89320,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: masked operators",scan when time allows,2022-11-21T15:35:54Z +issue,89301,Meta implementation for copy_ is wrong,https://github.com/pytorch/pytorch/issues/89301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2022-11-21T15:33:54Z +issue,89293,fbgemm_avx512 build failure,https://github.com/pytorch/pytorch/issues/89293,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-11-21T15:32:02Z +issue,51450,torch.median slower than torch.sort on cpu,https://github.com/pytorch/pytorch/issues/51450,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: reductions, module: sorting and selection",scan when time allows,2022-11-20T13:02:41Z +issue,75480,IndexError: Caught IndexError in replica 0 on device 0.,https://github.com/pytorch/pytorch/issues/75480,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-11-19T18:15:39Z +issue,89245,sm_80 support,https://github.com/pytorch/pytorch/issues/89245,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-11-19T05:17:55Z +issue,89124,Zero-copy way to make flat tensor into a nested_tensor given a shape,https://github.com/pytorch/pytorch/issues/89124,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-11-18T18:57:03Z +issue,86888,Broadcasting add for nested_tensor,https://github.com/pytorch/pytorch/issues/86888,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-11-18T18:55:56Z +issue,89218,"`torch.nn.LayerNorm` Abort with ""invalid device ordinal"" Error",https://github.com/pytorch/pytorch/issues/89218,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn, module: norms and normalization",scan when time allows,2022-11-18T08:22:04Z +issue,83313,torch.Tag doesn't have accurate mypy info,https://github.com/pytorch/pytorch/issues/83313,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2022-11-17T21:13:44Z +issue,88563,`nn.functional.embedding_bag` Trigger out-of-bound Read under Compute Sanitizer,https://github.com/pytorch/pytorch/issues/88563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-11-17T08:23:51Z +issue,88992,Incorrect version in the instructions on official website,https://github.com/pytorch/pytorch/issues/88992,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-11-16T16:43:29Z +issue,89080,Unable to backprop through dense weighted sum of sparse_coo_tensors,https://github.com/pytorch/pytorch/issues/89080,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-11-16T14:35:44Z +issue,7806,OOM Exception when using torch.nn.grad.conv2d_weight (apparently because CuDNN backwards is not used),https://github.com/pytorch/pytorch/issues/7806,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: memory usage, module: performance",scan when time allows,2022-11-16T09:58:16Z +issue,88949,`torch.set_rng_state` Trigger RuntimeError under UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88949,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,scan when time allows,2022-11-16T04:24:56Z +issue,88950,`torch.nn.functional.embedding_bag` Trigger RuntimeError under UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88950,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,scan when time allows,2022-11-16T04:12:05Z +issue,88931,INTERNAL ASSERT FAILED. Missing scalar type infromation.,https://github.com/pytorch/pytorch/issues/88931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-11-15T23:36:20Z +issue,88991,nvprims.native_batch_norm doesn't support fake tensor inputs,https://github.com/pytorch/pytorch/issues/88991,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-11-15T23:18:24Z +issue,88980,Glog macro redefinition problem when including headers from both libtorch and glog,https://github.com/pytorch/pytorch/issues/88980,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-11-15T23:18:20Z +issue,88947,`torch.Tensor.msort` Trigger RuntimeError under UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,scan when time allows,2022-11-15T23:16:13Z +issue,88945,`torch.linalg.eigvals` Trigger RuntimeError under UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88945,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,scan when time allows,2022-11-15T23:16:06Z +issue,88944,`torch.topk` Trigger RuntimError under UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88944,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,scan when time allows,2022-11-15T23:15:58Z +issue,89060,Extend test_proxy_tensor tests to support ops test non floating point types,https://github.com/pytorch/pytorch/issues/89060,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: proxytensor, module: tests",scan when time allows,2022-11-15T23:07:26Z +issue,86131,Debuggability++: Share instructions for building exotic CI configurations,https://github.com/pytorch/pytorch/issues/86131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-11-15T19:59:13Z +issue,88410,benchmark cache persist,https://github.com/pytorch/pytorch/issues/88410,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2022-11-15T19:24:54Z +issue,65698,torch.onnx.export error,https://github.com/pytorch/pytorch/issues/65698,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2022-11-15T16:07:30Z +issue,65718,Migrate current Windows CI scripts off of batch,https://github.com/pytorch/pytorch/issues/65718,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: windows",scan when time allows,2022-11-15T03:59:49Z +issue,86340,Add complex support for SparseAdam and LBFGS optimizers,https://github.com/pytorch/pytorch/issues/86340,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: optimizer",scan when time allows,2022-11-14T23:54:45Z +issue,73003,pin_memory hangs instead of throwing,https://github.com/pytorch/pytorch/issues/73003,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader",scan when time allows,2022-11-14T18:47:46Z +issue,88883,Investigate why `test_aot_autograd_symbolic_exhaustive_masked_median_cpu_float32` is flaky,https://github.com/pytorch/pytorch/issues/88883,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests,scan when time allows,2022-11-14T16:55:23Z +issue,88658,"nn.Linear allocate too many space which lead to CPUAllocator ""allocate memory failure"" if it's BF16. good for FP32.",https://github.com/pytorch/pytorch/issues/88658,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2022-11-14T12:54:22Z +issue,2001,Implement similar PyTorch function as model.summary() in keras?,https://github.com/pytorch/pytorch/issues/2001,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-11-14T10:19:06Z +issue,88565,What causes CPU to degrade when I load the weight with torch.hub.load(),https://github.com/pytorch/pytorch/issues/88565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: hub",scan when time allows,2022-11-14T08:25:20Z +issue,87957,VS2022Preview ParallelCommon.cpp.obj : fatal error LNK1161: invalid export specification,https://github.com/pytorch/pytorch/issues/87957,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: static linking",scan when time allows,2022-11-14T04:48:16Z +issue,62178,Wrong implementation of method log_prob in torch.distributions.negative_binomial,https://github.com/pytorch/pytorch/issues/62178,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs, module: numpy",scan when time allows,2022-11-13T17:59:39Z +issue,87019,Rewrite `narrow_copy_dense_cpu_out` using `copy_` and `narrow`,https://github.com/pytorch/pytorch/issues/87019,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2022-11-12T15:13:30Z +issue,88902,MPS test_numpy_ref_mps_nn_functional_group_norm_mps_float32 is flaky?,https://github.com/pytorch/pytorch/issues/88902,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: mps",scan when time allows,2022-11-11T21:14:55Z +issue,88691,forward AD for _euclidean_dist,https://github.com/pytorch/pytorch/issues/88691,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: forward ad, module: functorch",scan when time allows,2022-11-11T18:21:29Z +issue,45115,Standardized Distributions,https://github.com/pytorch/pytorch/issues/45115,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-11-10T21:01:45Z +issue,88612,Bernoulli uses legacy contiguous memory format,https://github.com/pytorch/pytorch/issues/88612,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2022-11-10T17:19:16Z +issue,88643,pack_sequence() always fail after set_default_tensor_type to CUDA,https://github.com/pytorch/pytorch/issues/88643,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2022-11-10T17:18:34Z +issue,88639,CUDA unknown error after suspend during debugging,https://github.com/pytorch/pytorch/issues/88639,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-11-10T17:17:58Z +issue,88598,"Despite having aten::diag_embed.out, torch.diag_embed doesn't support out= argument",https://github.com/pytorch/pytorch/issues/88598,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-11-10T17:16:07Z +issue,88591,`pack_padded_sequence` not compatible with deterministic mode it calls `torch.scatter`,https://github.com/pytorch/pytorch/issues/88591,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: rnn",scan when time allows,2022-11-10T17:15:38Z +issue,88614,The libtorch test TestScalarTensor.TestScalarTensorMPS fails on Apple Silicon,https://github.com/pytorch/pytorch/issues/88614,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: m1, module: mps",scan when time allows,2022-11-10T17:14:39Z +issue,88616,The libtorch test SequentialTest.ModuleForwardMethodOptionalArg fails on Apple Silicon,https://github.com/pytorch/pytorch/issues/88616,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: m1",scan when time allows,2022-11-10T17:13:19Z +issue,88474,Mixed precision training fails due to NaN in batch norm running_mean,https://github.com/pytorch/pytorch/issues/88474,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-11-08T06:36:19Z +issue,88309,ASAN shard 4 started to OOM after unrelated commit,https://github.com/pytorch/pytorch/issues/88309,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-11-07T23:15:33Z +issue,88002,Einsum Optimization Tracker,https://github.com/pytorch/pytorch/issues/88002,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2022-11-07T16:38:40Z +issue,39009,Missing OneCycleLR and MultiplicativeLR in lr_scheduler.pyi,https://github.com/pytorch/pytorch/issues/39009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2022-11-07T04:31:48Z +issue,88142,Refactor `torch.return_types.topk` to behave like a `namedtuple` or a `dict`,https://github.com/pytorch/pytorch/issues/88142,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2022-11-06T19:55:05Z +issue,88389,Whether to support libtorch source code compilation of C++11 ？,https://github.com/pytorch/pytorch/issues/88389,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-11-06T14:38:23Z +issue,82219,[feature request] DataLoader to accept num_threads argument to auto-set number of threads for OpenMP / intra-op parallelism,https://github.com/pytorch/pytorch/issues/82219,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: openmp",scan when time allows,2022-11-04T12:52:26Z +issue,61110,Incorrect exponential calculation on Jetson devices with float32 dtype,https://github.com/pytorch/pytorch/issues/61110,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: cuda, module: jetson",scan when time allows,2022-11-04T09:07:05Z +issue,88423,view_as_real and split_with_sizes links in Tensor Views docs are broken,https://github.com/pytorch/pytorch/issues/88423,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-11-03T18:18:14Z +issue,88408,Unit test with `--subprocess` command doesn't respect the `-k` filter flag and runs all available sub tests,https://github.com/pytorch/pytorch/issues/88408,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: testing, module: tests",scan when time allows,2022-11-03T18:12:30Z +issue,25767,"Model parallel with DDP get `Socket Timeout` error when using NCCL, while GLOO works fine",https://github.com/pytorch/pytorch/issues/25767,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-11-03T08:56:44Z +issue,88194,Add a config option to raise errors instead of warnings in nvFuser integration,https://github.com/pytorch/pytorch/issues/88194,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-11-02T16:50:59Z +issue,88004,link error happen when intergrate libtorch to other tool,https://github.com/pytorch/pytorch/issues/88004,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2022-11-02T16:09:19Z +issue,53261,Dimension argument names in torch.diag_embed/diagonal vs. transpose/transpose_,https://github.com/pytorch/pytorch/issues/53261,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy",scan when time allows,2022-11-02T14:28:55Z +issue,50274,torch.equal is divergent from np.equal,https://github.com/pytorch/pytorch/issues/50274,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy, module: testing",scan when time allows,2022-11-02T13:32:34Z +issue,88304,AttributeError: module 'tensorboard.compat.tensorflow_stub.io.gfile' has no attribute 'MakeDirs',https://github.com/pytorch/pytorch/issues/88304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2022-11-02T13:18:15Z +issue,86529,[NvFuser] JIT model with `mul+atan+sgn` will access illegal memory on cuda when computing gradient,https://github.com/pytorch/pytorch/issues/86529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-11-02T01:08:12Z +issue,88192,[docs] torch.is_neg/torch.Tensor.is_neg not documented,https://github.com/pytorch/pytorch/issues/88192,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-11-01T19:46:05Z +issue,86644,"[NvFuser] INTERNAL ASSERT FAIL ""ScalarType should be static for Tensors in fusion for amp optimization""",https://github.com/pytorch/pytorch/issues/86644,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: nvfuser",scan when time allows,2022-11-01T17:15:57Z +issue,88147,Conv2d is not deterministic when input tensor has different strides,https://github.com/pytorch/pytorch/issues/88147,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2022-11-01T15:30:36Z +issue,88191,`torch.nn.RReLU` not reporting `lower > upper` on CUDA,https://github.com/pytorch/pytorch/issues/88191,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-11-01T14:11:03Z +issue,88189,Moving tensor to GPU by .cuda() gets stucked when AMD Secure Encripted Virtualization (SEV) is activated,https://github.com/pytorch/pytorch/issues/88189,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: rocm",scan when time allows,2022-11-01T14:10:12Z +issue,88185,`torch.mm` Trigger RuntimeError with UndefinedBehaviorSanitizer,https://github.com/pytorch/pytorch/issues/88185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sanitizers,scan when time allows,2022-11-01T14:08:25Z +issue,87841,`max_unpool3d` will trigger an assertion fail under compute sanitizer,https://github.com/pytorch/pytorch/issues/87841,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: pooling, module: sanitizers",scan when time allows,2022-11-01T13:49:35Z +issue,87955,AttributeError: '_MultiProcessingDataLoaderIter' object has no attribute 'next',https://github.com/pytorch/pytorch/issues/87955,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-11-01T13:42:41Z +issue,88062,Cannot import `traverse_dps` from torch.data.utils.graph,https://github.com/pytorch/pytorch/issues/88062,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2022-11-01T13:39:38Z +issue,87902,Permute,https://github.com/pytorch/pytorch/issues/87902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-11-01T13:38:02Z +issue,87979,amp with `bf16`: backward happens in `f16` when using `@torch.cuda.amp.custom_bwd`,https://github.com/pytorch/pytorch/issues/87979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: bfloat16, module: half",scan when time allows,2022-10-31T23:32:55Z +issue,88081,out of memory with pytorch version after 1.8.1,https://github.com/pytorch/pytorch/issues/88081,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator, module: memory usage",scan when time allows,2022-10-31T21:46:14Z +issue,88027,"getting error error: namespace ""cub"" has no member ""Debug"" when try to build v1.8.2 with CUDA 11.6",https://github.com/pytorch/pytorch/issues/88027,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-10-31T19:20:33Z +issue,88053,Different behaviour in sparse matmul,https://github.com/pytorch/pytorch/issues/88053,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-10-31T19:12:08Z +issue,84234,Randomness should be consistent across devices with use_deterministic_algorithms,https://github.com/pytorch/pytorch/issues/84234,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: random",scan when time allows,2022-10-31T18:26:04Z +issue,88103,ProcessGroupNCCL watchdog can't catch NCCL comm initialization issues,https://github.com/pytorch/pytorch/issues/88103,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-10-31T17:56:55Z +issue,87706,C++ Adagrad optimizer doesn't initialize parameter state,https://github.com/pytorch/pytorch/issues/87706,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: optimizer",scan when time allows,2022-10-31T14:28:43Z +issue,62917,Use `linecache.lazycache` in torch.fx,https://github.com/pytorch/pytorch/issues/62917,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: fx",scan when time allows,2022-10-29T21:00:33Z +issue,18229,Add support for tuple type deduction in C++ custom operators,https://github.com/pytorch/pytorch/issues/18229,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: internals",scan when time allows,2022-10-29T14:11:39Z +issue,82259,[Reproducibility] Make tests say when unusual environment variables are set that change behavior of the test,https://github.com/pytorch/pytorch/issues/82259,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-10-28T21:56:45Z +issue,83081,[CI] Split up periodic.yml into forward-fixable.yml and periodic.yml,https://github.com/pytorch/pytorch/issues/83081,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-10-28T21:03:06Z +issue,39245,pytorch-crf model to onnx conversion,https://github.com/pytorch/pytorch/issues/39245,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2022-10-28T01:43:46Z +issue,87163,New APIs for cuda graph inspection and manipulation,https://github.com/pytorch/pytorch/issues/87163,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2022-10-27T19:39:06Z +issue,87539,RAM leak when copying tensor from cpu to cuda,https://github.com/pytorch/pytorch/issues/87539,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2022-10-27T19:04:42Z +issue,87642,`libtorch_cpu.so` is exposing some LLVM symbols,https://github.com/pytorch/pytorch/issues/87642,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions",scan when time allows,2022-10-27T14:40:00Z +issue,86449,torch.cuda.empty_cache() is not working,https://github.com/pytorch/pytorch/issues/86449,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-10-27T14:14:42Z +issue,39716,Do not modify global random state,https://github.com/pytorch/pytorch/issues/39716,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: random",scan when time allows,2022-10-27T02:52:13Z +issue,19126,weight_norm doesn't support eta and returns nan for zero weights,https://github.com/pytorch/pytorch/issues/19126,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: nn, module: norms and normalization",scan when time allows,2022-10-26T23:54:12Z +issue,87555,Improve `c10d::ReduceOp` & `torch.distributed.distributed_c10d.ReduceOp`,https://github.com/pytorch/pytorch/issues/87555,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-10-26T23:43:19Z +issue,87388,Document dist.new_subgroups,https://github.com/pytorch/pytorch/issues/87388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: distributed",scan when time allows,2022-10-26T17:15:48Z +issue,87634,Add tests for ProcessGroup cpp extensions,https://github.com/pytorch/pytorch/issues/87634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-10-26T16:32:55Z +issue,17798,Feature Request: deterministic CUDA torch.nn.CTCLoss,https://github.com/pytorch/pytorch/issues/17798,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: loss, module: nn",scan when time allows,2022-10-26T12:10:37Z +issue,87701,pytorch/pytorch cpu official Docker images,https://github.com/pytorch/pytorch/issues/87701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2022-10-25T20:55:10Z +issue,87589,"#error ""Expected GLOO_USE_CUDA to be defined""",https://github.com/pytorch/pytorch/issues/87589,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2022-10-25T02:13:38Z +issue,87571,Testing insufficient to catch incorrect dispatch key for bernoulli.p re functionalization,https://github.com/pytorch/pytorch/issues/87571,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2022-10-24T22:15:08Z +issue,68169,torch.histogram: Sum of PDFs is num of bins instead of 1,https://github.com/pytorch/pytorch/issues/68169,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sorting and selection",scan when time allows,2022-10-24T21:16:59Z +issue,87451,Some operations do not keep `channels_last` memory format which yields accuracy drop,https://github.com/pytorch/pytorch/issues/87451,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2022-10-24T19:08:09Z +issue,86020,Functorch memory_efficient_fusion gives wrong output batch size,https://github.com/pytorch/pytorch/issues/86020,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: functorch",scan when time allows,2022-10-24T17:25:20Z +issue,87267,The installation commands given on the pytorch website will not install properly,https://github.com/pytorch/pytorch/issues/87267,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-10-24T13:08:56Z +issue,87494,Missing docstring for resize_as,https://github.com/pytorch/pytorch/issues/87494,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-10-23T19:26:37Z +issue,87551,`bmm` will return wrong result on cpu with in-place,https://github.com/pytorch/pytorch/issues/87551,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-10-23T19:16:31Z +issue,87170,torch.linalg.cond gives inconsistent results on CPU/CUDA,https://github.com/pytorch/pytorch/issues/87170,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: nans and infs",scan when time allows,2022-10-21T15:19:33Z +issue,82687,Move nested-tensor tutorial from prototype,https://github.com/pytorch/pytorch/issues/82687,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-10-21T14:53:33Z +issue,87389,Installing PyTorch with BUILD_SPLIT_CUDA=ON and CUDNN fails on linker error,https://github.com/pytorch/pytorch/issues/87389,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: cudnn",scan when time allows,2022-10-21T14:36:01Z +issue,87371,Better type annotations for `torch.Tensor` subclasses,https://github.com/pytorch/pytorch/issues/87371,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2022-10-21T13:45:29Z +issue,87276,torch.save throws ValueError: ctypes objects containing pointers cannot be pickled,https://github.com/pytorch/pytorch/issues/87276,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2022-10-21T13:04:28Z +issue,87352,RuntimeError: Tensors of type TensorImpl do not have numel,https://github.com/pytorch/pytorch/issues/87352,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: docker",scan when time allows,2022-10-21T13:03:01Z +issue,28743,Sampler for IterableDataset,https://github.com/pytorch/pytorch/issues/28743,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-10-21T09:05:24Z +issue,84336,torch 1.12.1 cuda 10.2 runs slower than torch 1.8.2 cuda 10.2,https://github.com/pytorch/pytorch/issues/84336,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: performance",scan when time allows,2022-10-20T18:27:23Z +issue,87222,`custom_jvp` and `custom_vjp`,https://github.com/pytorch/pytorch/issues/87222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2022-10-20T15:06:08Z +issue,84415,Deepcopy of FX graph fails with nested make_fx and constant tensors,https://github.com/pytorch/pytorch/issues/84415,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-10-20T13:06:19Z +issue,74539,Making PyTorch more pythonic: the PyTorch functions should be inspectable.,https://github.com/pytorch/pytorch/issues/74539,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-10-20T04:44:09Z +issue,76613,MeanVarianceNormalization,https://github.com/pytorch/pytorch/issues/76613,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2022-10-19T23:11:28Z +issue,87041,"Segmentation fault: 11 when running ""import torch"" on Mac OS X",https://github.com/pytorch/pytorch/issues/87041,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: macos,scan when time allows,2022-10-19T18:25:55Z +issue,86265,TORCH_WARN is executed just once per set of parameters,https://github.com/pytorch/pytorch/issues/86265,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2022-10-18T21:23:24Z +issue,87087,Failing periodic tests: test_dense_mask_index_cpu (__main__.CpuTests) & est_expanded_reduction_cpu (__main__.CpuTests),https://github.com/pytorch/pytorch/issues/87087,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests,scan when time allows,2022-10-18T19:56:35Z +issue,87178,The autogenerated out variants via `autogen:` do not check that the dtype of the `out` kwarg via `canCast`.,https://github.com/pytorch/pytorch/issues/87178,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-10-18T19:55:55Z +issue,87031,Improve Readability of error(s) when provided unexpected keyword arguments.,https://github.com/pytorch/pytorch/issues/87031,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: python frontend",scan when time allows,2022-10-18T17:52:36Z +issue,87003,Multiprocessing DataLoader pickles multiprocessing.Queues incorrectly,https://github.com/pytorch/pytorch/issues/87003,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-10-18T17:26:10Z +issue,86968,Drop deprecated behavior from NumPy-style `T`,https://github.com/pytorch/pytorch/issues/86968,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy",scan when time allows,2022-10-18T17:22:36Z +issue,87076,Nandense layer for missing values,https://github.com/pytorch/pytorch/issues/87076,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: masked operators, module: nn, module: sparse",scan when time allows,2022-10-18T14:58:47Z +issue,87196,Missing `docker` directory in `tools/`,https://github.com/pytorch/pytorch/issues/87196,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, module: docs",scan when time allows,2022-10-18T14:44:49Z +issue,87080,cppextension host compiler check ignores executable symbolic link in CUDA bin directory,https://github.com/pytorch/pytorch/issues/87080,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions",scan when time allows,2022-10-18T02:21:29Z +issue,87056,Pipe conveys inconsistent value in GPU env,https://github.com/pytorch/pytorch/issues/87056,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2022-10-18T01:31:58Z +issue,50549,Channels last doesn't improve speed when using SyncBatchNorm,https://github.com/pytorch/pytorch/issues/50549,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: performance",scan when time allows,2022-10-17T22:47:54Z +issue,87033,Saving and loading from physical storage,https://github.com/pytorch/pytorch/issues/87033,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: serialization",scan when time allows,2022-10-17T16:09:26Z +issue,84616,Autogenerated out functions are missing at::cpu:: and co bindings,https://github.com/pytorch/pytorch/issues/84616,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-10-16T03:06:05Z +issue,86890,View-based advanced indexing (Integer array/LongTensor indexing) of nested_tensor,https://github.com/pytorch/pytorch/issues/86890,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2022-10-15T13:51:00Z +issue,83764,Missing header file,https://github.com/pytorch/pytorch/issues/83764,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-10-15T02:48:24Z +issue,67350,[FX] autowrap_functions doesn't work for module-scoped functions,https://github.com/pytorch/pytorch/issues/67350,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-10-14T19:23:15Z +issue,86965,Upgrade to a newer llvm-openmp version to avoid `/dev/shm` pollution,https://github.com/pytorch/pytorch/issues/86965,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: openmp",scan when time allows,2022-10-14T16:09:47Z +issue,86814,Expanding the parameters of `torch.svd_lowrank`,https://github.com/pytorch/pytorch/issues/86814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-10-14T14:00:23Z +issue,86918,torch.cat on empty tensor is bogus,https://github.com/pytorch/pytorch/issues/86918,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-10-14T13:57:13Z +issue,86819,Could not run select_backward [vmap] [dlrm] [functorch],https://github.com/pytorch/pytorch/issues/86819,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2022-10-13T19:48:59Z +issue,86798,TF32 conv_transpose2d with groups has bad precision compared to fp32,https://github.com/pytorch/pytorch/issues/86798,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability, module: tf32",scan when time allows,2022-10-13T14:47:09Z +issue,72556,"Clarify test dependencies (e.g., into a test-requirements.txt file)",https://github.com/pytorch/pytorch/issues/72556,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: ci",scan when time allows,2022-10-13T13:22:30Z +issue,86315,Feature request: Tests for `int` should be tests for `numbers.Integral`,https://github.com/pytorch/pytorch/issues/86315,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python frontend",scan when time allows,2022-10-12T19:10:51Z +issue,56485,Sparse tensor CSR layout for CUDA,https://github.com/pytorch/pytorch/issues/56485,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2022-10-12T16:45:45Z +issue,9222,[feature request] Implementing Block Sparse Operations,https://github.com/pytorch/pytorch/issues/9222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-10-12T16:26:38Z +issue,86381,OpInfo Tests To Validate that All Operators Are Being Tested With Strided Tensors,https://github.com/pytorch/pytorch/issues/86381,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-10-12T15:30:32Z +issue,86782,Poisson sampling on GPU fails for high rates,https://github.com/pytorch/pytorch/issues/86782,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: random",scan when time allows,2022-10-12T14:18:58Z +issue,75747,Depthwise Conv1d performance (a naive CUDA kernel is 10x faster),https://github.com/pytorch/pytorch/issues/75747,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-10-12T13:25:31Z +issue,66491,Normalize handling of scalar arguments,https://github.com/pytorch/pytorch/issues/66491,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-10-12T08:43:12Z +issue,86465,Wrong results with torch.linalg.inv on batched matrices when using cuda,https://github.com/pytorch/pytorch/issues/86465,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda, module: linear algebra, module: magma",scan when time allows,2022-10-12T00:57:00Z +issue,86717,NVFuser `FusionRootMappingMultipleBroadcast_CUDA` raises exception on sm_80+,https://github.com/pytorch/pytorch/issues/86717,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda, module: nvfuser",scan when time allows,2022-10-11T20:18:03Z +issue,86444,Dedicated function for shallow_copy_and_detach,https://github.com/pytorch/pytorch/issues/86444,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-10-11T16:11:55Z +issue,86676,Pytorch built for Jetson errors if CUDA is not found,https://github.com/pytorch/pytorch/issues/86676,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: cuda",scan when time allows,2022-10-11T13:06:22Z +issue,36990,[docs] Explain active_bytes in torch.cuda.memory_stats and Cuda Memory Management,https://github.com/pytorch/pytorch/issues/36990,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-10-11T07:44:02Z +issue,85841,Setting the cuda device when using start_processes in Jupyter on Ampere leads to CUDA reinitialization error,https://github.com/pytorch/pytorch/issues/85841,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2022-10-11T05:51:54Z +issue,86548,About autocast,https://github.com/pytorch/pytorch/issues/86548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-10-11T05:51:13Z +issue,86467,torch.tensor obj automatically moved to shared memory upon Process launch,https://github.com/pytorch/pytorch/issues/86467,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2022-10-11T05:48:56Z +issue,86601,NVFuser batch norm with prims: internal assert failure from test suite,https://github.com/pytorch/pytorch/issues/86601,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch, module: tests",scan when time allows,2022-10-11T05:41:44Z +issue,86660,libtorch make failed,https://github.com/pytorch/pytorch/issues/86660,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-10-11T05:41:28Z +issue,86544,Compile failed at allreduce without gloo,https://github.com/pytorch/pytorch/issues/86544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-10-11T00:03:16Z +issue,86493,`mem_get_info` reserves memory and can not be destroyed / deallocated.,https://github.com/pytorch/pytorch/issues/86493,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-10-10T20:38:09Z +issue,86271,AUTOGRAD is not working on IOS,https://github.com/pytorch/pytorch/issues/86271,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: ios, oncall: mobile",scan when time allows,2022-10-10T18:17:16Z +issue,86551,`MultiLabelMarginLoss` will return incorrect values in JIT after the first run on cuda,https://github.com/pytorch/pytorch/issues/86551,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nvfuser, oncall: jit",scan when time allows,2022-10-10T17:59:48Z +issue,86563,[functorch] colab links on functorch 0.2.0 website should be linked to a permalinked version of the colabs,https://github.com/pytorch/pytorch/issues/86563,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: functorch",scan when time allows,2022-10-10T16:54:17Z +issue,86554,[NvFuser] would change the output for some inaccurate dtype,https://github.com/pytorch/pytorch/issues/86554,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-10-10T15:07:41Z +issue,86192,fmt/src/os.cc: error: unknown type name 'error_code'; did you mean 'std::error_code'?,https://github.com/pytorch/pytorch/issues/86192,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-10-10T12:00:36Z +issue,86204,How to perform unstructured interpolation,https://github.com/pytorch/pytorch/issues/86204,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2022-10-10T11:59:06Z +issue,86194,path in WORKSPACE,https://github.com/pytorch/pytorch/issues/86194,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bazel,scan when time allows,2022-10-10T11:57:45Z +issue,15457,Improve one_hot,https://github.com/pytorch/pytorch/issues/15457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-10-10T04:29:41Z +issue,44938,DataParallel on CPU,https://github.com/pytorch/pytorch/issues/44938,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2022-10-08T14:32:22Z +issue,86111,When will the torch.sparse module be usable?,https://github.com/pytorch/pytorch/issues/86111,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-10-07T20:24:12Z +issue,86298,AOT Autograd Device Partitioning,https://github.com/pytorch/pytorch/issues/86298,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functorch,scan when time allows,2022-10-06T22:33:21Z +issue,86270,Autocast with BF16 on CPU slows down model more than 2X,https://github.com/pytorch/pytorch/issues/86270,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: bfloat16, module: performance",scan when time allows,2022-10-06T20:07:12Z +issue,86326,`torch.special.round` doesn't support the same dtypes as `torch.round`,https://github.com/pytorch/pytorch/issues/86326,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2022-10-06T17:20:27Z +issue,86356,`conv_transpose` is not similar to `nn.grad.conv_input` when `output_padding` is passed with non-default values.,https://github.com/pytorch/pytorch/issues/86356,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-10-06T17:18:58Z +issue,86339,Add `maximize` support to LBFGS optimizer,https://github.com/pytorch/pytorch/issues/86339,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-10-05T23:51:39Z +issue,32694,ninja: build stopped: subcommand failed.,https://github.com/pytorch/pytorch/issues/32694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-10-05T22:05:25Z +issue,85775,Timed out receiving the shared seed from the distribtued store on Rank 2,https://github.com/pytorch/pytorch/issues/85775,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",scan when time allows,2022-10-05T13:01:48Z +issue,86225,[Feature] Dispatching PyTorch Distributed Collectives,https://github.com/pytorch/pytorch/issues/86225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-10-05T11:31:18Z +issue,69505,"with the same environment ,pytorch 1.8 worked but 1.10 can't work",https://github.com/pytorch/pytorch/issues/69505,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: wsl",scan when time allows,2022-10-04T22:50:29Z +issue,85607,[Distributed: RPC] Failed to initialize RPC with >18 workers,https://github.com/pytorch/pytorch/issues/85607,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2022-10-04T20:51:17Z +issue,85377,CI fails for test_compare_cpu_nn_functional_embedding_cuda_float32 which is not reproducible locally,https://github.com/pytorch/pytorch/issues/85377,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: embedding, module: tests",scan when time allows,2022-10-04T19:24:36Z +issue,85889,High occupation on GPU 0 when converting Tensor to multi GPU,https://github.com/pytorch/pytorch/issues/85889,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2022-10-04T17:07:24Z +issue,5565,[feature request] Different interpolation algos for 'grid_sample' function,https://github.com/pytorch/pytorch/issues/5565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: interpolation",scan when time allows,2022-10-04T10:00:42Z +issue,85618,Build from source failed with error of different gpu architecture (compiler shows sm_30-related error but I use sm_86 GPU),https://github.com/pytorch/pytorch/issues/85618,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2022-10-04T08:25:23Z +issue,85625,How to install pytorch with cuda 11.7 in anaconda envirment?,https://github.com/pytorch/pytorch/issues/85625,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-10-04T08:23:30Z +issue,47568,Failed to get generated_cpp list,https://github.com/pytorch/pytorch/issues/47568,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-10-04T04:56:29Z +issue,86055,partial view/reshaping,https://github.com/pytorch/pytorch/issues/86055,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-10-03T23:22:37Z +issue,24470,Benchmark cudnn version of grid sampler,https://github.com/pytorch/pytorch/issues/24470,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: performance",scan when time allows,2022-10-03T22:27:28Z +issue,86074,torch.remainder and torch.fmod produce wrong results,https://github.com/pytorch/pytorch/issues/86074,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: numerical-stability",scan when time allows,2022-10-03T22:22:47Z +issue,56489,[torchdeploy] ar: _bz2module.o: No such file or directory,https://github.com/pytorch/pytorch/issues/56489,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deploy,scan when time allows,2022-10-03T17:48:52Z +issue,84316,Torch.fx tracing bug with dictionary.update calls on input,https://github.com/pytorch/pytorch/issues/84316,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-10-03T17:19:52Z +issue,83851,torch fx cannot trace assert for some cases,https://github.com/pytorch/pytorch/issues/83851,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-10-03T17:19:52Z +issue,84515,Torch.FX work with autograd.Function,https://github.com/pytorch/pytorch/issues/84515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-10-03T17:18:24Z +issue,85815,string interning for dispatcher operator names,https://github.com/pytorch/pytorch/issues/85815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2022-10-03T17:15:05Z +issue,59356,pytorch build failure on arm64,https://github.com/pytorch/pytorch/issues/59356,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: cuda, module: jetson",scan when time allows,2022-10-03T15:51:47Z +issue,86097,[Activation Checkpointing] Investigate pin_memory for CPU offload,https://github.com/pytorch/pytorch/issues/86097,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, oncall: distributed",scan when time allows,2022-10-03T06:17:05Z +issue,85698,AMP consumes 30x gpu memory with bmm,https://github.com/pytorch/pytorch/issues/85698,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-09-30T22:08:41Z +issue,82565,PyTorch for quantum mechanics,https://github.com/pytorch/pytorch/issues/82565,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scientific computing,scan when time allows,2022-09-30T09:19:33Z +issue,80942,[LTC] OOM on mnist example,https://github.com/pytorch/pytorch/issues/80942,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-09-30T02:39:48Z +issue,85834,[primTorch] Need to update data-dependent check policy,https://github.com/pytorch/pytorch/issues/85834,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-09-29T19:32:17Z +issue,85652,CUDA allocator feature requests,https://github.com/pytorch/pytorch/issues/85652,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudacachingallocator",scan when time allows,2022-09-29T17:37:56Z +issue,64254,[feature request] New copyseq_ method,https://github.com/pytorch/pytorch/issues/64254,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2022-09-29T08:16:09Z +issue,84537,Minimal example for torch.optim.SparseAdam,https://github.com/pytorch/pytorch/issues/84537,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sparse",scan when time allows,2022-09-28T16:11:18Z +issue,85642,Could not run 'aten::native_batch_norm' with arguments from the 'SparseCUDA' backend. using batch_norm,https://github.com/pytorch/pytorch/issues/85642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-09-28T15:29:36Z +issue,85544,NvFuser single mode changes the output,https://github.com/pytorch/pytorch/issues/85544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-09-28T14:15:09Z +issue,56144,libtorch static linking results in undefined references to onnx_torch and caffe2::EmbeddingLookup,https://github.com/pytorch/pytorch/issues/56144,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: undefined reference,scan when time allows,2022-09-28T09:12:26Z +issue,60295,Optimize torch.einsum,https://github.com/pytorch/pytorch/issues/60295,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2022-09-27T14:38:15Z +issue,14367,Error: DeviceGuardImpl for cpu is not available (static linking PyTorch),https://github.com/pytorch/pytorch/issues/14367,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: static linking",scan when time allows,2022-09-27T13:37:00Z +issue,85606,Creating NumPy array with `dtype=object` of PyTorch tensors fails,https://github.com/pytorch/pytorch/issues/85606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2022-09-27T08:28:47Z +issue,85558,torch.Tensor.transpose().contiguous() on dimension of size 1 gives wrong stride,https://github.com/pytorch/pytorch/issues/85558,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-09-26T20:36:16Z +issue,85520,Implement `rand_like` ref and implement nvfuser_impl for `uniform` prim,https://github.com/pytorch/pytorch/issues/85520,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-09-26T17:44:07Z +issue,85590,Solve default argument induced include cycles by not using defaults / moving the defaults to inl,https://github.com/pytorch/pytorch/issues/85590,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: internals",scan when time allows,2022-09-26T15:02:20Z +issue,52973,Add numerically stable methods to torch.distributions (e.g. logcdf),https://github.com/pytorch/pytorch/issues/52973,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-09-26T09:16:00Z +issue,85201,performance between manually created graph and CUDAGraph.replay,https://github.com/pytorch/pytorch/issues/85201,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2022-09-26T02:01:23Z +issue,34839,[torch.jit.script] Allow `range` to index into Tensor,https://github.com/pytorch/pytorch/issues/34839,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-09-24T09:28:56Z +issue,85227,topk returns different results with the same input in cuda and cpu,https://github.com/pytorch/pytorch/issues/85227,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sorting and selection",scan when time allows,2022-09-23T02:45:34Z +issue,31818,How to distinguish different layers in hook？,https://github.com/pytorch/pytorch/issues/31818,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-09-22T22:55:48Z +issue,85375,Inconsistency between geometric distributions,https://github.com/pytorch/pytorch/issues/85375,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: molly-guard",scan when time allows,2022-09-22T18:47:03Z +issue,85438,Prim Output Spec is Not Always Consistent With Eager,https://github.com/pytorch/pytorch/issues/85438,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-09-22T15:24:49Z +issue,69991,Composite Compliance Problems Tracker,https://github.com/pytorch/pytorch/issues/69991,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: internals, module: linear algebra",scan when time allows,2022-09-22T00:22:45Z +issue,85397,torch.nn.utils.prune.remove reorders the parameters of a module unexpectedly,https://github.com/pytorch/pytorch/issues/85397,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pruning,scan when time allows,2022-09-21T19:57:57Z +issue,85392,please report a bug to PyTorch. Expected Object but got PyObject,https://github.com/pytorch/pytorch/issues/85392,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: torchbind,scan when time allows,2022-09-21T19:56:32Z +issue,85390,Please put back missing rocm builds of Torch Vision.,https://github.com/pytorch/pytorch/issues/85390,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: rocm",scan when time allows,2022-09-21T19:55:16Z +issue,85387,very strange speed of torch.bmm with specific tensor shape,https://github.com/pytorch/pytorch/issues/85387,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",scan when time allows,2022-09-21T19:54:24Z +issue,85351,Point community docs to master,https://github.com/pytorch/pytorch/issues/85351,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2022-09-21T19:50:02Z +issue,85302,Remove `TypedStorage` and use only `UntypedStorage`,https://github.com/pytorch/pytorch/issues/85302,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-09-21T19:43:51Z +issue,85229,TracerWarning: Output nr 1. of the traced function does not match the corresponding output of the Python function.,https://github.com/pytorch/pytorch/issues/85229,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-09-21T19:34:33Z +issue,78050,RFC: [primTorch] Stride-agnostic Operator Semantics,https://github.com/pytorch/pytorch/issues/78050,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: primtorch, module: python frontend",scan when time allows,2022-09-21T16:31:43Z +issue,76288,Multi-GPU distributed training reports errors,https://github.com/pytorch/pytorch/issues/76288,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-09-21T02:30:04Z +issue,85258,"nvFuser support for {ceil,floor,round,trunc}(int)",https://github.com/pytorch/pytorch/issues/85258,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-09-20T15:24:52Z +issue,81648,`sparse_coo.to_dense()` produces different results between CPU and CUDA backends for boolean non-coalesced inputs.,https://github.com/pytorch/pytorch/issues/81648,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-09-20T11:51:09Z +issue,84265,torch.Tensor.to.dtype_layout overload is not available in Python,https://github.com/pytorch/pytorch/issues/84265,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: python frontend",scan when time allows,2022-09-19T23:25:20Z +issue,83733,BCELoss results in autocast CUDA warning,https://github.com/pytorch/pytorch/issues/83733,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-09-19T18:45:57Z +issue,85058,Selectively sync internal Meta discussions / posts to dev-discuss.pytorch.org,https://github.com/pytorch/pytorch/issues/85058,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-19T17:20:05Z +issue,32998,torch batchwise max with indices,https://github.com/pytorch/pytorch/issues/32998,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: reductions",scan when time allows,2022-09-19T07:45:44Z +issue,84835,Batch multiplication for torch.sparse matrix multiplication,https://github.com/pytorch/pytorch/issues/84835,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2022-09-17T16:24:00Z +issue,4954,[docs] Docs website search finds duplicates and produces bad snippets,https://github.com/pytorch/pytorch/issues/4954,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2022-09-16T23:45:38Z +issue,85147,torch::quantile performance?,https://github.com/pytorch/pytorch/issues/85147,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2022-09-16T20:56:48Z +issue,85122,"PT Dispatcher confusing error message ""There were no tensor arguments to this function""",https://github.com/pytorch/pytorch/issues/85122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2022-09-16T20:55:01Z +issue,85105,Python dispatch for PyOps needs to respect tensor subclasses,https://github.com/pytorch/pytorch/issues/85105,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: dispatch",scan when time allows,2022-09-16T20:54:52Z +issue,85082,"[NNC] loop vectorization fails, `Ramp` and `Broadcast` undefined",https://github.com/pytorch/pytorch/issues/85082,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-16T20:30:09Z +issue,85078,"libtorch create a tensor is very slow, who can tell me why",https://github.com/pytorch/pytorch/issues/85078,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: performance",scan when time allows,2022-09-16T20:28:03Z +issue,46386,"GPU memory leak when registering a forward hook with ""self"" access",https://github.com/pytorch/pytorch/issues/46386,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2022-09-16T19:40:23Z +issue,85121,make_traced() doesn't respect setting the seed,https://github.com/pytorch/pytorch/issues/85121,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: primtorch, module: tests",scan when time allows,2022-09-15T22:26:24Z +issue,85096,Update `use_deterministic_algorithms` documentation and tests to include `nn.functional` counterparts for all `nn` modules,https://github.com/pytorch/pytorch/issues/85096,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: docs, module: tests",scan when time allows,2022-09-15T18:47:18Z +issue,85081,primTorch/nvfuser: have a way to check that refs are added to __all__,https://github.com/pytorch/pytorch/issues/85081,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-09-15T18:35:36Z +issue,38240,RuntimeError: arg_types.size() == param_names.size() - (moduleSelf_ ? 1 : 0) INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/38240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: typing, oncall: jit",scan when time allows,2022-09-15T11:11:57Z +issue,84813,MPS: allow selecting specific MTLDevice by registryID via environment variable,https://github.com/pytorch/pytorch/issues/84813,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mps,scan when time allows,2022-09-14T21:22:12Z +issue,77514,`max_pool1d` can succeed when padding is negative for tensor requiring grad,https://github.com/pytorch/pytorch/issues/77514,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nn, module: pooling",scan when time allows,2022-09-14T21:03:45Z +issue,83817,[feature request] Add new device type works on CPU,https://github.com/pytorch/pytorch/issues/83817,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-14T07:10:57Z +issue,84717,Test aten decompositions match their alias information,https://github.com/pytorch/pytorch/issues/84717,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-09-13T23:52:13Z +issue,84864,torch.utils.checkpoint (with use_reentrant=False) doesn't work with all PyTorch features that set TLS,https://github.com/pytorch/pytorch/issues/84864,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: activation checkpointing",scan when time allows,2022-09-13T23:45:08Z +issue,84805,compiling failed from source,https://github.com/pytorch/pytorch/issues/84805,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: sparse",scan when time allows,2022-09-13T23:42:24Z +issue,84932,Separate doc and binaries build,https://github.com/pytorch/pytorch/issues/84932,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2022-09-13T23:36:41Z +issue,84840,"TypeError: finfo(): argument 'type' (position 1) must be torch.dtype, not HFProxy",https://github.com/pytorch/pytorch/issues/84840,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-09-13T21:54:09Z +issue,84721,PyTorch-DirectML RFC,https://github.com/pytorch/pytorch/issues/84721,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-13T21:50:36Z +issue,84710,About the different ways to print models,https://github.com/pytorch/pytorch/issues/84710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: printing,scan when time allows,2022-09-13T21:48:39Z +issue,79039,Add Autograd Support for Nested Tensor,https://github.com/pytorch/pytorch/issues/79039,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-09-13T18:10:48Z +issue,1927,[feature request] time-distributed layers for application of normal layers to sequence data,https://github.com/pytorch/pytorch/issues/1927,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-09-13T17:05:57Z +issue,72709,Feature: Use iterative refinement algorithm from cuSOLVER for the least-squares solver on CUDA,https://github.com/pytorch/pytorch/issues/72709,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2022-09-13T10:41:57Z +issue,84538,OpInfo tests should compare gpu to cpu implementations,https://github.com/pytorch/pytorch/issues/84538,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-09-12T18:36:11Z +issue,63921,More efficient colon backwards in advanced indexing,https://github.com/pytorch/pytorch/issues/63921,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: performance",scan when time allows,2022-09-12T17:13:39Z +issue,84620,torch.hub.load local model,https://github.com/pytorch/pytorch/issues/84620,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2022-09-12T15:10:40Z +issue,66162,Cuda Out of Memory - MMDMatic,https://github.com/pytorch/pytorch/issues/66162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2022-09-11T21:07:01Z +issue,84691,test_public_bindings is not robust to various build options,https://github.com/pytorch/pytorch/issues/84691,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-08T18:38:57Z +issue,84597,Accept SymInts and SymFloats For Scalar Inputs,https://github.com/pytorch/pytorch/issues/84597,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-08T18:35:11Z +issue,84565,Tensor Subclass that doesn't require grad may wrap a Tensor subclass that requires grad,https://github.com/pytorch/pytorch/issues/84565,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-07T23:35:19Z +issue,84630,"torch.nn.functional.interpolate fails on some degenerate shapes, but passes on others",https://github.com/pytorch/pytorch/issues/84630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2022-09-07T23:30:24Z +issue,25387,[DRAFT] Auto-casting in JIT - Automatic mixed precision,https://github.com/pytorch/pytorch/issues/25387,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-09-07T22:13:30Z +issue,77733,Adding Vulkan Support,https://github.com/pytorch/pytorch/issues/77733,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2022-09-07T21:52:49Z +issue,84625,Beta distribution behaves incorrectly for small parameters,https://github.com/pytorch/pytorch/issues/84625,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: edge cases",scan when time allows,2022-09-07T21:08:11Z +issue,32166,"C++ randint returns float32, python returns int64",https://github.com/pytorch/pytorch/issues/32166,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-09-07T17:59:29Z +issue,23233,"Unofficial ARMv6, ARMv7, ARMv8 builds",https://github.com/pytorch/pytorch/issues/23233,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2022-09-07T12:54:11Z +issue,59243,Can't install pytorch =1.7 with python 3.8.5 on Raspberry Pi,https://github.com/pytorch/pytorch/issues/59243,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: arm,scan when time allows,2022-09-07T12:50:58Z +issue,80541,[Prims+NVFuser] Prims with missing NVFuser ops,https://github.com/pytorch/pytorch/issues/80541,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-09-07T10:44:28Z +issue,84468,[c10d] Support a public API to retrieve default process group,https://github.com/pytorch/pytorch/issues/84468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-09-06T22:16:26Z +issue,83032,Support for CSR Tensor with NN layers,https://github.com/pytorch/pytorch/issues/83032,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sparse",scan when time allows,2022-09-06T18:36:13Z +issue,84529,"test_prims.py:test_nvfuser_no_args_cuda, memory leak",https://github.com/pytorch/pytorch/issues/84529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-09-06T13:41:41Z +issue,84539,list of tensors can't be converted to a torch tensor while list of lists gets easily converted to a pytorch tensor,https://github.com/pytorch/pytorch/issues/84539,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2022-09-06T13:40:22Z +issue,84560,[optim] asgd : handling of complex params as real params (NaN vs inf),https://github.com/pytorch/pytorch/issues/84560,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: optimizer",scan when time allows,2022-09-06T13:38:03Z +issue,83977,pytorch 1.12.1 doesn't build with ffmpeg 5.0,https://github.com/pytorch/pytorch/issues/83977,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-09-06T08:46:13Z +issue,75940,Add nesting of nested Tensor,https://github.com/pytorch/pytorch/issues/75940,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-09-05T18:43:22Z +issue,61781,Letting `_allgather_base` to support multiple tensors as inputs and outputs,https://github.com/pytorch/pytorch/issues/61781,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-09-05T08:41:36Z +issue,83902,"Bug in batch names with matmul (result tensor has names=('i', 'i', 'k')).",https://github.com/pytorch/pytorch/issues/83902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2022-09-05T07:05:49Z +issue,84335,Should enable skipped tests for `to` OpInfo,https://github.com/pytorch/pytorch/issues/84335,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-09-03T19:46:23Z +issue,84445,Strange cuda illegal memory allocation error,https://github.com/pytorch/pytorch/issues/84445,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-09-02T20:06:45Z +issue,84473,Install LibTorch by Conan or other C++ package manager,https://github.com/pytorch/pytorch/issues/84473,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-09-02T16:58:09Z +issue,84414,several questions about pytorch DDP,https://github.com/pytorch/pytorch/issues/84414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: nccl",scan when time allows,2022-09-02T16:52:37Z +issue,84412,Odd type-casting behaviour in prims.div,https://github.com/pytorch/pytorch/issues/84412,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: primtorch, module: type promotion",scan when time allows,2022-09-02T16:49:23Z +issue,84353,torch.Size should convert all elements to ints,https://github.com/pytorch/pytorch/issues/84353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-09-01T15:00:55Z +issue,84311,DecompositionInterpreter creates invalid graph,https://github.com/pytorch/pytorch/issues/84311,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-09-01T14:25:34Z +issue,84346,fix ATen tests that do not compile,https://github.com/pytorch/pytorch/issues/84346,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-09-01T14:24:24Z +issue,84259,would you like upload to the cpp libtorch to vcpkg package repo?,https://github.com/pytorch/pytorch/issues/84259,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cpp",scan when time allows,2022-09-01T09:37:04Z +issue,71022,installation of pytorch `cpuonly` from conda with `nomkl` installs `mkl`,https://github.com/pytorch/pytorch/issues/71022,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-08-30T12:49:41Z +issue,83968,"Python3 Depletes 2021 M1 Mac Memory Running Training Ops For Model's M, L and X",https://github.com/pytorch/pytorch/issues/83968,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: macos,scan when time allows,2022-08-30T12:10:43Z +issue,84194,Pytorch gets small bias on the result of different types of divisors while doing floating point division.,https://github.com/pytorch/pytorch/issues/84194,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability",scan when time allows,2022-08-30T06:02:45Z +issue,84247,Ensure ops account for offsets and strides,https://github.com/pytorch/pytorch/issues/84247,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-08-30T03:32:01Z +issue,84202,Gradient value calculation error in MultiLabelMarginLoss,https://github.com/pytorch/pytorch/issues/84202,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda, module: loss",scan when time allows,2022-08-29T23:32:33Z +issue,84140,Please include virtual/physical batch sizes in the tutorials,https://github.com/pytorch/pytorch/issues/84140,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-08-29T22:22:29Z +issue,80742,Add TorchDynamo as a submodule to Pytorch?,https://github.com/pytorch/pytorch/issues/80742,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-08-29T20:03:57Z +issue,84045,COREMLTOOLs/NNPACK Python Issue,https://github.com/pytorch/pytorch/issues/84045,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nnpack,scan when time allows,2022-08-29T13:56:00Z +issue,84050,I cannot install pytorch by Bad CRC-32 for file 'torch/lib/libtorch_cpu.so',https://github.com/pytorch/pytorch/issues/84050,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-08-29T13:43:02Z +issue,67590,Add `OptState.UPDATED` to `torch.cuda.amp.GradScaler`,https://github.com/pytorch/pytorch/issues/67590,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-08-29T11:58:17Z +issue,42812,checkpoint_sequential breaks backpropagation,https://github.com/pytorch/pytorch/issues/42812,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd",scan when time allows,2022-08-29T11:32:00Z +issue,80594,RuntimeError: DataLoader worker (pid 22822) is killed by signal: Aborted.,https://github.com/pytorch/pytorch/issues/80594,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-08-28T22:05:52Z +issue,44380,Support for NVIDIA UVM technology,https://github.com/pytorch/pytorch/issues/44380,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-08-28T21:53:19Z +issue,83775,[Nested Tensor] Move nested tensor specific ops to nested namespace,https://github.com/pytorch/pytorch/issues/83775,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-08-24T22:26:18Z +issue,81620,PyTorch 1.12 cu113 Illegal Memory Access or Internal Error instead of Out of Memory cases,https://github.com/pytorch/pytorch/issues/81620,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2022-08-24T21:26:16Z +issue,83923,Support primtorch view ops in functionalization,https://github.com/pytorch/pytorch/issues/83923,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: functionalization, module: primtorch, module: viewing and reshaping",scan when time allows,2022-08-24T20:58:07Z +issue,83914,"RAM not free when deleting a model in CPU? worse after inference, is there some cache hidden?",https://github.com/pytorch/pytorch/issues/83914,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2022-08-24T20:41:52Z +issue,84014,"fill_ OpInfo code not used, also, doesn't test the case where the second argument is a Tensor",https://github.com/pytorch/pytorch/issues/84014,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-08-24T20:40:39Z +issue,83931,primTorch: support refs and decompositions when ATen and Python disagree,https://github.com/pytorch/pytorch/issues/83931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-08-24T20:32:38Z +issue,83932,Hitting rate limits for pytorchbot token,https://github.com/pytorch/pytorch/issues/83932,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-08-24T20:32:20Z +issue,83979,Undefined reference in libtorch_cpu.so `...std::__cxx11::basic_string...`,https://github.com/pytorch/pytorch/issues/83979,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-08-24T20:31:14Z +issue,83980,Scatter min/max reduce operation that returns the corresponding indices,https://github.com/pytorch/pytorch/issues/83980,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2022-08-24T20:23:10Z +issue,84009,[Nested Tensor] Enable Nestedtensor to work with OpInfos,https://github.com/pytorch/pytorch/issues/84009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-08-24T19:27:12Z +issue,83910,Tracking nested tensor functions with backward kernels registered in derivatives.yaml,https://github.com/pytorch/pytorch/issues/83910,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-08-24T17:59:44Z +issue,83947,Are PyTorch Android nightly builds getting automatically published,https://github.com/pytorch/pytorch/issues/83947,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: ci",scan when time allows,2022-08-24T01:17:45Z +issue,83941,empty_quantized should probably be new_empty_quantized,https://github.com/pytorch/pytorch/issues/83941,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2022-08-23T21:42:29Z +issue,83826,test_lazy spuriously fails if LAPACK is not installed,https://github.com/pytorch/pytorch/issues/83826,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lazy, module: tests",scan when time allows,2022-08-23T14:39:39Z +issue,83714,Build from source failed on MacOS 10.6 with CUDA 10.1,https://github.com/pytorch/pytorch/issues/83714,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2022-08-23T14:04:00Z +issue,83726,nvfuser + prim stack generated illegal PTX code on hardware with sm <= 70,https://github.com/pytorch/pytorch/issues/83726,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-08-23T13:29:15Z +issue,76128,test_license_for_wheel always fails on my local dev copy,https://github.com/pytorch/pytorch/issues/76128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-08-22T15:45:32Z +issue,83320,Torch1.10.2 is slower than torch1.9.1,https://github.com/pytorch/pytorch/issues/83320,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-08-22T02:48:45Z +issue,61653,Move `lobpcg` into `torch.linalg`,https://github.com/pytorch/pytorch/issues/61653,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-08-21T10:39:09Z +issue,72146,Also allow dicts as type of `params=` field in param groups of optimizers,https://github.com/pytorch/pytorch/issues/72146,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-08-20T12:42:20Z +issue,83749,[Nested Tensor] Update TestCase.AssertEqual,https://github.com/pytorch/pytorch/issues/83749,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: testing",scan when time allows,2022-08-20T09:45:31Z +issue,73171,libtorch need operator= in torch::Device,https://github.com/pytorch/pytorch/issues/73171,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-08-19T15:15:10Z +issue,83657,[Discussion] Add custom device,https://github.com/pytorch/pytorch/issues/83657,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: dispatch",scan when time allows,2022-08-19T14:37:17Z +issue,83148,Cannot call CUDAGeneratorImpl::current_seed during CUDA graph capture,https://github.com/pytorch/pytorch/issues/83148,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-08-19T07:32:20Z +issue,18414,Add layer-wise adaptive rate scaling (LARS) optimizer,https://github.com/pytorch/pytorch/issues/18414,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-08-19T04:17:41Z +issue,83624,Unhelpful error message from torch.linalg.ldl_factor,https://github.com/pytorch/pytorch/issues/83624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: linear algebra",scan when time allows,2022-08-18T15:23:38Z +issue,68055,Release official Python 3.9 images,https://github.com/pytorch/pytorch/issues/68055,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-08-18T14:39:45Z +issue,62640,add gamma to CosineAnnealingWarmRestarts so max lr can decrease in cycles,https://github.com/pytorch/pytorch/issues/62640,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2022-08-18T14:38:22Z +issue,63488,MultiStepLR with different gammas for each parameter group,https://github.com/pytorch/pytorch/issues/63488,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2022-08-18T14:37:33Z +issue,70352,Rprop Optimizer: UnboundLocalError: local variable 'step_size_min' referenced before assignment,https://github.com/pytorch/pytorch/issues/70352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-08-18T14:31:15Z +issue,74022,Inconsistent implementation on SWA,https://github.com/pytorch/pytorch/issues/74022,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2022-08-18T14:24:46Z +issue,81162,SparseAdam performance issue during optimizer step,https://github.com/pytorch/pytorch/issues/81162,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: performance, module: sparse",scan when time allows,2022-08-18T14:08:34Z +issue,83579,Conv1d: NNPACK SpatialConvolution_updateOutput failed when batchsize or padding is too large,https://github.com/pytorch/pytorch/issues/83579,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn, module: nnpack",scan when time allows,2022-08-17T21:22:49Z +issue,83564,KL-divergence of two Generalized Dirichlet distributions,https://github.com/pytorch/pytorch/issues/83564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-08-17T14:18:26Z +issue,83583,Silent promotion of bool to int in the dispatcher,https://github.com/pytorch/pytorch/issues/83583,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: library, module: pybind, module: type promotion",scan when time allows,2022-08-17T13:13:04Z +issue,83493,Calling torch.linalg.cholesky on a CPU tensor requires compiling PyTorch with LAPACK.,https://github.com/pytorch/pytorch/issues/83493,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-08-17T06:54:24Z +issue,78018,Three memory copies of every dataloader cpu tensor,https://github.com/pytorch/pytorch/issues/78018,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader, module: multiprocessing",scan when time allows,2022-08-16T22:04:22Z +issue,83537,Add a new argument `check_inf=True` (by default) or check_pos_inf / check_neg_inf to anomaly mode,https://github.com/pytorch/pytorch/issues/83537,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-08-16T21:10:11Z +issue,83521,quantize_per_tensor/quantize_per_channel operators should honor the quant_min/quant_max from observer,https://github.com/pytorch/pytorch/issues/83521,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2022-08-16T16:44:10Z +issue,81317,Incorrect results for mean or sum kernels on aarch64 when building with gcc-7,https://github.com/pytorch/pytorch/issues/81317,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: arm,scan when time allows,2022-08-16T10:58:21Z +issue,50953,Dispatch-less structured wrapper / composite / alias kernels,https://github.com/pytorch/pytorch/issues/50953,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: structured kernels",scan when time allows,2022-08-16T08:12:10Z +issue,63079,"PyTorch 1.9.0, test_optim fails on Nvidia A10.",https://github.com/pytorch/pytorch/issues/63079,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: optimizer, module: tests",scan when time allows,2022-08-15T14:52:21Z +issue,83392,"How to turn off determinism just for specific operations, e.g. upsampling through bilinear interpolation?",https://github.com/pytorch/pytorch/issues/83392,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism",scan when time allows,2022-08-15T04:42:36Z +issue,83315,dataparallel function doesn't work,https://github.com/pytorch/pytorch/issues/83315,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2022-08-15T02:23:58Z +issue,79016,Corner cases of ShardedTensor checkpoint when using TorchRec,https://github.com/pytorch/pytorch/issues/79016,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, oncall: distributed",scan when time allows,2022-08-12T19:59:47Z +issue,83292,TorchVision testing in CI + test_fx,https://github.com/pytorch/pytorch/issues/83292,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-08-12T17:38:40Z +issue,83271,cannot import name 'ProcessGroup' from 'torch.distributed',https://github.com/pytorch/pytorch/issues/83271,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, oncall: distributed",scan when time allows,2022-08-12T13:03:44Z +issue,83272,bmm operator in bfloat16 has low TFLOPS for some tensor shapes with CUDA 11.6,https://github.com/pytorch/pytorch/issues/83272,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2022-08-12T06:37:34Z +issue,83193,"module 'torch.distributed' has no attribute 'pipeline' - macOS, PyTorch 1.12.1",https://github.com/pytorch/pytorch/issues/83193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-08-12T06:00:49Z +issue,83245,"During DDP training timm densenet121, mobilenetv2(v3) models do not save state_dict correctly.",https://github.com/pytorch/pytorch/issues/83245,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-08-12T05:51:20Z +issue,83214,torchgen.model.FunctionSchema.parse fails with following ops' schema,https://github.com/pytorch/pytorch/issues/83214,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-08-11T23:19:26Z +issue,83019,TestCommon.test_dtypes error message is confusing,https://github.com/pytorch/pytorch/issues/83019,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-08-11T21:56:54Z +issue,83024,New PR template suggests a pattern that does not close PR,https://github.com/pytorch/pytorch/issues/83024,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-08-11T21:56:36Z +issue,83152,"When padding is big int, torch.nn.functional.fold runs too long and can't return result",https://github.com/pytorch/pytorch/issues/83152,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-08-11T21:54:55Z +issue,83157,Fail to install torch for source,https://github.com/pytorch/pytorch/issues/83157,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-08-11T21:54:33Z +issue,83175,"torch.nn.GRU runs long time, when num_layers is large",https://github.com/pytorch/pytorch/issues/83175,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: nn",scan when time allows,2022-08-11T21:53:02Z +issue,83169,torch.nn.functional.softplus / torch.nn.Softplus parameter beta can be set to zero,https://github.com/pytorch/pytorch/issues/83169,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-08-11T21:52:59Z +issue,83232,torch.nn.PixelShuffle error message wrong,https://github.com/pytorch/pytorch/issues/83232,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-08-11T21:44:35Z +issue,83238,The type of parameter 'p' in torch.nn.TripletMarginLoss wrong,https://github.com/pytorch/pytorch/issues/83238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-08-11T21:44:29Z +issue,83229,torch.nn.MaxUnpool2d get negative size tensor,https://github.com/pytorch/pytorch/issues/83229,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-08-11T21:43:59Z +issue,83221,torch.nn.InstanceNorm{1|2|3}d doesn't verify the value type of parameter num_features,https://github.com/pytorch/pytorch/issues/83221,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-08-11T21:43:34Z +issue,83281,addcdiv_ (in and out of place) not implemented for torch.float16 and cpu,https://github.com/pytorch/pytorch/issues/83281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2022-08-11T21:41:43Z +issue,83234,torch.nn.ReplicationPad{1|2}d supports more input dimension than are written on documentation,https://github.com/pytorch/pytorch/issues/83234,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2022-08-11T21:40:55Z +issue,83257,PyYAML not listed as a dependency,https://github.com/pytorch/pytorch/issues/83257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2022-08-11T21:38:17Z +issue,79325,[NVFuser] hitting fallbacks on demucs (from torchbench + lazy tensor),https://github.com/pytorch/pytorch/issues/79325,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-08-11T17:21:16Z +issue,71403,[discussion] torch.flatten to allow unsqueeze of inexisting dimension,https://github.com/pytorch/pytorch/issues/71403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: batching,scan when time allows,2022-08-11T09:02:08Z +issue,83074,DPP training incompatibility with checkpoint and detach,https://github.com/pytorch/pytorch/issues/83074,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-08-11T01:34:34Z +issue,82960,torch.bitwise_xor argument 'other' documentation wrong,https://github.com/pytorch/pytorch/issues/82960,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-08-10T13:23:41Z +issue,83060,torch.empty_strided argument 'size'and 'stride' documentation wrong,https://github.com/pytorch/pytorch/issues/83060,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-08-10T13:17:29Z +issue,83064,Updating the LTS version of the torch (1.8.2 -> 1.10.2\1.11.2?),https://github.com/pytorch/pytorch/issues/83064,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-08-10T13:07:21Z +issue,83082,[CI] Create periodic fuzzy testing for PyTorch build flags,https://github.com/pytorch/pytorch/issues/83082,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-08-10T13:02:55Z +issue,83135,torch.nn.functional.avg_pool{1|2|3}d error message does not match what is described in the documentation,https://github.com/pytorch/pytorch/issues/83135,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2022-08-10T12:57:45Z +issue,7686,Only one thread is used on macOS (super slow on CPU),https://github.com/pytorch/pytorch/issues/7686,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, module: multithreading",scan when time allows,2022-08-10T10:18:22Z +issue,82793,Profiling results on CPU is not reliable,https://github.com/pytorch/pytorch/issues/82793,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2022-08-10T05:46:13Z +issue,82382,Problems in built-from-source pytorch with USE_DEPLOY=1 in Ubuntu,https://github.com/pytorch/pytorch/issues/82382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deploy,scan when time allows,2022-08-10T00:59:28Z +issue,82802,"[ROCm] build instruction is haphazard missing information unclear, build does not work",https://github.com/pytorch/pytorch/issues/82802,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rocm",scan when time allows,2022-08-09T18:49:48Z +issue,75703,Let's host NVIDIA dependencies in our own S3,https://github.com/pytorch/pytorch/issues/75703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-08-09T17:16:19Z +issue,10043,Sparse tensor use cases,https://github.com/pytorch/pytorch/issues/10043,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-08-09T16:05:28Z +issue,82577,RFC: Add flag for RNN decomposition to all RNN modules,https://github.com/pytorch/pytorch/issues/82577,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2022-08-09T14:41:32Z +issue,82494,Implement torch.clamp() on sparse tensors with SparseCPU backend,https://github.com/pytorch/pytorch/issues/82494,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-08-08T19:46:39Z +issue,82727,DecompositionInterpreter creates invalid graphs for FX graph modules created with torch.fx.symbolic_trace,https://github.com/pytorch/pytorch/issues/82727,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-08-08T15:57:43Z +issue,82926,"Slice operation on ""ragged"" dimension in NestedTensor",https://github.com/pytorch/pytorch/issues/82926,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2022-08-08T15:57:16Z +issue,73160,Preprocessing function for backend coreml is not registered,https://github.com/pytorch/pytorch/issues/73160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,scan when time allows,2022-08-08T09:26:22Z +issue,39917,build QT program use libtorch-cxx11-abi-shared-with-deps-1.5.0+cu101 ok with CPU but error with cuda GPU,https://github.com/pytorch/pytorch/issues/39917,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-08-08T05:36:42Z +issue,82871,GPU arch 8.6 is not covered by the `TORCH_CUDA_ARCH_LIST = All` option,https://github.com/pytorch/pytorch/issues/82871,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2022-08-05T21:34:35Z +issue,81680,Provide an option to disable CUDA_GCC_VERSIONS,https://github.com/pytorch/pytorch/issues/81680,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-08-05T19:30:25Z +issue,82545,Incorrect CPU implementation of CTCLoss backward step,https://github.com/pytorch/pytorch/issues/82545,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: loss",scan when time allows,2022-08-05T18:56:58Z +issue,7857,Feature Request: Logistic Distribution,https://github.com/pytorch/pytorch/issues/7857,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-08-05T17:19:53Z +issue,82669,unittest.subTest and way to selectively mark subTests as expected failures,https://github.com/pytorch/pytorch/issues/82669,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-08-05T15:12:52Z +issue,81297,Files downloaded with torch.hub should respect umask,https://github.com/pytorch/pytorch/issues/81297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2022-08-05T14:50:17Z +issue,82823,getDLContext in DLConvertor.h cannot be found,https://github.com/pytorch/pytorch/issues/82823,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dlpack,scan when time allows,2022-08-05T13:30:49Z +issue,82789,[LibTorch] the C++ api needs detailed error reports like pytorch,https://github.com/pytorch/pytorch/issues/82789,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2022-08-05T13:04:56Z +issue,82465,Does torch.utils.checkpoint compatible with torch.cuda.make_graphed_callables?,https://github.com/pytorch/pytorch/issues/82465,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: cuda graphs",scan when time allows,2022-08-04T13:53:46Z +issue,53438,Pipe should automatically build appropriate partitions.,https://github.com/pytorch/pytorch/issues/53438,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-08-04T02:03:20Z +issue,82764,Subclass of Tensor doesn't support __format__,https://github.com/pytorch/pytorch/issues/82764,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-08-03T23:24:28Z +issue,82761,torch.Tensor.bag() should automatically implement bagging,https://github.com/pytorch/pytorch/issues/82761,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-08-03T23:23:59Z +issue,82710,build fail when using lto with gcc,https://github.com/pytorch/pytorch/issues/82710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-08-03T21:55:47Z +issue,82751,Refactor how errors decide whether to append C++ stacktrace,https://github.com/pytorch/pytorch/issues/82751,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-08-03T20:28:56Z +issue,82610,contiguous() not work for rank 1 length 1 tensor.,https://github.com/pytorch/pytorch/issues/82610,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dlpack,scan when time allows,2022-08-03T16:42:23Z +issue,82542,Is there Doc that explains how to call an extension op in another extension implementation?,https://github.com/pytorch/pytorch/issues/82542,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2022-08-03T15:18:05Z +issue,38310,Compiling errors when trying to cross-compile the C++ API for RTOS (QNX),https://github.com/pytorch/pytorch/issues/38310,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2022-08-02T20:39:53Z +issue,82583,DataLoader parameter pin_memory_device should accept torch.device type,https://github.com/pytorch/pytorch/issues/82583,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-08-02T18:38:44Z +issue,82303,Race condition between torch.tensor's view and /= (/= returns incorrect result),https://github.com/pytorch/pytorch/issues/82303,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: partial aliasing,scan when time allows,2022-08-02T17:42:20Z +issue,82510,Complex addition result in NaN when it shouldn't,https://github.com/pytorch/pytorch/issues/82510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",scan when time allows,2022-08-02T17:29:40Z +issue,78759,Add `inplace_view` tag for `resize_`,https://github.com/pytorch/pytorch/issues/78759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2022-08-02T17:06:26Z +issue,56500,Marking variable not required in backward calculation if it is not needed,https://github.com/pytorch/pytorch/issues/56500,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-08-02T15:45:19Z +issue,78482,RuntimeError: Event device type CUDA does not match blocking stream’s device type CPU,https://github.com/pytorch/pytorch/issues/78482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda, module: tests",scan when time allows,2022-08-02T01:34:14Z +issue,72749,"pytorchmergebot doesn't react to comments left from ""files"" tab",https://github.com/pytorch/pytorch/issues/72749,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-08-01T18:08:27Z +issue,81955,Investigate adding shell linter/checker to CI,https://github.com/pytorch/pytorch/issues/81955,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-08-01T18:05:45Z +issue,81945,Investigate adding Dockerfile linter hadolint to CI,https://github.com/pytorch/pytorch/issues/81945,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-08-01T18:01:18Z +issue,71896,CUDA Graph API Improvement,https://github.com/pytorch/pytorch/issues/71896,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2022-08-01T16:51:10Z +issue,82145,set_grad_enabled not respected when running on a web server,https://github.com/pytorch/pytorch/issues/82145,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: dependency bug",scan when time allows,2022-08-01T16:40:26Z +issue,39570,Dataloader._shutdown_workers hangs,https://github.com/pytorch/pytorch/issues/39570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2022-08-01T09:10:11Z +issue,82377,masked_scatter_ is very lacking,https://github.com/pytorch/pytorch/issues/82377,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: scatter & gather ops",scan when time allows,2022-07-30T17:11:05Z +issue,26957,Batched Dataloader,https://github.com/pytorch/pytorch/issues/26957,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-07-30T16:11:13Z +issue,36914,Problem with c10/utils/variant.h,https://github.com/pytorch/pytorch/issues/36914,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-07-29T12:17:40Z +issue,50328,RTX3090 performs no better than 1080ti,https://github.com/pytorch/pytorch/issues/50328,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2022-07-29T06:26:36Z +issue,69663,"After updated to pytorch1.10.0 cuda11.1, NCCL is not available",https://github.com/pytorch/pytorch/issues/69663,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-07-29T05:22:15Z +issue,82357,ufmt and flake8 lints race,https://github.com/pytorch/pytorch/issues/82357,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-28T20:47:00Z +issue,82419,The torch::deploy document is not updated,https://github.com/pytorch/pytorch/issues/82419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deploy,scan when time allows,2022-07-28T20:46:53Z +issue,82081,Could be clearer that Cross Entropy takes logits as input,https://github.com/pytorch/pytorch/issues/82081,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: loss, module: nn",scan when time allows,2022-07-28T16:11:16Z +issue,54408,It is strange that PyTorch is slow on RTX 3090,https://github.com/pytorch/pytorch/issues/54408,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2022-07-28T07:40:13Z +issue,69025,[subgraph_rewriter] Support skipping matching for some ops,https://github.com/pytorch/pytorch/issues/69025,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-27T22:44:23Z +issue,69577,[subgraph_rewriter] Add match_filter for the matches,https://github.com/pytorch/pytorch/issues/69577,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-27T22:30:53Z +issue,81541,[vulkan]compiling VulkanOpContext.cpp with some errors,https://github.com/pytorch/pytorch/issues/81541,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vulkan",scan when time allows,2022-07-27T20:47:31Z +issue,81669,Register refs for CompositeImplicitAutograd ops as decompositions,https://github.com/pytorch/pytorch/issues/81669,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-07-27T20:34:13Z +issue,82185,DataLoader: `pin_memory` should respect object attributes before object collection type,https://github.com/pytorch/pytorch/issues/82185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-07-27T15:48:00Z +issue,82276,DEBUG=1 env var doesn't actually set DEBUG preprocessor macro,https://github.com/pytorch/pytorch/issues/82276,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-07-27T13:33:20Z +issue,82242,logspace inconsistently casts inputs to int before performing computation,https://github.com/pytorch/pytorch/issues/82242,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2022-07-26T23:56:51Z +issue,82230,logspace and linspace off by one on cuda for integer dtypes for some inputs,https://github.com/pytorch/pytorch/issues/82230,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2022-07-26T23:56:34Z +issue,82239,primtorch refs should be composite compliant,https://github.com/pytorch/pytorch/issues/82239,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: primtorch",scan when time allows,2022-07-26T23:55:06Z +issue,81568,Improve interaction of PyTorch downstream libraries and torchdeploy,https://github.com/pytorch/pytorch/issues/81568,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deploy,scan when time allows,2022-07-26T21:03:57Z +issue,82197,Provide error handling for ops that don't yet support Dynamic Shape,https://github.com/pytorch/pytorch/issues/82197,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-26T20:30:01Z +issue,81996,linspace cpu and sometimes cuda is wrong on integral types,https://github.com/pytorch/pytorch/issues/81996,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: edge cases, module: python frontend",scan when time allows,2022-07-26T17:54:25Z +issue,82212,"RuntimeError: ""reflection_pad2d"" not implemented for 'Half' in autocast enabled region",https://github.com/pytorch/pytorch/issues/82212,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-07-26T16:48:41Z +issue,80061,pow CUDA tensor raised to CPU scalar tensor result can't backward properly,https://github.com/pytorch/pytorch/issues/80061,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-07-26T15:09:20Z +issue,82200,"model.to(device) takes time forever on A40-8Q, NVIDIA. cuda11.1, torch1.9.1.",https://github.com/pytorch/pytorch/issues/82200,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-07-26T14:07:22Z +issue,634,Feature Request: NegativeSampling and HierarchicalSoftmax loss functions,https://github.com/pytorch/pytorch/issues/634,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn, module: primtorch",scan when time allows,2022-07-26T04:14:39Z +issue,35798,Dropout on sparse tensors,https://github.com/pytorch/pytorch/issues/35798,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sparse",scan when time allows,2022-07-26T04:03:48Z +issue,76206,Update NCCL to 2.12,https://github.com/pytorch/pytorch/issues/76206,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: distributed",scan when time allows,2022-07-26T01:59:15Z +issue,82033,[PyTorch/XLA] Improve the XLA PR landing process,https://github.com/pytorch/pytorch/issues/82033,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: xla",scan when time allows,2022-07-25T23:19:53Z +issue,76785,"`addr, baddmm, dist, l1_loss` will backward fail when input tensors have different dtypes",https://github.com/pytorch/pytorch/issues/76785,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex",scan when time allows,2022-07-25T23:11:47Z +issue,80762,`atan2` will gradcheck fail when `other` is a tensor with `int8` dtype,https://github.com/pytorch/pytorch/issues/80762,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",scan when time allows,2022-07-25T23:10:16Z +issue,68208,Make `torch.Tensor.view` support autograd for appropriate cases,https://github.com/pytorch/pytorch/issues/68208,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: viewing and reshaping",scan when time allows,2022-07-25T22:53:47Z +issue,74677,`check_batched_forward_grad` fails for `torch.norm` and related ops,https://github.com/pytorch/pytorch/issues/74677,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: batching, module: norms and normalization",scan when time allows,2022-07-25T22:44:31Z +issue,75586,Add ZeroTensor support for `mm`,https://github.com/pytorch/pytorch/issues/75586,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: performance",scan when time allows,2022-07-25T22:43:54Z +issue,63623,Expose a SavedTensorsHooks nn.Module for users to register saved tensors hooks,https://github.com/pytorch/pytorch/issues/63623,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2022-07-25T22:22:35Z +issue,81417,Add a check to detect mutation of the inputs during backward,https://github.com/pytorch/pytorch/issues/81417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: molly-guard",scan when time allows,2022-07-25T22:09:55Z +issue,82061,Documentation for torch.cuda.Event(blocking=True) is wrong,https://github.com/pytorch/pytorch/issues/82061,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2022-07-25T21:10:13Z +issue,82132,External libraries cannot have a requirements.txt that needs to install a cpp_extension,https://github.com/pytorch/pytorch/issues/82132,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2022-07-25T21:06:43Z +issue,82156,[Checkpoint] Support multiple unpack in saved tensor hooks,https://github.com/pytorch/pytorch/issues/82156,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2022-07-25T21:06:02Z +issue,82140,Stop manually binding sparse factory functions,https://github.com/pytorch/pytorch/issues/82140,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-07-25T20:57:57Z +issue,25310,Handling of packed_sequence by activation functions and linear layers,https://github.com/pytorch/pytorch/issues/25310,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2022-07-25T20:39:06Z +issue,82093,"When using libtorch v1.10.2, calling at::slow_conv_dilated3d directly returns wrong results on cpu backend",https://github.com/pytorch/pytorch/issues/82093,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cpp",scan when time allows,2022-07-25T20:17:42Z +issue,59629,Potentially misleading note in documentation for PackedSequence,https://github.com/pytorch/pytorch/issues/59629,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nestedtensor, module: nn, module: rnn",scan when time allows,2022-07-25T20:12:01Z +issue,81768,TestTagsCPU.test_tags__refs_constant_pad_nd_cpu_float32 flaky with dynamo & pytest,https://github.com/pytorch/pytorch/issues/81768,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-07-25T18:27:32Z +issue,38767,weight_decay in Adam is not an L2 Penalty,https://github.com/pytorch/pytorch/issues/38767,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-07-25T18:06:24Z +issue,81982,test_sparse_matmul_cpu_complex128 fails on my local copy,https://github.com/pytorch/pytorch/issues/81982,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: sparse",scan when time allows,2022-07-25T16:20:12Z +issue,81979,test_sparse_spdiags_cpu_bool fails on my local working copy,https://github.com/pytorch/pytorch/issues/81979,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-07-25T16:20:10Z +issue,82095,torch.concat type hints fail for keyword argument,https://github.com/pytorch/pytorch/issues/82095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2022-07-25T15:30:12Z +issue,82088,linear.matrix_power is not composite compliant,https://github.com/pytorch/pytorch/issues/82088,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-07-25T14:12:12Z +issue,81935,implement sym_numel,https://github.com/pytorch/pytorch/issues/81935,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-25T12:39:43Z +issue,82041,[Misleading] The doc started using Tensorflow terminology in the document to explain how to use the Pytorch code.,https://github.com/pytorch/pytorch/issues/82041,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-07-24T16:35:11Z +issue,81988,Unify c10::Event and at::cuda::CUDAEvent,https://github.com/pytorch/pytorch/issues/81988,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-07-24T16:31:39Z +issue,36107,Allow grid_sample to accept pixel units (absolute coordinates),https://github.com/pytorch/pytorch/issues/36107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: vision",scan when time allows,2022-07-23T13:51:57Z +issue,73141,libtorch: `DistributedRandomSampler` uses the same random order in every epoch,https://github.com/pytorch/pytorch/issues/73141,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2022-07-22T18:57:38Z +issue,81983,backwards compatibility ALLOWLIST is misused,https://github.com/pytorch/pytorch/issues/81983,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-07-22T17:17:34Z +issue,73469,torch.fx failed when tracing functions from other Libs.,https://github.com/pytorch/pytorch/issues/73469,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-22T09:00:24Z +issue,81959,Overloading multiple signatures for a single ref,https://github.com/pytorch/pytorch/issues/81959,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-07-22T03:37:52Z +issue,81654,[packaging] Conda install missing python local version label (+cu123 or +cpu),https://github.com/pytorch/pytorch/issues/81654,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2022-07-22T01:14:19Z +issue,81650,RFC: auto-generated plain Tensor argument only sparse primitives,https://github.com/pytorch/pytorch/issues/81650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-07-22T01:12:02Z +issue,81801,"Rename DispatchKey Dense/Sparse/etc to DenseFunctionality/SparseFunctionality, use original name for alias",https://github.com/pytorch/pytorch/issues/81801,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-07-22T01:10:31Z +issue,81938,"Devirtualize sym_sizes, virtualize sym_sizes_custom",https://github.com/pytorch/pytorch/issues/81938,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-22T01:07:04Z +issue,81943,Investigate if it's okay to throw a RuntimeError instead of TypeError here : https://github.com/pytorch/pytorch/pull/79560/files#diff-415017bcad4fa6cd6d3dfe5f6ea1caffcd7122b46b8c1e4825f7d889efc80a62R1816,https://github.com/pytorch/pytorch/issues/81943,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-22T01:06:52Z +issue,81937,Add more autograd tests with symints,https://github.com/pytorch/pytorch/issues/81937,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-22T01:05:09Z +issue,81625,[bug] the output shape from torch::mean and torch::var is different in libtorch,https://github.com/pytorch/pytorch/issues/81625,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-07-22T01:02:36Z +issue,81649,Idiom for PrimTorch refs for Tensor methods,https://github.com/pytorch/pytorch/issues/81649,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-07-22T00:52:24Z +issue,81932,Make sure we always redispatch through a dispatcher for all SymInt ops,https://github.com/pytorch/pytorch/issues/81932,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-21T22:37:03Z +issue,79138,[AUTOGRAD] support implicit reductions with SymInts in autograd.,https://github.com/pytorch/pytorch/issues/79138,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-21T22:32:45Z +issue,39820,I would like to install pytorch-nightly from requirements.txt,https://github.com/pytorch/pytorch/issues/39820,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-21T13:12:38Z +issue,8276,[Installation]: Support conda/pip install with ppc64le(power8),https://github.com/pytorch/pytorch/issues/8276,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: power",scan when time allows,2022-07-21T13:12:38Z +issue,81448,torch.utils.checkpoint optimization opportunity,https://github.com/pytorch/pytorch/issues/81448,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-07-21T10:10:26Z +issue,80765,Let torch.utils.tensorboard support multiprocessing,https://github.com/pytorch/pytorch/issues/80765,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: tensorboard",scan when time allows,2022-07-21T02:09:04Z +issue,80966,[Prims+NVFuser] Aten2Aten decomp hurting performance,https://github.com/pytorch/pytorch/issues/80966,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-07-20T23:14:10Z +issue,81307,[Prims+NvFuser] Non-fusible ops Tracker,https://github.com/pytorch/pytorch/issues/81307,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-07-20T22:53:51Z +issue,63076,Tracing models all the way down to basic functions,https://github.com/pytorch/pytorch/issues/63076,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T22:00:16Z +issue,68301,[FX] [BUG] Tensor.{inplace_method}_(.) is eliminated as dead code,https://github.com/pytorch/pytorch/issues/68301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:58:46Z +issue,77332,"wrong overload resolved for `torch.mul(x, 4)` in `__torch_dispatch__`",https://github.com/pytorch/pytorch/issues/77332,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-07-20T21:52:22Z +issue,81244,CapabilityBasedPartitioner does not work correctly with mutating operations,https://github.com/pytorch/pytorch/issues/81244,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: capabilitybasedpartitioner, module: fx",scan when time allows,2022-07-20T21:48:35Z +issue,81539,CapabilityBasedPartitioner treats non-compute ops inconsistently,https://github.com/pytorch/pytorch/issues/81539,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: capabilitybasedpartitioner, module: fx, module: fx.passes",scan when time allows,2022-07-20T21:47:52Z +issue,81229,torch.fx.node.map_aggregate and torch.utils._pytree.tree_map do the same thing,https://github.com/pytorch/pytorch/issues/81229,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: pytree",scan when time allows,2022-07-20T21:25:29Z +issue,80427,Torch fx print line number of each node,https://github.com/pytorch/pytorch/issues/80427,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:29Z +issue,80903,make_fx doesn't work with truly dynamic argument functions (e.g. fx.Interpreter),https://github.com/pytorch/pytorch/issues/80903,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:28Z +issue,80172,Torch.fx: add reporting of the name of a module not found during tracing,https://github.com/pytorch/pytorch/issues/80172,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:28Z +issue,78435,fx.Tracer with param_shapes_constant=True not working for RobertaForMaskedLM,https://github.com/pytorch/pytorch/issues/78435,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:26Z +issue,64997,FX failed to normalize op `block_diag` and `broadcast_tensors`,https://github.com/pytorch/pytorch/issues/64997,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:25Z +issue,78201,nn.Sequential causes fx.replace_pattern to not find any match.,https://github.com/pytorch/pytorch/issues/78201,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:24Z +issue,74058,torch.fx.symbolic_trace is non-deterministic,https://github.com/pytorch/pytorch/issues/74058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:24Z +issue,76186,fx: cannot find module when using apex.amp,https://github.com/pytorch/pytorch/issues/76186,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:21Z +issue,77016,Where is fx2trt fx to tensorrt tool?,https://github.com/pytorch/pytorch/issues/77016,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:20Z +issue,75342,`torch.fx.operator_schemas.normalize_function` is too permissive,https://github.com/pytorch/pytorch/issues/75342,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:20Z +issue,75090,torch.fx deepcopy bug,https://github.com/pytorch/pytorch/issues/75090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:18Z +issue,74218,Cannot get op through FX when using nn.Sequential,https://github.com/pytorch/pytorch/issues/74218,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:18Z +issue,73992,[FX] Tensor constants are not lifted to attributes in direct Graph construction,https://github.com/pytorch/pytorch/issues/73992,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:18Z +issue,73304,"fx.symbolic_trace docs says that it ignores different values of concrete_args, but it doesn't",https://github.com/pytorch/pytorch/issues/73304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:16Z +issue,71715,[FX] Support call_method in NormalizeArgs,https://github.com/pytorch/pytorch/issues/71715,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:15Z +issue,70892,Memory leak while training model generated by torch.fx.symbolic_trace() in data parallel mode,https://github.com/pytorch/pytorch/issues/70892,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:15Z +issue,70485,Cannot run FX tracer on a vision transformer model,https://github.com/pytorch/pytorch/issues/70485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:15Z +issue,70449,[FX] Detect attribute mutation during tracing,https://github.com/pytorch/pytorch/issues/70449,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:14Z +issue,69095,[FX] `concrete_args` with unpacking breaks `fx.Interpreter`,https://github.com/pytorch/pytorch/issues/69095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:14Z +issue,68513,torch.fx cannot trace torch.Size() properly,https://github.com/pytorch/pytorch/issues/68513,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:13Z +issue,68190,[FX][docs] Document gotcha about `training` flag,https://github.com/pytorch/pytorch/issues/68190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:13Z +issue,67662,"""self.graph.owning_module not set for purity check"" error when trying to remove a node from torch.fx.graph",https://github.com/pytorch/pytorch/issues/67662,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:11Z +issue,67402,[Documentation] Incomplete FX module,https://github.com/pytorch/pytorch/issues/67402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:11Z +issue,67543,[FX] Support `int` and similar variants in FX,https://github.com/pytorch/pytorch/issues/67543,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:04Z +issue,67364,[FX] FX torchvision tests fail with torchvision 0.10,https://github.com/pytorch/pytorch/issues/67364,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:03Z +issue,67030,fx graph mode quantizing error,https://github.com/pytorch/pytorch/issues/67030,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:02Z +issue,66335,`torch.fx.symbolic_trace()` loses module class information,https://github.com/pytorch/pytorch/issues/66335,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:02Z +issue,66197,`torch.fx.replace_pattern` doesn't work with untraceable wrapped functions,https://github.com/pytorch/pytorch/issues/66197,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:01Z +issue,65096,[FX] `passes.split_module` doesn't have documentation,https://github.com/pytorch/pytorch/issues/65096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:01Z +issue,64497,FX tracing on Fairseq-based roberta fails.,https://github.com/pytorch/pytorch/issues/64497,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:25:00Z +issue,63815,[FX] Documentation for Graph.flatten_inps and Graph.unflatten_outs,https://github.com/pytorch/pytorch/issues/63815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:24:58Z +issue,63490,[FX] Data type of target should be validated on Node construction and lint,https://github.com/pytorch/pytorch/issues/63490,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:24:58Z +issue,44851,fx: Python range function used on tensor shape is not symbolically traceable,https://github.com/pytorch/pytorch/issues/44851,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:24:31Z +issue,44049,[FX] Can't symbolically trace callsites to ScriptFunction,https://github.com/pytorch/pytorch/issues/44049,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:40Z +issue,43679,[FX] List unpacking is broken,https://github.com/pytorch/pytorch/issues/43679,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:40Z +issue,43659,Exploring supporting some cases of `__bool__` on Proxy,https://github.com/pytorch/pytorch/issues/43659,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:40Z +issue,44901,for loop can't be symbolically traced,https://github.com/pytorch/pytorch/issues/44901,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:39Z +issue,44316,[FX] Schema normalization tooling,https://github.com/pytorch/pytorch/issues/44316,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: internals",scan when time allows,2022-07-20T21:23:39Z +issue,45534,"[FX] Make tracer return a Graph, not a GraphModule",https://github.com/pytorch/pytorch/issues/45534,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:38Z +issue,45220,FX should preserve type annotations and not break TorchScript-ability,https://github.com/pytorch/pytorch/issues/45220,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, oncall: jit",scan when time allows,2022-07-20T21:23:38Z +issue,45683,[FX] Configurability of nested module representation,https://github.com/pytorch/pytorch/issues/45683,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:37Z +issue,45682,[FX] Allow customization of the behavior of Proxy,https://github.com/pytorch/pytorch/issues/45682,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:37Z +issue,47481,[fx] scripting a model with tensor list as input fails,https://github.com/pytorch/pytorch/issues/47481,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:36Z +issue,46210,[FX]Add support for len() on Proxy,https://github.com/pytorch/pytorch/issues/46210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:36Z +issue,49265,[FX] Update `placeholder` docs to better reflect the role of `node` and `target`,https://github.com/pytorch/pytorch/issues/49265,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:35Z +issue,48813,inception_v3 is not symbolically traceable,https://github.com/pytorch/pytorch/issues/48813,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:35Z +issue,48209,[FX] Write an inference optimization pass,https://github.com/pytorch/pytorch/issues/48209,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:35Z +issue,50143,FX call_function does not work if target is a class member function,https://github.com/pytorch/pytorch/issues/50143,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:34Z +issue,49618,Support scope in fx Node,https://github.com/pytorch/pytorch/issues/49618,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:34Z +issue,50177,[FX] Static specialization of callsites (e.g. chunk),https://github.com/pytorch/pytorch/issues/50177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:33Z +issue,52538,Helper methods not preserved during AST rewrite,https://github.com/pytorch/pytorch/issues/52538,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:32Z +issue,52022,[FX] Interpreter doesn't GC values,https://github.com/pytorch/pytorch/issues/52022,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:32Z +issue,52564,Submodules not rewritten during AST rewrite,https://github.com/pytorch/pytorch/issues/52564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:31Z +issue,53678,[FX] Regression from 1.8: FX can no longer trace functions where the first element of an int list is a Proxy,https://github.com/pytorch/pytorch/issues/53678,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:30Z +issue,53121,[fx] Handle models with Optional inputs,https://github.com/pytorch/pytorch/issues/53121,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:30Z +issue,53794,[fx] GraphModule __init__ when provided a Module w/ inconsistent buffer handling,https://github.com/pytorch/pytorch/issues/53794,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:29Z +issue,53732,[FX] Allow customizing code generation for certain Nodes,https://github.com/pytorch/pytorch/issues/53732,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:29Z +issue,54071,[FX] Add guards in files in `test/fx` to make sure they're not run directly,https://github.com/pytorch/pytorch/issues/54071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:28Z +issue,53935,SubgraphRewriter doesn't handle tensor constants in replacement graph,https://github.com/pytorch/pytorch/issues/53935,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:28Z +issue,54602,if cannot be symbolically traced,https://github.com/pytorch/pytorch/issues/54602,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:27Z +issue,54574,"Add OpInfo metadata for ""is_torch_functional"" and a skip for these ops in TestOperatorSignatures.test_get_torch_func_signature_exhaustive",https://github.com/pytorch/pytorch/issues/54574,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: tests",scan when time allows,2022-07-20T21:23:27Z +issue,54503,torch.fx may have problem for annotations,https://github.com/pytorch/pytorch/issues/54503,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:27Z +issue,54899,[fx] forward references in annotations do not produce correct source code,https://github.com/pytorch/pytorch/issues/54899,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: fx",scan when time allows,2022-07-20T21:23:26Z +issue,54785,[FX] Cannot symbolically trace class method,https://github.com/pytorch/pytorch/issues/54785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:26Z +issue,54913,FX tracer doesn't support returning a new instance of `object` subclass from module forward,https://github.com/pytorch/pytorch/issues/54913,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:25Z +issue,55395,[FX][testing] Test tracing into all the standard `torch.nn.functional` instances,https://github.com/pytorch/pytorch/issues/55395,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:24Z +issue,55394,[FX][testing] Test tracing into all the standard `nn.Module` instances,https://github.com/pytorch/pytorch/issues/55394,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:24Z +issue,55198,[FX] to_folder breaks with qualified type name,https://github.com/pytorch/pytorch/issues/55198,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: fx",scan when time allows,2022-07-20T21:23:24Z +issue,55400,[FX][testing] Add symbolic tracing tests for torchtext,https://github.com/pytorch/pytorch/issues/55400,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:23Z +issue,55397,[FX][testing] Run symbolic tracing tests over torch benchmark in CI,https://github.com/pytorch/pytorch/issues/55397,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:23Z +issue,55668,GraphModule.to_folder generates code with syntax error (imports in the middle of a class) [torch.fx],https://github.com/pytorch/pytorch/issues/55668,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:22Z +issue,55544,[FX] See if we can type-annotate attribute accesses on proxy,https://github.com/pytorch/pytorch/issues/55544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:22Z +issue,55401,[FX][testing] Add symbolic tracing tests for torchaudio,https://github.com/pytorch/pytorch/issues/55401,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:22Z +issue,55694,symbolic tracing does not support patterns with multiple arguments?,https://github.com/pytorch/pytorch/issues/55694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:21Z +issue,56778,can't find user's fuser method in quantization.fx.fusion_patterns.py when run fx graph qat.,https://github.com/pytorch/pytorch/issues/56778,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:20Z +issue,56688,Torch FX does not work with torchvision model,https://github.com/pytorch/pytorch/issues/56688,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:20Z +issue,56191,[FX] Issues with names of functions in user packages,https://github.com/pytorch/pytorch/issues/56191,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: fx",scan when time allows,2022-07-20T21:23:20Z +issue,56938,remove the dependence on node names in fx graph mode quant,https://github.com/pytorch/pytorch/issues/56938,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:19Z +issue,58051,quantize_per_tensor not symbolically traceable with FX if scale+zp are proxied,https://github.com/pytorch/pytorch/issues/58051,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:18Z +issue,57124,Allow `ScriptFunction`s to be symbolically traced,https://github.com/pytorch/pytorch/issues/57124,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:18Z +issue,58402,fx OpInfo tests don't work with nn.functional test names,https://github.com/pytorch/pytorch/issues/58402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:17Z +issue,62021,[FX] Ability to provide autowrap function to Tracer init,https://github.com/pytorch/pytorch/issues/62021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:14Z +issue,62554,`torch.fx.node.normalized_arguments` does not pass `normalize_to_only_use_kwargs`,https://github.com/pytorch/pytorch/issues/62554,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: fx",scan when time allows,2022-07-20T21:23:13Z +issue,71404,[feature request] [discussion] Generalize / recommend behavior of reset_parameters (and potentially rename),https://github.com/pytorch/pytorch/issues/71404,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fx, module: nn",scan when time allows,2022-07-20T21:23:08Z +issue,69804,typing error in the signatures of `torch.unbind` when getting them using `torch.fx`.,https://github.com/pytorch/pytorch/issues/69804,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fx,scan when time allows,2022-07-20T21:23:08Z +issue,81682,Python operator registration API for subclasses,https://github.com/pytorch/pytorch/issues/81682,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: dispatch",scan when time allows,2022-07-20T13:03:59Z +issue,81722,Inconsistent naming convention for end of enum in DispatchKey,https://github.com/pytorch/pytorch/issues/81722,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2022-07-20T02:19:29Z +issue,80985,Nested tensor: Support Noncontiguous Buffer,https://github.com/pytorch/pytorch/issues/80985,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-07-20T00:07:11Z +issue,69685,Wrong PyTorch version in Docker image,https://github.com/pytorch/pytorch/issues/69685,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2022-07-19T21:36:25Z +issue,58036,[feature request] torch.as_tensor to support any object that NumPy's asarray or array can consume (consume __array_interface__),https://github.com/pytorch/pytorch/issues/58036,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2022-07-19T16:53:25Z +issue,42444,torch.random.randperm stuck in multiprocess,https://github.com/pytorch/pytorch/issues/42444,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: random",scan when time allows,2022-07-19T12:00:12Z +issue,80986,"[Prims+NVFuser] nvFuser running into ""Tensors of type SparseTensorImpl do not have strides""",https://github.com/pytorch/pytorch/issues/80986,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-07-19T04:02:56Z +issue,79164,PyTorch/XLA's DDP XLABackend is broken by upstream change,https://github.com/pytorch/pytorch/issues/79164,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: xla, oncall: distributed",scan when time allows,2022-07-18T07:51:33Z +issue,81543,Linking pytorch libraries causes sstream behavior to be overridden globally,https://github.com/pytorch/pytorch/issues/81543,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-07-18T04:21:48Z +issue,81544,Torch does not build with Lazy TS disabled,https://github.com/pytorch/pytorch/issues/81544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-07-15T12:15:30Z +issue,81531,CapabilityBasedPartitioner doesn't support horizontal (vertical?) fusion,https://github.com/pytorch/pytorch/issues/81531,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-07-15T12:04:56Z +issue,81459,Missing corner case handling in ATen ctc_loss implementation,https://github.com/pytorch/pytorch/issues/81459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: loss",scan when time allows,2022-07-15T11:58:33Z +issue,76743,Supporting torch.tensor.apply_ over GPU,https://github.com/pytorch/pytorch/issues/76743,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-07-14T22:06:49Z +issue,75625,[ONNX] Support unit tests in scripting that we already support in tracing,https://github.com/pytorch/pytorch/issues/75625,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2022-07-14T18:40:48Z +issue,80377,Modify _add_docstr to also set the correct module for the APIs,https://github.com/pytorch/pytorch/issues/80377,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-07-14T17:27:32Z +issue,34977,`conv2d` is slow with specific shapes of channels_last tensors,https://github.com/pytorch/pytorch/issues/34977,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: memory format, module: performance",scan when time allows,2022-07-14T13:50:00Z +issue,73537,Functionalization doesn't work when applied twice with wrapper tensor,https://github.com/pytorch/pytorch/issues/73537,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2022-07-14T04:08:04Z +issue,81385,"""Attempted to resize a view tensor to a larger size. This is not allowed in the functionalization pass"" reported on non view tensor",https://github.com/pytorch/pytorch/issues/81385,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2022-07-14T03:16:47Z +issue,81358,[Prims+NvFuser] Issue with aten.where.ScalarSelf,https://github.com/pytorch/pytorch/issues/81358,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-07-14T03:16:09Z +issue,81428,torch.stft does not normalize non-rectangular windows correctly,https://github.com/pytorch/pytorch/issues/81428,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: fft",scan when time allows,2022-07-14T03:09:40Z +issue,81413,torch.searchsorted error message and documentation is unclear,https://github.com/pytorch/pytorch/issues/81413,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sorting and selection",scan when time allows,2022-07-14T03:09:11Z +issue,75147,Dataloader hangs. Potential deadlock with `set_num_threads` in worker processes?,https://github.com/pytorch/pytorch/issues/75147,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: deadlock",scan when time allows,2022-07-13T18:48:06Z +issue,42303,frame_length reference in pytorch.stft documentation,https://github.com/pytorch/pytorch/issues/42303,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-07-13T18:38:15Z +issue,81046,three typing inconsistencies on Tensor methods,https://github.com/pytorch/pytorch/issues/81046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2022-07-13T00:46:06Z +issue,52279,[docs] Official apex -> torch.cuda.amp migration guide,https://github.com/pytorch/pytorch/issues/52279,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: docs",scan when time allows,2022-07-12T23:12:39Z +issue,81323,`torch.overrides.get_testing_overrides` does not function as intended for native tensor methods/operations,https://github.com/pytorch/pytorch/issues/81323,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2022-07-12T20:07:28Z +issue,80921,[primTorch] `|` operator does not work with FakeTensor in _refs,https://github.com/pytorch/pytorch/issues/80921,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: meta tensors, module: primtorch",scan when time allows,2022-07-12T19:22:26Z +issue,81186,grad not preserved during copying or pickling,https://github.com/pytorch/pytorch/issues/81186,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-07-12T18:21:07Z +issue,80808,"The result of doing a dot product between two vectors, using einsum, depends on another unrelated vector",https://github.com/pytorch/pytorch/issues/80808,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility,scan when time allows,2022-07-12T17:27:50Z +issue,80940,Wrong example of sliced computation in doc page Numerical Accuracy,https://github.com/pytorch/pytorch/issues/80940,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numerical-reproducibility",scan when time allows,2022-07-12T17:23:59Z +issue,81167,move bazel files out of pytorch repo root,https://github.com/pytorch/pytorch/issues/81167,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bazel,scan when time allows,2022-07-12T17:14:03Z +issue,81172,build libtorch with the same mkl as Matlab,https://github.com/pytorch/pytorch/issues/81172,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: mkl",scan when time allows,2022-07-12T17:13:54Z +issue,81259,Refactor linter adapters to avoid code duplication,https://github.com/pytorch/pytorch/issues/81259,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2022-07-12T16:46:10Z +issue,81240,Functionalization and fake tensors failure in torture test,https://github.com/pytorch/pytorch/issues/81240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: functionalization",scan when time allows,2022-07-12T16:44:50Z +issue,80929,"Interpolation artifacts when using nn.interpolate, trilinear mode for 3D label images",https://github.com/pytorch/pytorch/issues/80929,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn",scan when time allows,2022-07-12T07:24:28Z +issue,74991,Tensor Subclass not preserved for Tensor subclasses created via inheritance of a TensorImpl subclass,https://github.com/pytorch/pytorch/issues/74991,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-07-11T14:10:12Z +issue,48626,Missing super().__init__() call in nn.Module,https://github.com/pytorch/pytorch/issues/48626,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-07-11T13:30:51Z +issue,81110,CI: Run cpu tests in parallel processes?,https://github.com/pytorch/pytorch/issues/81110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-07-08T22:36:00Z +issue,51782,[discussion] [python] Generic way for generating module-wrappers for stateless parametrized functions,https://github.com/pytorch/pytorch/issues/51782,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-07-08T14:09:01Z +issue,80954,ExpandedWeights sometimes fail silently and doesn't compute .grad_sample attribute,https://github.com/pytorch/pytorch/issues/80954,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-07-07T22:10:50Z +issue,80951,ExpandedWeights can't handle modules with tied weights,https://github.com/pytorch/pytorch/issues/80951,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-07-07T22:09:32Z +issue,80932,Position embedding aware global circular convolution,https://github.com/pytorch/pytorch/issues/80932,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-07-07T21:58:01Z +issue,80187,[NVFuser] Investigate modules with bad performance relative to eager,https://github.com/pytorch/pytorch/issues/80187,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-07-07T20:16:01Z +issue,79202,[bug] Device dispatcher can choose CPU path for CUDA tensors.,https://github.com/pytorch/pytorch/issues/79202,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: codegen, module: dispatch",scan when time allows,2022-07-07T02:29:00Z +issue,75172,Runtime configuration to disable TORCH_WARN temporally?,https://github.com/pytorch/pytorch/issues/75172,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-07-07T01:42:23Z +issue,79977,[CI] Do we run all cpp tests on CI?,https://github.com/pytorch/pytorch/issues/79977,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-07-06T23:51:47Z +issue,80230,[Prims+NVFuser] Aten2Prim refs tracking items,https://github.com/pytorch/pytorch/issues/80230,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-07-06T23:19:32Z +issue,79604,Some unit tests are failing,https://github.com/pytorch/pytorch/issues/79604,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, oncall: distributed",scan when time allows,2022-07-06T17:50:21Z +issue,79787,Improve clarity by making sharding a static nightly update,https://github.com/pytorch/pytorch/issues/79787,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-07-06T16:34:58Z +issue,80577,2-dimensional arange,https://github.com/pytorch/pytorch/issues/80577,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor, module: tensor creation",scan when time allows,2022-07-06T15:39:03Z +issue,80804,`torch.renorm` gives wrong gradient for 0-valued input when `p` is even and `maxnorm=0`.,https://github.com/pytorch/pytorch/issues/80804,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",scan when time allows,2022-07-05T15:53:14Z +issue,80803,`hardshrink` gives wrong gradient for 0 input when `lambd` is 0.,https://github.com/pytorch/pytorch/issues/80803,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",scan when time allows,2022-07-05T15:53:02Z +issue,80551,NVFuser should extend caching to remove necessity for PrimTorch's executor to Provide Tensor Contiguity Info,https://github.com/pytorch/pytorch/issues/80551,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-07-05T08:59:24Z +issue,80549,Allow parameterization of Layouts,https://github.com/pytorch/pytorch/issues/80549,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: sparse",scan when time allows,2022-07-04T20:33:35Z +issue,80776,`torch.inverse()` crash in cuda,https://github.com/pytorch/pytorch/issues/80776,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: linear algebra",scan when time allows,2022-07-03T03:12:21Z +issue,80605,pad_sequence and pack_sequence should support length zero tensors,https://github.com/pytorch/pytorch/issues/80605,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2022-07-02T14:37:38Z +issue,80771,Anaconda is not a package manager,https://github.com/pytorch/pytorch/issues/80771,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-07-01T15:31:05Z +issue,80574,`bmm_sparse_cuda` kernel for `bfloat16`,https://github.com/pytorch/pytorch/issues/80574,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda, module: sparse",scan when time allows,2022-07-01T07:37:02Z +issue,79171,DistributedDataParallel `static_graph=True` fails to handle unused parameters,https://github.com/pytorch/pytorch/issues/79171,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-07-01T03:56:02Z +issue,68394,"INTERNAL ASSERT FAILED at ""../aten/src/ATen/MapAllocator.cpp"":323, please report a bug to PyTorch. unable to mmap 68 bytes from file : Cannot allocate memory (12) ---------------------------------------------------------------------------",https://github.com/pytorch/pytorch/issues/68394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multiprocessing",scan when time allows,2022-06-30T03:02:41Z +issue,80439,scatter_reduce choosed indices,https://github.com/pytorch/pytorch/issues/80439,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2022-06-29T22:45:09Z +issue,80458,Revisit OpInfo samples for nn.functional.max_poolNd,https://github.com/pytorch/pytorch/issues/80458,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling, module: testing",scan when time allows,2022-06-29T22:26:07Z +issue,64692,Remove the dependency with CUDA_LIBRARIES in TorchConfig.cmake,https://github.com/pytorch/pytorch/issues/64692,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-06-28T22:58:41Z +issue,49406,Bitcode enable for iOS,https://github.com/pytorch/pytorch/issues/49406,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: ios",scan when time allows,2022-06-28T20:37:13Z +issue,80167,AMP step() enforce synchronization,https://github.com/pytorch/pytorch/issues/80167,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-06-28T13:33:49Z +issue,80420,[DDP] output_device argument appears completely unused,https://github.com/pytorch/pytorch/issues/80420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-06-28T01:31:58Z +issue,80417,[c10d] Async object-based collectives,https://github.com/pytorch/pytorch/issues/80417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-06-28T01:09:11Z +issue,79629,Missing the time unit in duration time of DDP logging,https://github.com/pytorch/pytorch/issues/79629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ddp,scan when time allows,2022-06-27T23:16:10Z +issue,80204,`max_unpool` gives wrong gradient when `indices` has duplicate,https://github.com/pytorch/pytorch/issues/80204,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases, module: nn, module: pooling",scan when time allows,2022-06-27T19:21:53Z +issue,80154,[primTorch] No _refs support for torch.Tensor.requires_grad.__get__,https://github.com/pytorch/pytorch/issues/80154,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-06-27T18:22:44Z +issue,78744,Unable to programmatically update models using references from model.named_modules()...requires additional parsing,https://github.com/pytorch/pytorch/issues/78744,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-06-27T17:32:17Z +issue,80238,test_conv_backend tests OOMing in 10.2 slow_gradcheck CI,https://github.com/pytorch/pytorch/issues/80238,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: convolution, module: nn",scan when time allows,2022-06-27T17:18:23Z +issue,80022,memory leaking when doing all_to_all_single communication,https://github.com/pytorch/pytorch/issues/80022,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-06-27T17:00:37Z +issue,80189,[NVFuser] Investigate models without any fusion groups found,https://github.com/pytorch/pytorch/issues/80189,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-06-27T13:17:42Z +issue,80206,`soft_margin_loss` gives wrong gradient when `target` with dtype uint8,https://github.com/pytorch/pytorch/issues/80206,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: loss, module: nn",scan when time allows,2022-06-27T13:15:07Z +issue,80221,OpInfos for torch.ops.aten operations,https://github.com/pytorch/pytorch/issues/80221,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-06-27T13:14:31Z +issue,80226,Support tensor subclasses as `UninitializedParameter`s,https://github.com/pytorch/pytorch/issues/80226,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lazy, module: nn",scan when time allows,2022-06-27T13:14:14Z +issue,80231,[Prims+NVFuser] Supports 0-sized inputs,https://github.com/pytorch/pytorch/issues/80231,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nvfuser, module: primtorch",scan when time allows,2022-06-27T13:00:11Z +issue,79351,[bazel] ability to run gpu tests on gpu machines in RBE,https://github.com/pytorch/pytorch/issues/79351,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bazel,scan when time allows,2022-06-25T17:36:54Z +issue,80242,Write some torch.distributed.nn.* tests for the new dispatcher passable ops,https://github.com/pytorch/pytorch/issues/80242,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-06-24T20:34:53Z +issue,80241,Change c10d APIs in ProcessGroup to accept const std::vector&,https://github.com/pytorch/pytorch/issues/80241,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-06-24T20:32:25Z +issue,80134,[feature request] Add support for a custom DatasetFetcher in DataLoader,https://github.com/pytorch/pytorch/issues/80134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2022-06-24T15:37:24Z +issue,80208,F.binary_cross_entropy_with_logits unexpected behaviour,https://github.com/pytorch/pytorch/issues/80208,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2022-06-24T14:44:44Z +issue,80132,Expose more MAGMA backends for solve_triangular,https://github.com/pytorch/pytorch/issues/80132,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: magma",scan when time allows,2022-06-24T13:21:58Z +issue,79847,Multi-node training meets unknown error,https://github.com/pytorch/pytorch/issues/79847,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-06-23T23:54:36Z +issue,80157,Elliptic Functions and Integrals,https://github.com/pytorch/pytorch/issues/80157,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2022-06-23T18:51:29Z +issue,80118,"Allow a user provided ""test name - test time"" mapping file work with pytorch's test sharding mechanism",https://github.com/pytorch/pytorch/issues/80118,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-06-23T18:32:59Z +issue,80152,Orthogonal Polynomials,https://github.com/pytorch/pytorch/issues/80152,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2022-06-23T18:32:51Z +issue,80080,Complex support in DDP,https://github.com/pytorch/pytorch/issues/80080,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-06-23T15:14:49Z +issue,76838,Windows CUDA TTS tracking task,https://github.com/pytorch/pytorch/issues/76838,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-06-22T23:29:08Z +issue,79464,Iteration # 1-offset in DDP logging,https://github.com/pytorch/pytorch/issues/79464,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-06-22T17:29:35Z +issue,59592,Build failed with TBB,https://github.com/pytorch/pytorch/issues/59592,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: tbb",scan when time allows,2022-06-22T14:27:39Z +issue,42838,"RuntimeError: ""triangular_solve_cuda"" not implemented for 'Half'",https://github.com/pytorch/pytorch/issues/42838,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: half",scan when time allows,2022-06-22T13:06:30Z +issue,50723,[PyTorch Mobile] Can't use Vulkan backend,https://github.com/pytorch/pytorch/issues/50723,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: vulkan, oncall: mobile",scan when time allows,2022-06-22T06:46:23Z +issue,79997,Comprehensive documentation for Tensor indexing?,https://github.com/pytorch/pytorch/issues/79997,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: docs",scan when time allows,2022-06-22T06:26:53Z +issue,76656,Clarify dependency on NumPy (related to maskedtensor?),https://github.com/pytorch/pytorch/issues/76656,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-22T00:55:28Z +issue,77900,Floating point exception in slow_conv3d,https://github.com/pytorch/pytorch/issues/77900,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: edge cases",scan when time allows,2022-06-21T23:26:16Z +issue,77901,Heap corruption in slow_conv_transpose3d,https://github.com/pytorch/pytorch/issues/77901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: edge cases",scan when time allows,2022-06-21T23:26:02Z +issue,54035,Web support?,https://github.com/pytorch/pytorch/issues/54035,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: language binding,scan when time allows,2022-06-21T17:33:13Z +issue,25091,JavaScript (Web Assembly) target for trained models,https://github.com/pytorch/pytorch/issues/25091,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-21T17:32:46Z +issue,79853,[bazel] [ci] `//:lazy_tests` Could not run 'aten::mul.Tensor' with arguments from the 'Lazy' backend,https://github.com/pytorch/pytorch/issues/79853,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: cuda, module: lazy",scan when time allows,2022-06-21T13:41:15Z +issue,79867,All {view}_scatter variants should support all (or most) dtypes,https://github.com/pytorch/pytorch/issues/79867,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: functionalization,scan when time allows,2022-06-21T13:39:39Z +issue,79888,`CosineAnnealingWarmRestarts` does not update parameters added with `add_param_group`,https://github.com/pytorch/pytorch/issues/79888,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2022-06-21T13:33:12Z +issue,79893,Write lint for isGreen,https://github.com/pytorch/pytorch/issues/79893,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-06-21T13:32:25Z +issue,79895,Modify update-viable-strict GHA to use internal version of checkout,https://github.com/pytorch/pytorch/issues/79895,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-06-21T13:32:02Z +issue,79848,Automatically calculate output_shape of sequential model (or any other fCNN),https://github.com/pytorch/pytorch/issues/79848,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2022-06-21T13:27:59Z +issue,79788,Parameter.__deepcopy__ doesn't preserve view relationships,https://github.com/pytorch/pytorch/issues/79788,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: nn",scan when time allows,2022-06-21T13:17:00Z +issue,79542,Unable to use a parameter with torch.sparse_coo layout with DDP,https://github.com/pytorch/pytorch/issues/79542,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: sparse, oncall: distributed",scan when time allows,2022-06-21T00:19:09Z +issue,78367,Split up and reorganize RPC tests,https://github.com/pytorch/pytorch/issues/78367,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2022-06-20T23:49:06Z +issue,62094,"[FR] Support for non-Parameter, non-Buffer objects in state_dict",https://github.com/pytorch/pytorch/issues/62094,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-06-20T21:18:59Z +issue,79421,Mismatch in clang toolchain lead to binary incompatibilities on M1 between torch and torchvision,https://github.com/pytorch/pytorch/issues/79421,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: ci, module: macos",scan when time allows,2022-06-20T17:41:36Z +issue,77538,Failed to run on iOS - Couldn't find an operator for `aten::conv1d`,https://github.com/pytorch/pytorch/issues/77538,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ios,scan when time allows,2022-06-20T15:47:47Z +issue,36560,Allow `__array__` to automatically detach and move to CPU,https://github.com/pytorch/pytorch/issues/36560,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numpy",scan when time allows,2022-06-20T13:52:05Z +issue,77256,CrossEntropyLoss computes SoftMax always across the second dimension,https://github.com/pytorch/pytorch/issues/77256,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2022-06-20T09:31:01Z +issue,79705,gradgradcheck fails for torch.native_layer_norm,https://github.com/pytorch/pytorch/issues/79705,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: double backwards",scan when time allows,2022-06-19T17:27:45Z +issue,79703,Float and double tensors randomly initialized with the same seed get different values for size >= 16,https://github.com/pytorch/pytorch/issues/79703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2022-06-19T17:27:37Z +issue,79647,tensorboard SummaryWriter.add_graph fails when model uses empty tuples,https://github.com/pytorch/pytorch/issues/79647,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2022-06-19T17:26:09Z +issue,79336,Batches are being duplicated from go http call,https://github.com/pytorch/pytorch/issues/79336,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-17T20:56:51Z +issue,60028,RReLU and PReLU do not propagate input strides to outputs,https://github.com/pytorch/pytorch/issues/60028,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: nn",scan when time allows,2022-06-17T18:30:50Z +issue,76232,Observing a strange behavior - Row parallelism,https://github.com/pytorch/pytorch/issues/76232,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: nn, module: numerical-stability",scan when time allows,2022-06-17T14:34:23Z +issue,79359,[build] No documented way to install C++ binaries for pure-python development of pytorch,https://github.com/pytorch/pytorch/issues/79359,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2022-06-16T18:18:23Z +issue,79018,Random functions should infer device from user-specified Generator,https://github.com/pytorch/pytorch/issues/79018,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2022-06-16T14:59:53Z +issue,79592,[LTC] Introduce a `MetricsReport` python binding and allow backend to add their report as string,https://github.com/pytorch/pytorch/issues/79592,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-16T04:52:42Z +issue,79606,PyTorch Preview (Nightly) version number does not comply with Conda conventions,https://github.com/pytorch/pytorch/issues/79606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-06-15T18:52:34Z +issue,79563,__torch__dispatch does not return new output in inplace function,https://github.com/pytorch/pytorch/issues/79563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-06-15T14:28:28Z +issue,79425,API for accessing SymIntNode mandates refcount bump even when it is unnecessary,https://github.com/pytorch/pytorch/issues/79425,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-15T05:40:08Z +issue,79395,SymInt equality tests are unsound,https://github.com/pytorch/pytorch/issues/79395,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-15T05:40:01Z +issue,79387,caffe2_nvrtc is produced even when it won't be used,https://github.com/pytorch/pytorch/issues/79387,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: selective build",scan when time allows,2022-06-15T05:23:32Z +issue,79355,[bazel] build spams warnings,https://github.com/pytorch/pytorch/issues/79355,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bazel,scan when time allows,2022-06-15T05:20:10Z +issue,79453,[LTC] Make `torch::lazy::BackendImplInterface::ExecuteComputation` takes `ComputationPtr` instead of `Computation`,https://github.com/pytorch/pytorch/issues/79453,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-14T18:33:45Z +issue,79349,PyTorch get positive log_prob of a multivariate normal distribution,https://github.com/pytorch/pytorch/issues/79349,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-14T18:07:43Z +issue,79407,Conda enviroment,https://github.com/pytorch/pytorch/issues/79407,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-14T18:07:24Z +issue,79518,"Display a ""reference"" link for ops that points to primTorch implementations",https://github.com/pytorch/pytorch/issues/79518,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: primtorch",scan when time allows,2022-06-14T18:03:52Z +issue,79469,Add a new _broadcast_coalesced op for DDP,https://github.com/pytorch/pytorch/issues/79469,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-13T22:44:22Z +issue,79468,Ensure the guards in distributed_c10d.py wrappers get executed in the replay of the graph,https://github.com/pytorch/pytorch/issues/79468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-06-13T22:43:24Z +issue,79467,Add autograd support for dispatch passable c10d ops,https://github.com/pytorch/pytorch/issues/79467,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-06-13T22:42:09Z +issue,79222,scripted fft Convolutions are faster than nn.Conv1d with large kernels,https://github.com/pytorch/pytorch/issues/79222,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: performance",scan when time allows,2022-06-13T20:59:41Z +issue,78274,Memory allocation errors when attempting to initialize a large number of small feed-forward networks in RAM with shared memory despite having enough memory,https://github.com/pytorch/pytorch/issues/78274,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2022-06-13T20:54:20Z +issue,79244,[NVFuser] bad performance on pyhpc_equation_of_state,https://github.com/pytorch/pytorch/issues/79244,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-06-13T18:43:23Z +issue,62545,Incosistency with args for `nn.functional.max_poolNd` vs `nn.MaxPoolNd` functions,https://github.com/pytorch/pytorch/issues/62545,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2022-06-13T18:40:17Z +issue,79382,torch failure to open libcuda.so.1 on macOS,https://github.com/pytorch/pytorch/issues/79382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: macos,scan when time allows,2022-06-13T18:19:39Z +issue,79083,CI workflow creates too many tags in RSS feed,https://github.com/pytorch/pytorch/issues/79083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-06-13T17:47:28Z +issue,73113,PyTorch fails to compile on gcc 11.2 due to breakpad,https://github.com/pytorch/pytorch/issues/73113,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-06-12T12:15:56Z +issue,31945,"Function request: logerfc, logerfcx special functions",https://github.com/pytorch/pytorch/issues/31945,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: numpy, module: special",scan when time allows,2022-06-11T23:54:47Z +issue,47916,Hash mismatch for METADATA file,https://github.com/pytorch/pytorch/issues/47916,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-06-11T16:42:58Z +issue,30532,Version 1.3 no longer supporting Tesla K40m?,https://github.com/pytorch/pytorch/issues/30532,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: docs",scan when time allows,2022-06-11T11:15:57Z +issue,78878,Improving clarity in the docs of different losses,https://github.com/pytorch/pytorch/issues/78878,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: python frontend",scan when time allows,2022-06-10T21:10:22Z +issue,8921,[feature request] More methods for PackedSequence,https://github.com/pytorch/pytorch/issues/8921,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-06-10T15:51:51Z +issue,79261,[NVFuser] bad performance on pyhpc_isoneutral_mixing,https://github.com/pytorch/pytorch/issues/79261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-06-10T15:01:37Z +issue,79246,[NVFuser] bad performance on mobilenet_v2 and mobilenet_v3_large,https://github.com/pytorch/pytorch/issues/79246,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-06-10T14:59:33Z +issue,56901,Static prebuilt libraries contain shared libraries,https://github.com/pytorch/pytorch/issues/56901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: static linking",scan when time allows,2022-06-10T06:48:48Z +issue,79089,Improve PrimTorch testing for view consistency.,https://github.com/pytorch/pytorch/issues/79089,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: primtorch, module: tests, module: viewing and reshaping",scan when time allows,2022-06-09T23:37:32Z +issue,79177,_make_elementwise_unary_reference and other function factories in torch._refs don't set __name__ correctly,https://github.com/pytorch/pytorch/issues/79177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-06-09T23:36:45Z +issue,78987,DataLoader leaking resources?,https://github.com/pytorch/pytorch/issues/78987,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-06-09T23:08:27Z +issue,79032,Add a test that shows that lazy_ir reuse breaks SizeNodes,https://github.com/pytorch/pytorch/issues/79032,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-09T22:02:45Z +issue,79120,Hide or fuse TupleConstruct / TupleUnpack from tensorboard graph,https://github.com/pytorch/pytorch/issues/79120,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2022-06-09T00:11:12Z +issue,79130,torchvision.models.mobilenetv3 can't save pre-trained model to custom dir?,https://github.com/pytorch/pytorch/issues/79130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2022-06-09T00:04:43Z +issue,79118,About the source code of,https://github.com/pytorch/pytorch/issues/79118,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-09T00:04:26Z +issue,52260,Unlucky batch->device split by DataParallel causes an exception,https://github.com/pytorch/pytorch/issues/52260,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2022-06-08T18:21:25Z +issue,79137,[AUTOGRAD] codegen to use sym_sizes for ops w/ symint overloads in derivative formulas,https://github.com/pytorch/pytorch/issues/79137,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-08T18:05:14Z +issue,79021,Building PyTorch from Source with BUILD_LAZY_TS_BACKEND_ON,https://github.com/pytorch/pytorch/issues/79021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-06-07T22:32:03Z +issue,79073,"Multi-node, Multi-GPU set up tutorial for Slurm cluster",https://github.com/pytorch/pytorch/issues/79073,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-06-07T22:31:58Z +issue,79053,[MetaIssue] Propagating SymInts through Autograd,https://github.com/pytorch/pytorch/issues/79053,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-07T20:19:29Z +issue,79050,[MetaIssue] Investigate if we should be reusing primtorch formulas for `is_dynamic`,https://github.com/pytorch/pytorch/issues/79050,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-07T20:17:14Z +issue,79052,Mirror and implement `SymbolicIntNode` API for `SymInt` so we can trace in C++,https://github.com/pytorch/pytorch/issues/79052,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-07T20:04:22Z +issue,79031,Implement SymbolicIntNode interface for lazy (i.e. lazy::SymbolicIntNode),https://github.com/pytorch/pytorch/issues/79031,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-07T17:41:02Z +issue,79030,Devirtualize `sym_sizes`. It still has to work for python tensor subclasses and LTC/Xla,https://github.com/pytorch/pytorch/issues/79030,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-07T17:31:53Z +issue,79020,"When setting sizes and strides on a tensor subclass in `THPVariable_make_wrapper_subclass`, also make offset symbolic",https://github.com/pytorch/pytorch/issues/79020,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-06-07T17:19:29Z +issue,79013,Multi30k can't be downloaded the destination domain can't be reached,https://github.com/pytorch/pytorch/issues/79013,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-06-07T13:47:09Z +issue,79004,Doc on index of CPU Device seems wrong,https://github.com/pytorch/pytorch/issues/79004,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-06-07T13:30:36Z +issue,78961,[forwardAD] torch.no_grad has no effect under forward_ad,https://github.com/pytorch/pytorch/issues/78961,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2022-06-07T13:27:52Z +issue,78876,Remove const from function return type if returning const value,https://github.com/pytorch/pytorch/issues/78876,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-06-07T13:27:44Z +issue,78917,LambdaLR changes the learning rate in an undesired way,https://github.com/pytorch/pytorch/issues/78917,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2022-06-07T03:14:56Z +issue,78921,Do we really need sampler for IterableDataset?,https://github.com/pytorch/pytorch/issues/78921,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: performance",scan when time allows,2022-06-06T21:41:52Z +issue,78884,Abnormal GPU memory usage when using CUDA tensors with multiprocessing,https://github.com/pytorch/pytorch/issues/78884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: jiterator, module: memory usage, module: multiprocessing",scan when time allows,2022-06-06T19:04:22Z +issue,78882,Cannot build master on AWS cluster: error: ‘__fatDeviceText’ was not declared in this scope,https://github.com/pytorch/pytorch/issues/78882,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2022-06-06T19:00:21Z +issue,78581,RFC: Improve the performance and usability of linear algebra on CUDA devices,https://github.com/pytorch/pytorch/issues/78581,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: magma",scan when time allows,2022-06-06T17:23:56Z +issue,78808,Feature request: Integer system decompositions,https://github.com/pytorch/pytorch/issues/78808,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-06-06T14:51:56Z +issue,5434,RuntimeError: $ Torch: not enough memory: you tried to allocate 72GB. Buy new RAM!,https://github.com/pytorch/pytorch/issues/5434,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2022-06-06T08:01:09Z +issue,77200,[RFC] Allow device override during Tensor unpickling without torch.load,https://github.com/pytorch/pytorch/issues/77200,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pickle, module: serialization",scan when time allows,2022-06-05T09:28:44Z +issue,78871,[RecordFunction] Hold a durable schema reference,https://github.com/pytorch/pytorch/issues/78871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,scan when time allows,2022-06-04T03:48:01Z +issue,78845,"[Modes] no_dispatch is not the same as DisableTorchFunction, causing differences in modes",https://github.com/pytorch/pytorch/issues/78845,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: __torch_function__",scan when time allows,2022-06-03T23:22:45Z +issue,78737,New c10 constants,https://github.com/pytorch/pytorch/issues/78737,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-06-03T22:49:02Z +issue,78800,Efficiency of unary operations on CPU for large tensors,https://github.com/pytorch/pytorch/issues/78800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2022-06-03T22:08:10Z +issue,78761,Extend tag testing for aliases,https://github.com/pytorch/pytorch/issues/78761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-06-03T22:04:40Z +issue,78742,Add a gallery of examples with sphinx-gallery,https://github.com/pytorch/pytorch/issues/78742,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-06-03T22:00:17Z +issue,78741,Test approximation and numerical stability of numerical operators,https://github.com/pytorch/pytorch/issues/78741,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: testing",scan when time allows,2022-06-03T21:59:29Z +issue,78708,BuildExtension does not choose correct CUDA installation,https://github.com/pytorch/pytorch/issues/78708,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: cuda",scan when time allows,2022-06-03T21:55:24Z +issue,63140,[documentation] torch.distributed.elastic: illustrate how to write load_checkpoint and save_checkpoint in Train Script,https://github.com/pytorch/pytorch/issues/63140,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: elastic, oncall: r2p",scan when time allows,2022-06-03T20:47:29Z +issue,78249,`max_unpool2d` is not deterministic,https://github.com/pytorch/pytorch/issues/78249,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: nn, module: numerical-stability, module: pooling",scan when time allows,2022-06-03T18:58:50Z +issue,78786,Deprecate hardtanh type promotion behavior.,https://github.com/pytorch/pytorch/issues/78786,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: primtorch",scan when time allows,2022-06-03T16:05:17Z +issue,73531,Mention docker build process in RELEASE.md and automate building those for release,https://github.com/pytorch/pytorch/issues/73531,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docker",scan when time allows,2022-06-03T09:10:10Z +issue,78738,[primTorch] Sensible Error Messages,https://github.com/pytorch/pytorch/issues/78738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-06-03T03:13:12Z +issue,38138,"TorchScript to support == None, != None",https://github.com/pytorch/pytorch/issues/38138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-06-03T01:55:37Z +issue,49846,torch init_process_group always hangs,https://github.com/pytorch/pytorch/issues/49846,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-06-02T19:52:13Z +issue,78729,[Better Engineering] Make OpInfo-based test failures easy to reproduce,https://github.com/pytorch/pytorch/issues/78729,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-06-02T19:11:41Z +issue,78743,Expose docs from the yaml for each torch.Tag in Python,https://github.com/pytorch/pytorch/issues/78743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: docs, module: library",scan when time allows,2022-06-02T17:18:44Z +issue,78634,Debug job does not build in debug mode,https://github.com/pytorch/pytorch/issues/78634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-06-02T00:16:56Z +issue,78624,linalg.pinv_singular tests are slow,https://github.com/pytorch/pytorch/issues/78624,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra, module: tests",scan when time allows,2022-06-02T00:16:38Z +issue,77050,ENORMOUS OVERHEAD from mp.get_context('spawn'),https://github.com/pytorch/pytorch/issues/77050,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: performance, module: power",scan when time allows,2022-06-01T21:40:00Z +issue,76920,Avoid Self-loops on Module Creation,https://github.com/pytorch/pytorch/issues/76920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-06-01T18:21:20Z +issue,78519,test_python_dispatch fails on DEBUG=1,https://github.com/pytorch/pytorch/issues/78519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2022-06-01T18:21:00Z +issue,77515,Inplace Bool API + `sum` will trigger INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/77515,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",scan when time allows,2022-06-01T18:00:08Z +issue,77963,`logaddexp2` fails to backward,https://github.com/pytorch/pytorch/issues/77963,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",scan when time allows,2022-06-01T17:35:15Z +issue,78422,Permutation of Sparse Tensor,https://github.com/pytorch/pytorch/issues/78422,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-06-01T17:20:44Z +issue,78413,torch.angle differs from np.angle for -0.,https://github.com/pytorch/pytorch/issues/78413,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: primtorch",scan when time allows,2022-06-01T08:11:22Z +issue,78530,LibTorch cannot be used without nvcc,https://github.com/pytorch/pytorch/issues/78530,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-06-01T04:29:01Z +issue,77052,Building from source results in broken __version__,https://github.com/pytorch/pytorch/issues/77052,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2022-06-01T02:44:51Z +issue,77808,`Could not start gRPC server` flakiness in XLA tests,https://github.com/pytorch/pytorch/issues/77808,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xla,scan when time allows,2022-05-31T21:02:50Z +issue,78507,Parallel execution of multiple unrelated statements written sequentially,https://github.com/pytorch/pytorch/issues/78507,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-31T20:39:19Z +issue,78450,"Softmax, LogSoftmax are over parameterized",https://github.com/pytorch/pytorch/issues/78450,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-05-31T20:23:28Z +issue,78346,`gradcheck` fails for `torch.distribution.transform` APIs in forward mode,https://github.com/pytorch/pytorch/issues/78346,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: distributions, module: forward ad",scan when time allows,2022-05-31T19:06:17Z +issue,75926,Disable TracerWarnings on NVFuser opinfo tests,https://github.com/pytorch/pytorch/issues/75926,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",scan when time allows,2022-05-31T17:39:56Z +issue,78047,DDP multi host with single GPU each.,https://github.com/pytorch/pytorch/issues/78047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-05-31T17:26:59Z +issue,78513,clear input shape declaration on pytorch model inputs and outputs,https://github.com/pytorch/pytorch/issues/78513,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python frontend, module: shape checking",scan when time allows,2022-05-31T16:14:33Z +issue,78261,"pytorch-android-lite use its own libfbjni.so, which is not compatible with any other version at all..",https://github.com/pytorch/pytorch/issues/78261,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: binaries",scan when time allows,2022-05-31T15:38:29Z +issue,78518,Exponentiating floating number with cuda tensor is slow,https://github.com/pytorch/pytorch/issues/78518,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-05-31T15:37:29Z +issue,74143,Tracing model parameter shapes without instantiating the model parameters,https://github.com/pytorch/pytorch/issues/74143,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-05-31T15:36:57Z +issue,78075,torch.multiprocessing.spawn raise PicklingError inside a decorator,https://github.com/pytorch/pytorch/issues/78075,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: serialization",scan when time allows,2022-05-31T00:50:28Z +issue,19143,Support memoryview() method on torch.Tensor,https://github.com/pytorch/pytorch/issues/19143,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2022-05-30T14:35:08Z +issue,43546,"NCCL WARN Your program may be hanging, this may be caused by a collective mismatch around rank 2. Please check your collective calls at and around this rank. You can use NCCL_DEBUG=INFO and NCCL_DEBUG_SUBSYS=COLL to see the collective logs",https://github.com/pytorch/pytorch/issues/43546,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nccl, oncall: distributed",scan when time allows,2022-05-30T12:33:22Z +issue,33385,Illegal instruction: 4 - OSX 10.13.6 install from source,https://github.com/pytorch/pytorch/issues/33385,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2022-05-26T16:52:17Z +issue,78253,Any plan to add Noam scheduling?,https://github.com/pytorch/pytorch/issues/78253,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler,scan when time allows,2022-05-26T02:56:28Z +issue,46803,About 2 minor bug fixes on CUDA macOSX 10.13.6,https://github.com/pytorch/pytorch/issues/46803,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: macos, module: nnpack",scan when time allows,2022-05-26T00:05:50Z +issue,78170,[chalf] reference_testing: low quality test for fast growing ops,https://github.com/pytorch/pytorch/issues/78170,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: half",scan when time allows,2022-05-25T21:11:16Z +issue,78153,`pack_sequence` crash,https://github.com/pytorch/pytorch/issues/78153,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T20:35:43Z +issue,78151,`ctc_loss` will backward crash,https://github.com/pytorch/pytorch/issues/78151,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: edge cases",scan when time allows,2022-05-25T20:35:19Z +issue,78157,[Optimizer Overlap] Custom optimizer registration,https://github.com/pytorch/pytorch/issues/78157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-05-25T19:06:45Z +issue,78159,[Optimizer Overlap] Parameter group support,https://github.com/pytorch/pytorch/issues/78159,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-05-25T19:04:50Z +issue,78255,Floating point exception in _conv_depthwise2d,https://github.com/pytorch/pytorch/issues/78255,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:43:16Z +issue,78122,Segfault in bincount,https://github.com/pytorch/pytorch/issues/78122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:37:39Z +issue,78185,test_to (__main__.TestTorch) fails with multiple gpus,https://github.com/pytorch/pytorch/issues/78185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multi-gpu,scan when time allows,2022-05-25T18:32:33Z +issue,78131,Segfault in _pad_packed_sequence,https://github.com/pytorch/pytorch/issues/78131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:30:40Z +issue,78129,Segfault in _embedding_bag_forward_only,https://github.com/pytorch/pytorch/issues/78129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:30:39Z +issue,78127,Segfault in torch._C._nn.reflection_pad2d,https://github.com/pytorch/pytorch/issues/78127,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:30:39Z +issue,78126,Segfault in max_unpool3d,https://github.com/pytorch/pytorch/issues/78126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:30:38Z +issue,78125,Segfault in grid_sampler_3d,https://github.com/pytorch/pytorch/issues/78125,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:30:38Z +issue,78130,Segfault in _grid_sampler_2d_cpu_fallback,https://github.com/pytorch/pytorch/issues/78130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:29:46Z +issue,78128,Segfault in torch._C._nn.thnn_conv2d,https://github.com/pytorch/pytorch/issues/78128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: edge cases,scan when time allows,2022-05-25T18:29:25Z +issue,78260,[CI] Detect when tests are no longer running from CI,https://github.com/pytorch/pytorch/issues/78260,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-05-25T15:25:50Z +issue,77736,TimeSeriesDataset retrieve columns,https://github.com/pytorch/pytorch/issues/77736,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-25T09:44:59Z +issue,77463,conda CPU installation for LTS fails with UnsatisfiableError,https://github.com/pytorch/pytorch/issues/77463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lts,scan when time allows,2022-05-24T21:45:56Z +issue,77839,BUG: reference count leak when using `THPLayout_New` and `THPMemoryFormat_New` (static analyzer reports),https://github.com/pytorch/pytorch/issues/77839,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: python frontend",scan when time allows,2022-05-23T17:21:32Z +issue,77946,[failing test] test_foreach::test_binary_op_scalarlist_fastpath,https://github.com/pytorch/pytorch/issues/77946,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-23T16:43:51Z +issue,78070,[primTorch] Meta function for item creates a dummy value,https://github.com/pytorch/pytorch/issues/78070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-05-23T14:31:01Z +issue,77552,Functional Jacobian does not work with Torchdiffeq,https://github.com/pytorch/pytorch/issues/77552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-05-23T04:07:54Z +issue,78071,[primTorch] item prim can't return a bool properly,https://github.com/pytorch/pytorch/issues/78071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-05-23T04:01:36Z +issue,77951,`topk` returns different results with the same input twice in cuda,https://github.com/pytorch/pytorch/issues/77951,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: python frontend",scan when time allows,2022-05-23T03:41:53Z +issue,77801,`torch.utils.benchmark.examples.blas_compare` can not be parsed by Python-3.7 runtime,https://github.com/pytorch/pytorch/issues/77801,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: benchmark,scan when time allows,2022-05-23T03:39:10Z +issue,77899,Floating point exception in native_channel_shuffle,https://github.com/pytorch/pytorch/issues/77899,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-23T03:38:33Z +issue,77894,Floating point exception in channel_shuffle,https://github.com/pytorch/pytorch/issues/77894,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-23T03:37:01Z +issue,77838,Sporadic convolution error with dilation=0,https://github.com/pytorch/pytorch/issues/77838,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2022-05-23T03:27:19Z +issue,77844,Conv2D with large different number of input and output channels gives a CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/77844,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2022-05-23T03:26:53Z +issue,77973,"forward-mode support for ""logically composite"" operators",https://github.com/pytorch/pytorch/issues/77973,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: derivatives, module: forward ad",scan when time allows,2022-05-22T22:13:19Z +issue,77981,Override sym_sizes to create LTC IR for SymIntNode,https://github.com/pytorch/pytorch/issues/77981,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-22T22:13:02Z +issue,78061,Python memory allocator called without holding the GIL when running torchrun under Python debug version,https://github.com/pytorch/pytorch/issues/78061,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: r2p,scan when time allows,2022-05-22T22:12:09Z +issue,78053,toleranceOverride should override atol and rtol even when explicitly specified in a test,https://github.com/pytorch/pytorch/issues/78053,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: primtorch, module: tests",scan when time allows,2022-05-22T00:39:47Z +issue,76783,`gradcheck` fails for `torch.trace`,https://github.com/pytorch/pytorch/issues/76783,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-05-21T13:58:01Z +issue,77614,PyTorch fails to build on gcc 12 due to gloo,https://github.com/pytorch/pytorch/issues/77614,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2022-05-20T15:50:31Z +issue,77967,Inference Tensors should not be allowed to hold `grad_fn`,https://github.com/pytorch/pytorch/issues/77967,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-20T15:16:13Z +issue,77869,Throw warning if python optimise flags are enabled,https://github.com/pytorch/pytorch/issues/77869,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-05-20T04:25:12Z +issue,47702,GPU acceleration for Apple's M1 chip?,https://github.com/pytorch/pytorch/issues/47702,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2022-05-18T21:40:22Z +issue,77675,PrimTorch refs do not match argument naming with their PyTorch counterparts,https://github.com/pytorch/pytorch/issues/77675,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-05-18T17:37:44Z +issue,77546,Build check for AVX512 fails with AMD CPU and march=native,https://github.com/pytorch/pytorch/issues/77546,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vectorization",scan when time allows,2022-05-18T17:36:08Z +issue,77544,Modernize LoggingTensorMode,https://github.com/pytorch/pytorch/issues/77544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-05-18T17:34:20Z +issue,77668,Extend BC test to test for __torch_function__ overridability,https://github.com/pytorch/pytorch/issues/77668,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2022-05-18T17:30:54Z +issue,77731,complex abs strides are wrong on empty tensors and tensors with 1 dimension,https://github.com/pytorch/pytorch/issues/77731,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: primtorch",scan when time allows,2022-05-18T16:42:41Z +issue,77738,net_observer_reporter_print.h missing,https://github.com/pytorch/pytorch/issues/77738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-05-18T16:40:46Z +issue,28341,LinearOperator Abstraction / Structure-Exploiting LazyTensors for Linear Algebra,https://github.com/pytorch/pytorch/issues/28341,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lazy, module: linear algebra",scan when time allows,2022-05-18T16:00:34Z +issue,77231,`torch.scatter_add` will succeed when the `index` is a complex tensor,https://github.com/pytorch/pytorch/issues/77231,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2022-05-18T12:05:57Z +issue,62695,Use fast gpuAtomicAdd for FP16 data type in CUDA kernels,https://github.com/pytorch/pytorch/issues/62695,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half",scan when time allows,2022-05-17T22:35:50Z +issue,63034,[bug] nn.functional.pad (circular) ubsan failure,https://github.com/pytorch/pytorch/issues/63034,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",scan when time allows,2022-05-17T22:00:20Z +issue,74519,Lazy TS Test Failures,https://github.com/pytorch/pytorch/issues/74519,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-17T21:53:18Z +issue,24810,"Doesn't install the python module ""torch""",https://github.com/pytorch/pytorch/issues/24810,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-05-17T18:51:45Z +issue,77265,Subclasses with unwrapping `__torch_dispatch__` impls as parameters,https://github.com/pytorch/pytorch/issues/77265,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: nn",scan when time allows,2022-05-17T15:57:06Z +issue,77589,How to handle __module__ attribute for Public API bindings,https://github.com/pytorch/pytorch/issues/77589,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-05-17T14:37:45Z +issue,77548,lintrunner doesn't give good error message suggesting lintrunner init,https://github.com/pytorch/pytorch/issues/77548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2022-05-17T07:13:29Z +issue,77529,softmarginloss should use `log1p` and has an incorrect out= behaviour.,https://github.com/pytorch/pytorch/issues/77529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2022-05-17T07:05:33Z +issue,77454,More clarity in doc for `torch.cuda.Event.record`?,https://github.com/pytorch/pytorch/issues/77454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-05-17T06:46:27Z +issue,58813,Single-Process Multi-GPU is not the recommended mode for DDP,https://github.com/pytorch/pytorch/issues/58813,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-05-17T01:57:11Z +issue,76558,[jiterator] perf regression when jiterating few ops for complex dtype,https://github.com/pytorch/pytorch/issues/76558,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: jiterator, module: performance",scan when time allows,2022-05-16T06:28:24Z +issue,52332,Reciprocals of complex tensors with infinities are different from NumPy.,https://github.com/pytorch/pytorch/issues/52332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs, module: numpy",scan when time allows,2022-05-16T05:27:35Z +issue,57690,torch.cdist returns high diagonal values with CUDA,https://github.com/pytorch/pytorch/issues/57690,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: tf32",scan when time allows,2022-05-15T15:27:09Z +issue,77415,Suboptimal error message - nn.Linear with double argument,https://github.com/pytorch/pytorch/issues/77415,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2022-05-15T14:35:03Z +issue,64145,F.conv2d: confusing error message when using uint8 input instead of float32,https://github.com/pytorch/pytorch/issues/64145,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: error checking",scan when time allows,2022-05-15T03:00:30Z +issue,31141,[JIT] tensor(device=...) and tensor.to(device = ...) does not work properly in traced functions and modules,https://github.com/pytorch/pytorch/issues/31141,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-05-15T02:04:53Z +issue,74748,[jiterator] Jiterate Complex Ops,https://github.com/pytorch/pytorch/issues/74748,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: jiterator",scan when time allows,2022-05-14T17:00:59Z +issue,77397,Large numerical inconsistency for `torch.einsum` on RTX30 series GPU.,https://github.com/pytorch/pytorch/issues/77397,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tf32",scan when time allows,2022-05-14T01:26:40Z +issue,77411,Allow force building with/without AVX,https://github.com/pytorch/pytorch/issues/77411,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-05-13T14:03:01Z +issue,77380,microbenchmark-style tests,https://github.com/pytorch/pytorch/issues/77380,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-05-13T13:55:45Z +issue,76354,NVFuser failing extremal opinfos,https://github.com/pytorch/pytorch/issues/76354,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-05-12T22:40:53Z +issue,76780,`gradcheck` should support the comparison of NaN,https://github.com/pytorch/pytorch/issues/76780,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-05-12T20:36:16Z +issue,77340,Disable issue doesn't disable multiple dtypes correctly,https://github.com/pytorch/pytorch/issues/77340,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-05-12T14:36:03Z +issue,77223,torch.ops.aten.ceil(1.5) returns Tensor rather than scalar,https://github.com/pytorch/pytorch/issues/77223,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2022-05-11T23:36:17Z +issue,62675,dynamic shapes,https://github.com/pytorch/pytorch/issues/62675,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-05-11T20:49:40Z +issue,7795,[feature request] Add cudaification API for distributions,https://github.com/pytorch/pytorch/issues/7795,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-05-11T19:57:35Z +issue,77216,[primTorch] Reduction references don't return views consistent with their original operators,https://github.com/pytorch/pytorch/issues/77216,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-05-11T19:35:36Z +issue,77241,libtorch1.8 torch::sigmoid is wrong,https://github.com/pytorch/pytorch/issues/77241,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: jetson",scan when time allows,2022-05-11T17:46:49Z +issue,77184,Undefined symbol error when compiling and loading C++ extension,https://github.com/pytorch/pytorch/issues/77184,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cpp-extensions",scan when time allows,2022-05-11T15:46:41Z +issue,77251,The codegen unconditionaly generate code even when it is not going to be used,https://github.com/pytorch/pytorch/issues/77251,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-05-11T14:28:24Z +issue,77166,"torch.cholesky has been deprecated in favour of torch.linalg.cholesky. However, torch.cholesky_inverse remains as is. It should also be moved to torch.linalg",https://github.com/pytorch/pytorch/issues/77166,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-05-11T10:00:30Z +issue,75659,Misleading documentation for cholesky_inverse,https://github.com/pytorch/pytorch/issues/75659,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: linear algebra",scan when time allows,2022-05-10T23:07:03Z +issue,77140,"torch.randperm uses too much cpu, but not efficient.",https://github.com/pytorch/pytorch/issues/77140,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: performance, module: random",scan when time allows,2022-05-10T18:37:05Z +issue,69080,Generalized matmul,https://github.com/pytorch/pytorch/issues/69080,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-05-10T16:44:48Z +issue,76709,`reshape` for distributions.,https://github.com/pytorch/pytorch/issues/76709,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-05-10T16:40:33Z +issue,77161,SymInt shouldn't be in dynamic_type.h,https://github.com/pytorch/pytorch/issues/77161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-05-10T13:44:46Z +issue,77118,__name__ on OpOverload should not contain period,https://github.com/pytorch/pytorch/issues/77118,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-05-10T13:27:05Z +issue,77067,Unable to continue adding modules to `nn.Sequential` after using `del` method,https://github.com/pytorch/pytorch/issues/77067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-05-09T21:34:40Z +issue,48845,memory_format kwarg doesn't work on most factory functions,https://github.com/pytorch/pytorch/issues/48845,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: memory format",scan when time allows,2022-05-09T20:45:54Z +issue,67565,Adding a new kwarg to a torch.nn.functional function breaks FC for JIT,https://github.com/pytorch/pytorch/issues/67565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, oncall: jit",scan when time allows,2022-05-09T19:05:56Z +issue,76853,Inconsistent results between Pow and Float Pow with their numpy references for complex types,https://github.com/pytorch/pytorch/issues/76853,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",scan when time allows,2022-05-09T18:29:22Z +issue,76779,"Strange warning from `matmul(..., out=...)`",https://github.com/pytorch/pytorch/issues/76779,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-05-09T17:09:46Z +issue,76887,Multiprocessing DataLoader hangs on exception inside iterator when using a simple queue and a producer thread,https://github.com/pytorch/pytorch/issues/76887,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2022-05-09T13:57:04Z +issue,77047,Stop calling sizes/numel/dim/is_contiguous on undefined tensors,https://github.com/pytorch/pytorch/issues/77047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-05-09T04:33:35Z +issue,76972,Display EC2 information,https://github.com/pytorch/pytorch/issues/76972,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-05-09T04:29:49Z +issue,76966,[bug] `NATIVE` and `OMP` `parallel_for` implementations are inconsistent.,https://github.com/pytorch/pytorch/issues/76966,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: openmp",scan when time allows,2022-05-09T04:29:21Z +issue,76936,"Private API for accessing all ""internal"" attributes on Tensors",https://github.com/pytorch/pytorch/issues/76936,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-05-09T04:16:23Z +issue,76933,NVFuser opinfos - check for CudaFusionGroup in the graph,https://github.com/pytorch/pytorch/issues/76933,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-05-09T04:16:03Z +issue,76906,torch.nn.functional.linear sometimes incorrectly accepts arguments of the different type,https://github.com/pytorch/pytorch/issues/76906,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: type promotion",scan when time allows,2022-05-09T04:14:46Z +issue,76891,Unwanted behavior with some in-place operations on CPU,https://github.com/pytorch/pytorch/issues/76891,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2022-05-09T04:11:19Z +issue,76786,"`cumprod, prod` will backward fail if `dtype` argument is different than the dtype of input tensor",https://github.com/pytorch/pytorch/issues/76786,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex",scan when time allows,2022-05-09T04:07:42Z +issue,75599,kthvalue 20x slower than sort,https://github.com/pytorch/pytorch/issues/75599,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: sorting and selection",scan when time allows,2022-05-08T20:39:01Z +issue,77046,torch.stack test_conj_view and test_neg_view are failing after 77043,https://github.com/pytorch/pytorch/issues/77046,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: primtorch, module: viewing and reshaping",scan when time allows,2022-05-08T08:28:10Z +issue,42402,Reading data speed slower than tensorflow,https://github.com/pytorch/pytorch/issues/42402,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: performance",scan when time allows,2022-05-07T11:37:41Z +issue,70348,nan return by nn.CrossEntropyLoss when all the labels are ignore_index in torch 1.11,https://github.com/pytorch/pytorch/issues/70348,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-05-06T18:17:43Z +issue,76944,primTorch references don't handle scalar x scalar inputs correctly,https://github.com/pytorch/pytorch/issues/76944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: primtorch, module: type promotion",scan when time allows,2022-05-06T06:22:48Z +issue,76441,[checkpoint] Switch away from pickle-base serialization,https://github.com/pytorch/pytorch/issues/76441,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-05-05T23:42:15Z +issue,76733,test_python_reference_meta_functions takes too long to run,https://github.com/pytorch/pytorch/issues/76733,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-05-05T21:10:51Z +issue,76871,PrimTorch binary refs do not handle CUDA + CPU scalar tensors correctly,https://github.com/pytorch/pytorch/issues/76871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-05-05T18:58:23Z +issue,76638,Design API for accessing sparse tensor indices,https://github.com/pytorch/pytorch/issues/76638,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-05-05T18:05:11Z +issue,15963,ProcessGroupGlooTest.test_scatter_stress_cuda is flaky,https://github.com/pytorch/pytorch/issues/15963,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: flaky-tests, oncall: distributed",scan when time allows,2022-05-05T15:55:25Z +issue,16295,ProcessGroupGlooTest.test_gather_stress is flaky,https://github.com/pytorch/pytorch/issues/16295,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: flaky-tests, oncall: distributed",scan when time allows,2022-05-05T15:52:10Z +issue,52230,Removing mutation on block input should print any meaning message.,https://github.com/pytorch/pytorch/issues/52230,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2022-05-04T23:23:54Z +issue,76483,Expand pow and float_pow sampling function for more coverage,https://github.com/pytorch/pytorch/issues/76483,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-05-04T22:25:15Z +issue,76844,Unify torch.ops argument parsing code with PythonArgParser,https://github.com/pytorch/pytorch/issues/76844,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: python frontend",scan when time allows,2022-05-04T21:41:55Z +issue,76827,TYPEIGNORE lint run locally disagrees with CI,https://github.com/pytorch/pytorch/issues/76827,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: macos",scan when time allows,2022-05-04T21:32:37Z +issue,76659,Remove all docstrings when python is running in optimization mode,https://github.com/pytorch/pytorch/issues/76659,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: performance, module: python frontend",scan when time allows,2022-05-04T17:14:41Z +issue,76806,torch.Tensor.__rdiv__ long x scalar float type promotion is incorrect,https://github.com/pytorch/pytorch/issues/76806,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: primtorch, module: type promotion",scan when time allows,2022-05-04T10:03:33Z +issue,76804,"torch.add bool x bool allows integer alpha, inconsistent with other dtype type checking",https://github.com/pytorch/pytorch/issues/76804,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: primtorch,scan when time allows,2022-05-04T09:53:29Z +issue,75733,"`torch.sum, prod, cumsum, cumprod, sparse.sum` INTERNAL ASSERT FAIL",https://github.com/pytorch/pytorch/issues/75733,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: reductions",scan when time allows,2022-05-04T08:15:09Z +issue,76705,"Torch `x += y.bmm(z)` is faster than `x.baddbmm_(y, z)`",https://github.com/pytorch/pytorch/issues/76705,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2022-05-03T21:24:19Z +issue,46164,Deprecate spmm and dsmm functions,https://github.com/pytorch/pytorch/issues/46164,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: linear algebra, module: sparse",scan when time allows,2022-05-03T18:32:05Z +issue,76636,WeightNorm: Add reset_parameters Linear override,https://github.com/pytorch/pytorch/issues/76636,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2022-05-03T18:27:31Z +issue,76747,[checkpoint] Stable file format for checkpoints,https://github.com/pytorch/pytorch/issues/76747,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-05-03T16:46:53Z +issue,76734,__torch_function__ callers should always pass kwargs,https://github.com/pytorch/pytorch/issues/76734,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2022-05-03T16:45:42Z +issue,76745,[checkpoint] Handle overlapping storage during save and load,https://github.com/pytorch/pytorch/issues/76745,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-05-03T16:40:11Z +issue,76644,`torch.smm` backward fail with strange error message,https://github.com/pytorch/pytorch/issues/76644,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-05-03T16:18:22Z +issue,76732,Adding Polyloss to `torch`,https://github.com/pytorch/pytorch/issues/76732,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2022-05-03T16:06:03Z +issue,76338,torch.bucketize doc typo on the left boundary when 'right=True',https://github.com/pytorch/pytorch/issues/76338,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sorting and selection",scan when time allows,2022-05-03T09:28:40Z +issue,76612,OpInfo incorrectly advertises lu_solve support on CUDA even when compiled without magma,https://github.com/pytorch/pytorch/issues/76612,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: tests",scan when time allows,2022-05-03T09:04:18Z +issue,76585,`torch.linalg.cond` has different results for tensor requiring autograd,https://github.com/pytorch/pytorch/issues/76585,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra",scan when time allows,2022-05-03T09:02:18Z +issue,76618,pylint segfault,https://github.com/pytorch/pytorch/issues/76618,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2022-05-03T08:57:35Z +issue,76611,OpInfo CUDA bfloat16 support detection is buggy,https://github.com/pytorch/pytorch/issues/76611,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-05-03T08:57:15Z +issue,76627,`Tensor.logit`'s signature in doc misses `eps` argument,https://github.com/pytorch/pytorch/issues/76627,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: python frontend",scan when time allows,2022-05-03T08:56:22Z +issue,76649,Add `ldl_unpack` functionality,https://github.com/pytorch/pytorch/issues/76649,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-05-03T08:50:59Z +issue,76655,[feature request] no-param sort to exploit parallelization,https://github.com/pytorch/pytorch/issues/76655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multithreading, module: performance, module: sorting and selection",scan when time allows,2022-05-03T08:50:18Z +issue,76654,`torch.sort` does not exploit parallelization when invoked without the `dim` parameter.,https://github.com/pytorch/pytorch/issues/76654,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: sorting and selection",scan when time allows,2022-05-03T08:49:35Z +issue,76646,`torch.nn.HuberLoss` backwards unexpectedly fail,https://github.com/pytorch/pytorch/issues/76646,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn, module: type promotion",scan when time allows,2022-05-03T00:19:42Z +issue,42407,torch.distributions.multinomial.Multinomial cannot be used in batch,https://github.com/pytorch/pytorch/issues/42407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-05-02T19:42:15Z +issue,76555,[numpy] Missing Tensor-Scalar support for multiple binary ops,https://github.com/pytorch/pytorch/issues/76555,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python frontend",scan when time allows,2022-05-02T18:14:43Z +issue,71249,[docs] nn.Sequential docs should list member functions,https://github.com/pytorch/pytorch/issues/71249,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-05-02T14:51:12Z +issue,76519,Fix layout of masked output when all sparse dimensions are reduced,https://github.com/pytorch/pytorch/issues/76519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-05-02T11:55:59Z +issue,17113,Distributed training jobs do not terminate properly if there is a crash,https://github.com/pytorch/pytorch/issues/17113,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-04-30T01:53:53Z +issue,76514,Clean up PyTorch's private operators,https://github.com/pytorch/pytorch/issues/76514,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-04-30T00:47:48Z +issue,64041,torch.unique acting up for a binary tensor,https://github.com/pytorch/pytorch/issues/64041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: correctness (silent), module: numpy, module: sorting and selection",scan when time allows,2022-04-29T14:00:59Z +issue,76487,`F.interpolate` uses incorrect size when `align_corners=True`,https://github.com/pytorch/pytorch/issues/76487,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2022-04-28T21:08:30Z +issue,76451,[checkpoint] Extension hooks to support logging and telemetry,https://github.com/pytorch/pytorch/issues/76451,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-27T22:51:24Z +issue,76491,[NVFuser] Automated generation of microbenchmarks,https://github.com/pytorch/pytorch/issues/76491,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-04-27T22:49:42Z +issue,76465,Pyre type checking fails,https://github.com/pytorch/pytorch/issues/76465,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2022-04-27T21:31:15Z +issue,76433,Remove _log_softmax/_softmax in favor of log_softmax and softmax respectively.,https://github.com/pytorch/pytorch/issues/76433,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-04-27T21:25:38Z +issue,54135,torch.kron of a transposed input error,https://github.com/pytorch/pytorch/issues/54135,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2022-04-27T20:26:58Z +issue,76160,test_wishart_log_prob fails locally for me,https://github.com/pytorch/pytorch/issues/76160,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: tests",scan when time allows,2022-04-27T02:11:10Z +issue,76347,`index_select` allows negative `index` for sparse but not for strided `self`,https://github.com/pytorch/pytorch/issues/76347,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: sparse",scan when time allows,2022-04-26T21:58:26Z +issue,76362,Whether 'targetSize' in inferExpandGeometryImpl needs to be checked when it is less than 0,https://github.com/pytorch/pytorch/issues/76362,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2022-04-26T21:58:20Z +issue,74554,Specifying left-right-padding as tuple for asymmetric padding,https://github.com/pytorch/pytorch/issues/74554,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",scan when time allows,2022-04-26T21:26:43Z +issue,76225,"RuntimeError: bucket_count == per_bucket_sizes.size()INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1646755853042/work/torch/csrc/distributed/c10d/reducer.cpp"":980, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/76225,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-04-26T17:26:10Z +issue,76331,[checkpoint] Avoid loading whole tensor when resharding,https://github.com/pytorch/pytorch/issues/76331,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-26T17:02:18Z +issue,76327,[checkpoint] Add extension points to avoid the default serialization behavior,https://github.com/pytorch/pytorch/issues/76327,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-26T17:02:00Z +issue,76326,[checkpoint] Support models with different cross-rank metadata,https://github.com/pytorch/pytorch/issues/76326,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-26T17:01:55Z +issue,76323,[Checkpoint] Add module documentation,https://github.com/pytorch/pytorch/issues/76323,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-26T17:01:45Z +issue,76294,torch._remove_batch_dim is interceptable by __torch_function__ / batch tensors don't print correctly,https://github.com/pytorch/pytorch/issues/76294,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: vmap",scan when time allows,2022-04-26T02:54:43Z +issue,76300,Unexpected _LinAlgError appeared only on my device,https://github.com/pytorch/pytorch/issues/76300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-04-25T23:26:23Z +issue,76175,Many dispatch keys do not print to string correctly,https://github.com/pytorch/pytorch/issues/76175,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2022-04-25T17:19:01Z +issue,76166,at::real and at::imag as methods,https://github.com/pytorch/pytorch/issues/76166,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2022-04-25T17:17:22Z +issue,76127,run_test.py option to write out failed tests,https://github.com/pytorch/pytorch/issues/76127,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-04-25T17:15:53Z +issue,76106,Allow any operation that takes a Storage to also take a contiguous Tensor instead,https://github.com/pytorch/pytorch/issues/76106,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python frontend,scan when time allows,2022-04-25T17:13:09Z +issue,75986,gql_mocks.json has really long lines,https://github.com/pytorch/pytorch/issues/75986,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-25T17:09:55Z +issue,75911,'python setup.py build' failed but succeed using 'pip install -v .' which calls 'python setup.py build'.,https://github.com/pytorch/pytorch/issues/75911,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-04-25T17:09:39Z +issue,70906,`full` should take an integer size,https://github.com/pytorch/pytorch/issues/70906,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: python array api, module: tensor creation",scan when time allows,2022-04-25T14:22:43Z +issue,42332,RuntimeError: ProcessGroupNCCL does not support recv (torch-nightly),https://github.com/pytorch/pytorch/issues/42332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2022-04-22T06:32:33Z +issue,75701,Einsum should have an `out=` parameter,https://github.com/pytorch/pytorch/issues/75701,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-04-22T00:39:29Z +issue,64864,Base parallelism on CPU count available to process rather than system total,https://github.com/pytorch/pytorch/issues/64864,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2022-04-21T17:32:10Z +issue,72711,Feature request: Implement `gelsd` backend for `linalg.lstsq` via `linalg.svd`,https://github.com/pytorch/pytorch/issues/72711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-04-21T07:12:08Z +issue,76012,torch.nn.LayerNorm is very slow on GPU (much slower than a custom LayerNorm version in the ConvNext model),https://github.com/pytorch/pytorch/issues/76012,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2022-04-20T23:45:05Z +issue,75912,"interaction with psychopy during imports, script exits with: free(): invalid pointer. Aborted (core dumped)",https://github.com/pytorch/pytorch/issues/75912,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-20T23:12:06Z +issue,76138,Coverage test is only checking packages and not all submodules,https://github.com/pytorch/pytorch/issues/76138,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: python frontend",scan when time allows,2022-04-20T20:43:04Z +issue,76011,backcompat tests in test_nn.py are slow,https://github.com/pytorch/pytorch/issues/76011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: tests",scan when time allows,2022-04-20T20:03:47Z +issue,75680,Addressing skips in OpInfo nn.functional.binary_cross_entropy_with_logits,https://github.com/pytorch/pytorch/issues/75680,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: tests",scan when time allows,2022-04-20T19:19:45Z +issue,70160,NewOperatorRegistrationTest.testImplNoDefGetsCaught failed.,https://github.com/pytorch/pytorch/issues/70160,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: dispatch, module: tests",scan when time allows,2022-04-20T13:02:46Z +issue,76070,NVFuser microbenchmark classifier - hash on memory formats,https://github.com/pytorch/pytorch/issues/76070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-04-20T12:24:33Z +issue,76030,Dirichlet with small concentration,https://github.com/pytorch/pytorch/issues/76030,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-04-19T18:59:12Z +issue,75923,aten::_softmax.out doesn't work with non-contiguous Tensors,https://github.com/pytorch/pytorch/issues/75923,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-04-19T17:41:27Z +issue,76007,"Build a default NVFuser comparison callback, e.g. for use with torchbench",https://github.com/pytorch/pytorch/issues/76007,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-04-19T17:40:22Z +issue,76025,Numerical instability: matrix multiplication got different results on cpu and gpu,https://github.com/pytorch/pytorch/issues/76025,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tf32,scan when time allows,2022-04-19T16:00:17Z +issue,76047,NNC failing opinfo accuracy tests,https://github.com/pytorch/pytorch/issues/76047,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-19T15:57:25Z +issue,76028,A bug in instructions for building PyTorch with ASAN,https://github.com/pytorch/pytorch/issues/76028,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: docs",scan when time allows,2022-04-19T15:57:19Z +issue,76037,__torch_function__ and generator input hazard,https://github.com/pytorch/pytorch/issues/76037,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2022-04-19T15:56:41Z +issue,76043,RuntimeError: bucket_count == per_bucket_sizes.size() INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/76043,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: ddp",scan when time allows,2022-04-19T15:55:37Z +issue,76040,SSL certificate error: urlopen error [SSL: WRONG_VERSION_NUMBER] wrong version number (_ssl.c:1091)>,https://github.com/pytorch/pytorch/issues/76040,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-19T15:29:37Z +issue,75798,Fix workaround `__module__` used to appease public binding checks,https://github.com/pytorch/pytorch/issues/75798,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-18T18:29:44Z +issue,75652,Op segfaults with ForwardAD and Subclassed Tensor as Tangent,https://github.com/pytorch/pytorch/issues/75652,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2022-04-18T15:34:04Z +issue,74616,"__rpow__(self, other) OpInfo should not test the case where `other` is a Tensor",https://github.com/pytorch/pytorch/issues/74616,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-04-18T02:34:55Z +issue,32881,CUDNN_STATUS_EXECUTION_FAILED,https://github.com/pytorch/pytorch/issues/32881,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2022-04-17T05:16:42Z +issue,74788,[Structured] Make it possible to override create_out implementation on a per dispatch key basis (structured sparse),https://github.com/pytorch/pytorch/issues/74788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2022-04-17T03:09:26Z +issue,47201,torch.save has a maximum size regardless of RAM,https://github.com/pytorch/pytorch/issues/47201,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: serialization",scan when time allows,2022-04-16T00:03:17Z +issue,75925,autogen-58 microbenchmark fails on NNC gpu fusion,https://github.com/pytorch/pytorch/issues/75925,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-15T22:51:08Z +issue,75904,Move _SKIP_PYTHON_BINDINGS to native_functions.yaml,https://github.com/pytorch/pytorch/issues/75904,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-04-15T21:55:10Z +issue,75909,Add batching rules for `{view}_copy` operators,https://github.com/pytorch/pytorch/issues/75909,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: batching,scan when time allows,2022-04-15T21:53:43Z +issue,75895,add -D_GLIBCXX_ASSERTIONS in debug mode,https://github.com/pytorch/pytorch/issues/75895,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-04-15T21:36:00Z +issue,75494,`torch.pinverse` on CUDA tensors produces non-optimal output for certain type of (invertible) matrix on torch > 1.7.1 but not on torch <= 1.7.1,https://github.com/pytorch/pytorch/issues/75494,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: numerical-stability",scan when time allows,2022-04-15T18:24:25Z +issue,66271,1.10.0-rc1 fails to compile with gcc 6.4.0,https://github.com/pytorch/pytorch/issues/66271,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-04-15T13:27:08Z +issue,74907,[Proposal] Use batched oprations to accelerate PowerSGD,https://github.com/pytorch/pytorch/issues/74907,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: mta, module: optimizer, module: performance",scan when time allows,2022-04-13T22:31:22Z +issue,75582,Add `balance` flag to `random_split`,https://github.com/pytorch/pytorch/issues/75582,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2022-04-13T20:42:56Z +issue,75737,torch.device missing doctring,https://github.com/pytorch/pytorch/issues/75737,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-04-13T17:24:30Z +issue,72257,`torch.hasnan`,https://github.com/pytorch/pytorch/issues/72257,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: nans and infs",scan when time allows,2022-04-13T16:19:45Z +issue,75565,make lint should advertise make setup_lint,https://github.com/pytorch/pytorch/issues/75565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: lint",scan when time allows,2022-04-12T16:28:26Z +issue,75562,"clang-tidy ""error: do not use const_cast"" cppcoreguidelines-pro-type-const-cast is counterproductive",https://github.com/pytorch/pytorch/issues/75562,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: lint",scan when time allows,2022-04-12T16:27:48Z +issue,75667,Implement histc for bfloat16 on CPU,https://github.com/pytorch/pytorch/issues/75667,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: sorting and selection",scan when time allows,2022-04-12T15:43:51Z +issue,75474,[NVFuser] call kernels with informative names (e.g. pow_mul_add),https://github.com/pytorch/pytorch/issues/75474,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-04-12T07:54:23Z +issue,75503,make_dual errors out when primal is a Tensor and tangent is a subclass Tensor,https://github.com/pytorch/pytorch/issues/75503,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: forward ad,scan when time allows,2022-04-12T07:08:23Z +issue,55277,Small model occupies too much GPU in CUDA11.1 + Torch1.8.1 but is normal in Torch 1.6 + CUDA10.1,https://github.com/pytorch/pytorch/issues/55277,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: performance",scan when time allows,2022-04-12T04:54:00Z +issue,56396,Tests in CI are run from the test/ directory,https://github.com/pytorch/pytorch/issues/56396,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: tests",scan when time allows,2022-04-11T21:06:43Z +issue,75213,Improve wording in _store_based_barrier logging and identify ranks that have not joined,https://github.com/pytorch/pytorch/issues/75213,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-04-11T17:42:00Z +issue,65230,RuntimeError: Trying to create tensor with negative dimension -1741885395: [-1741885395],https://github.com/pytorch/pytorch/issues/65230,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-04-11T17:24:33Z +issue,74014,torch.utils.data.Dataset combined with pycuda issue,https://github.com/pytorch/pytorch/issues/74014,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2022-04-11T14:43:09Z +issue,75568,mypy typing strategy for Tensor-likes (`__torch_function__`),https://github.com/pytorch/pytorch/issues/75568,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: typing",scan when time allows,2022-04-11T14:21:19Z +issue,75549,`torch.linalg.lstsq` raises `CUBLAS_STATUS_EXECUTION_FAILED` for large `B` in CUDA tensors,https://github.com/pytorch/pytorch/issues/75549,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2022-04-11T13:41:00Z +issue,75554,torch.overrides testing is not catching people adding new kwargs and not passing on to handle_torch_function,https://github.com/pytorch/pytorch/issues/75554,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: tests",scan when time allows,2022-04-11T13:40:16Z +issue,72345,Add ZeroTensor fastpath for torch.addmm,https://github.com/pytorch/pytorch/issues/72345,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-11T01:02:12Z +issue,72362,Add ZeroTensor fastpath for torch.bmm,https://github.com/pytorch/pytorch/issues/72362,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-11T00:59:00Z +issue,72359,Add ZeroTensor fastpath for torch.addmv,https://github.com/pytorch/pytorch/issues/72359,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-11T00:58:55Z +issue,73930,"User raised TypeError from __torch_dispatch__ gets turned into generic ""unsupported operand type"" message",https://github.com/pytorch/pytorch/issues/73930,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: bootcamp",scan when time allows,2022-04-10T23:13:48Z +issue,74876,Failure to set number of threads on AWS Lambda,https://github.com/pytorch/pytorch/issues/74876,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multithreading",scan when time allows,2022-04-10T05:31:33Z +issue,75371,"NVFuser bad ""reshape"" performance",https://github.com/pytorch/pytorch/issues/75371,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nvfuser,scan when time allows,2022-04-08T21:47:20Z +issue,75025,torch.cuda.init() unstacks existing CUDA contexts,https://github.com/pytorch/pytorch/issues/75025,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-04-08T21:15:08Z +issue,75510,"[doc] view appears to mean different things, `view/reshape` vs `transpose/permute`.",https://github.com/pytorch/pytorch/issues/75510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy, module: partial aliasing, module: viewing and reshaping",scan when time allows,2022-04-08T19:24:20Z +issue,75419,Require PyTorch test suite to be warnings clean,https://github.com/pytorch/pytorch/issues/75419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-04-08T17:47:51Z +issue,75427,`torch.cuda.is_bf16_supported()` seem to not work properly,https://github.com/pytorch/pytorch/issues/75427,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cuda",scan when time allows,2022-04-08T07:55:41Z +issue,75495,Forward AD convolution fails for the empty backend,https://github.com/pytorch/pytorch/issues/75495,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: forward ad",scan when time allows,2022-04-08T05:26:24Z +issue,75462,NotImplemented confusion between __torch_function__ and __rpow__ (and other dunder magic methods),https://github.com/pytorch/pytorch/issues/75462,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2022-04-08T03:57:51Z +issue,75363,conv3d has numerical issue where same input produces output that are not bit-wise identical,https://github.com/pytorch/pytorch/issues/75363,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: determinism",scan when time allows,2022-04-07T18:46:29Z +issue,21567,torch.bernoulli's parameter generator not documented,https://github.com/pytorch/pytorch/issues/21567,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: random",scan when time allows,2022-04-07T11:06:14Z +issue,75251,Potential runtime optimization of Mish activation,https://github.com/pytorch/pytorch/issues/75251,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: python frontend",scan when time allows,2022-04-06T21:27:54Z +issue,75198,`torch.cuda.get_device_name` fails to identify RTX 3090 Ti,https://github.com/pytorch/pytorch/issues/75198,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-04-05T16:24:40Z +issue,10375,"Sending CUDA tensor to process, and then back, does not work",https://github.com/pytorch/pytorch/issues/10375,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: cuda, module: multiprocessing",scan when time allows,2022-04-05T13:33:11Z +issue,74185,[DDP] Parallelize initialization collectives,https://github.com/pytorch/pytorch/issues/74185,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-04-04T19:26:48Z +issue,74485,dir(torch._VF) doesn't work,https://github.com/pytorch/pytorch/issues/74485,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-04-04T17:19:51Z +issue,75180,Add BUILD_LAZY_CUDA_SPARSE option,https://github.com/pytorch/pytorch/issues/75180,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: sparse",scan when time allows,2022-04-04T15:56:17Z +issue,74052,assertEqual gives confusing error when comparing tuple with Tensor with Tensor,https://github.com/pytorch/pytorch/issues/74052,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-04-04T14:51:33Z +issue,74951,[feature request] `coalesce` to support `dim` argument.,https://github.com/pytorch/pytorch/issues/74951,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: sparse",scan when time allows,2022-04-04T00:40:04Z +issue,71446,Discussion of TorchQuantum and QuantumNAS,https://github.com/pytorch/pytorch/issues/71446,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2022-04-03T21:13:00Z +issue,71636,`torch.median` will return -2147483648 when input is an empty tensor,https://github.com/pytorch/pytorch/issues/71636,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2022-04-02T17:18:20Z +issue,74148,powersgd can not get linear growth due to extra 2 times allreduce,https://github.com/pytorch/pytorch/issues/74148,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-04-02T17:14:56Z +issue,75129,[BE] add documentation for adjust learning rate when going to distributed training,https://github.com/pytorch/pytorch/issues/75129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-04-01T19:35:53Z +issue,45125,Boolean indexing of an ndarray with a torch.tensor mask breaks for size=1,https://github.com/pytorch/pytorch/issues/45125,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: numpy",scan when time allows,2022-04-01T14:10:44Z +issue,71629,"`torch.{max,min}` have strange error message when `input.numel()==0`",https://github.com/pytorch/pytorch/issues/71629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2022-04-01T13:50:04Z +issue,74896,Install PyTorch from source on power machine,https://github.com/pytorch/pytorch/issues/74896,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2022-03-31T22:58:24Z +issue,65423,Enable is_contiguous slowpath for torch_dispatch,https://github.com/pytorch/pytorch/issues/65423,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-03-31T20:21:27Z +issue,32524,Pytorch 1.4 does not detect gpu,https://github.com/pytorch/pytorch/issues/32524,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-03-31T18:37:49Z +issue,73182,Segmentation fault in _sobol_engine_scramble_,https://github.com/pytorch/pytorch/issues/73182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: random",scan when time allows,2022-03-31T16:53:59Z +issue,74911,"Include finfo(dtype).[min, max, eps, tiny] in the extremal test case",https://github.com/pytorch/pytorch/issues/74911,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-03-30T04:58:30Z +issue,74942,Store SourceDataset in MapDataset using pointer,https://github.com/pytorch/pytorch/issues/74942,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: dataloader",scan when time allows,2022-03-30T03:46:43Z +issue,74937,[docs] RandomSampler has unrendered back-ticks,https://github.com/pytorch/pytorch/issues/74937,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-03-30T03:45:40Z +issue,64477,torch.onnx.export crash while export quantized model,https://github.com/pytorch/pytorch/issues/64477,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2022-03-30T01:08:14Z +issue,74222,Autograd API to get saved for backwards tensors for an autograd graph,https://github.com/pytorch/pytorch/issues/74222,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: double backwards",scan when time allows,2022-03-29T20:30:13Z +issue,74604,"Training, Forward / backward pass with _different_ batch-size, no speedup observed when backward pass has smaller batch-size",https://github.com/pytorch/pytorch/issues/74604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-03-29T17:29:26Z +issue,74884,Problematic ASGD Optimizer,https://github.com/pytorch/pytorch/issues/74884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2022-03-29T13:23:18Z +issue,74167,Slower performance of `torch.mm` method with sparse CSR tensor,https://github.com/pytorch/pytorch/issues/74167,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkl, module: sparse",scan when time allows,2022-03-29T09:51:27Z +issue,74491,__torch_function__ is hitting private functions in some cases,https://github.com/pytorch/pytorch/issues/74491,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2022-03-28T23:10:02Z +issue,74500,One should be able to query if a DecorateInfo is a xfail or skip,https://github.com/pytorch/pytorch/issues/74500,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-03-28T22:20:58Z +issue,74503,Compilation error on M1 Mac,https://github.com/pytorch/pytorch/issues/74503,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: macos",scan when time allows,2022-03-28T22:19:47Z +issue,74514,Figure out a lint for directly returning a nullptr instead of raising python_error,https://github.com/pytorch/pytorch/issues/74514,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-28T22:19:19Z +issue,74855,Enable dtype keyword argument for to_dense method,https://github.com/pytorch/pytorch/issues/74855,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-03-28T22:16:45Z +issue,69033,Lazy Tensor Core Documentation Out-of-Date,https://github.com/pytorch/pytorch/issues/69033,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: lazy, module: xla",scan when time allows,2022-03-28T18:42:19Z +issue,73504,`ge` and `div` behaves differently when converting an overflow number,https://github.com/pytorch/pytorch/issues/73504,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: type promotion",scan when time allows,2022-03-28T18:42:15Z +issue,74544,Many inplace operators are not being tested for variant consistency (test_variant_consistency_eager),https://github.com/pytorch/pytorch/issues/74544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-03-28T13:10:17Z +issue,74679,Embedding isn't determinstic on linux-xenial-cuda11.3-py3.7-gcc7,https://github.com/pytorch/pytorch/issues/74679,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: embedding",scan when time allows,2022-03-27T02:53:07Z +issue,30896,"FATAL_ERROR ""Failed to determine the source files for the regular expression backend""",https://github.com/pytorch/pytorch/issues/30896,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2022-03-26T03:28:39Z +issue,74771,[bug] `torch.multinomial` should throw error as documented,https://github.com/pytorch/pytorch/issues/74771,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-25T22:33:19Z +issue,51062,test_profiler_with_remote_builtin (__main__.TensorPipeRpcTestWithSpawn) is flaky,https://github.com/pytorch/pytorch/issues/51062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: rpc, oncall: distributed",scan when time allows,2022-03-25T22:25:57Z +issue,64093,test_backward_accumulate_grads (__main__.TensorPipeDistAutogradTest) is flaky,https://github.com/pytorch/pytorch/issues/64093,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: rpc, oncall: distributed",scan when time allows,2022-03-25T22:16:07Z +issue,74634,Sampling Issue With Distributions,https://github.com/pytorch/pytorch/issues/74634,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2022-03-25T21:22:49Z +issue,23547,model use dilated conv backward in v1.1.0 is ~3x slower than in v0.4.1 on 1080Ti,https://github.com/pytorch/pytorch/issues/23547,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn, module: performance",scan when time allows,2022-03-25T07:05:09Z +issue,71518,Compilation instructions are not exhaustive: <> on Fedora 35/CUDA 11.6,https://github.com/pytorch/pytorch/issues/71518,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-03-25T02:06:20Z +issue,12435,`pstrf` on positive semi-definite matrices,https://github.com/pytorch/pytorch/issues/12435,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-03-25T00:21:47Z +issue,46157,Ability to disable cusolver,https://github.com/pytorch/pytorch/issues/46157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: linear algebra",scan when time allows,2022-03-25T00:00:09Z +issue,72448,Use Module `__getattr__` in `torch.ops` and friends,https://github.com/pytorch/pytorch/issues/72448,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-24T23:20:48Z +issue,74642,Some easy way to add xfails to OpInfos,https://github.com/pytorch/pytorch/issues/74642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-03-24T18:25:10Z +issue,69009,Mamba does not respect `cpuonly` when creating a conda environment from YAML file,https://github.com/pytorch/pytorch/issues/69009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2022-03-24T08:54:53Z +issue,74384,Package and Deploy: Generate External Python Registration files at runtime [WIP],https://github.com/pytorch/pytorch/issues/74384,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deploy,scan when time allows,2022-03-23T20:45:11Z +issue,50345,torch.special tracking issue,https://github.com/pytorch/pytorch/issues/50345,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: special",scan when time allows,2022-03-23T20:40:32Z +issue,934,dataloader parallels over elements vs over batches,https://github.com/pytorch/pytorch/issues/934,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-03-23T20:05:35Z +issue,41292,"[RFC, Tracker] DataLoader improvements",https://github.com/pytorch/pytorch/issues/41292,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-03-23T18:25:11Z +issue,45111,(prototype) Graph Mode Dynamic Quantization on BERT failure on quantize_dynamic_jit(...) call,https://github.com/pytorch/pytorch/issues/45111,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,scan when time allows,2022-03-23T16:12:35Z +issue,36108,Add a flow_sample function to sample optical flows and displacement fields,https://github.com/pytorch/pytorch/issues/36108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-03-23T12:58:02Z +issue,73180,Segmentation fault in _sobol_engine_ff_,https://github.com/pytorch/pytorch/issues/73180,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: random",scan when time allows,2022-03-23T03:34:41Z +issue,73181,Segmentation fault in _sobol_engine_initialize_state_,https://github.com/pytorch/pytorch/issues/73181,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: random",scan when time allows,2022-03-23T03:34:29Z +issue,73190,Segmentation fault in max_pool1d,https://github.com/pytorch/pytorch/issues/73190,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: error checking, module: pooling",scan when time allows,2022-03-23T03:30:58Z +issue,74320,Make OpInfo repr more useful,https://github.com/pytorch/pytorch/issues/74320,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-03-22T11:20:53Z +issue,74328,instantiate_device_type_tests is misnamed,https://github.com/pytorch/pytorch/issues/74328,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-03-22T08:47:57Z +issue,74531,Feature Request: torch.testing.make_scalar (make_tensor for scalars),https://github.com/pytorch/pytorch/issues/74531,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-03-22T08:41:46Z +issue,37332,Concerning default confiugration for distribution packages,https://github.com/pytorch/pytorch/issues/37332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2022-03-21T20:53:31Z +issue,64025,RuntimeError: cuDNN error: CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/64025,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2022-03-21T20:45:43Z +issue,74411,"UserWarning: size_average and reduce args will be deprecated, please use reduction='mean' instead.",https://github.com/pytorch/pytorch/issues/74411,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-03-21T16:35:50Z +issue,72960,Refactor/Cleanup lazy IR codegen,https://github.com/pytorch/pytorch/issues/72960,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-03-21T16:33:15Z +issue,74398,_run_ninja_build failing with clang,https://github.com/pytorch/pytorch/issues/74398,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-03-21T15:51:25Z +issue,74391,Can this cudaDeviceSynchronize call be removed?,https://github.com/pytorch/pytorch/issues/74391,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-03-21T15:50:11Z +issue,74421,"can not build pytorch, failing due to missing _ctypes module",https://github.com/pytorch/pytorch/issues/74421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-03-21T15:06:21Z +issue,74337,None returned from data loader causes debugging difficulty from collate function,https://github.com/pytorch/pytorch/issues/74337,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-03-21T13:52:57Z +issue,74389,torch.package unpickling transforms: ModuleNotFoundError: No module named 'torch._C._linalg'; 'torch._C' is not a package,https://github.com/pytorch/pytorch/issues/74389,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: pickle, module: serialization",scan when time allows,2022-03-20T07:57:35Z +issue,61909,Support matrix operations between complex and real tensors,https://github.com/pytorch/pytorch/issues/61909,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: linear algebra, module: type promotion",scan when time allows,2022-03-19T16:21:53Z +issue,41323,[discussion] Comparison operator chaining,https://github.com/pytorch/pytorch/issues/41323,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor,scan when time allows,2022-03-19T15:14:49Z +issue,74301,Error in lobpcg when using largest=False,https://github.com/pytorch/pytorch/issues/74301,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: masked operators, module: sparse",scan when time allows,2022-03-19T00:24:32Z +issue,74368,"Can't download torch vision models for older versions via torch.hub.load: ""multiple choices""",https://github.com/pytorch/pytorch/issues/74368,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: hub, module: models",scan when time allows,2022-03-18T21:52:07Z +issue,74329,CompositeImplicitAutograd addr implementation does not have correct behavior for bool (meta gives wrong dtype),https://github.com/pytorch/pytorch/issues/74329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2022-03-17T16:22:10Z +issue,73710,Refactor/Cleanup Lazy Tensor Core,https://github.com/pytorch/pytorch/issues/73710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-03-16T23:41:50Z +issue,62011,Enable clang-tidy on all of master,https://github.com/pytorch/pytorch/issues/62011,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-16T23:37:33Z +issue,74296,Time Gated Lstm,https://github.com/pytorch/pytorch/issues/74296,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2022-03-16T21:59:41Z +issue,74248,dir() on torch.ops.aten doesn't work,https://github.com/pytorch/pytorch/issues/74248,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-16T20:28:31Z +issue,74295,"TypeError: new_empty(): argument 'size' (position 1) must be tuple of ints, not list",https://github.com/pytorch/pytorch/issues/74295,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2022-03-16T20:23:36Z +issue,74300,repeat_interleave is not meta friendly; not transform friendly either,https://github.com/pytorch/pytorch/issues/74300,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-16T20:22:53Z +issue,74299,Gradients tests are very time consuming,https://github.com/pytorch/pytorch/issues/74299,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: ci",scan when time allows,2022-03-16T19:46:09Z +issue,74256,Create secure credential storage for metrics credentials and associated documentation on how to regenerate them if needed,https://github.com/pytorch/pytorch/issues/74256,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-03-16T17:30:02Z +issue,74281,Move torchbench workflow to `workflow_dispatch`?,https://github.com/pytorch/pytorch/issues/74281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-03-16T05:03:41Z +issue,74187,[docker] test_corrcoef_cpu_complex64 fails on CPU build,https://github.com/pytorch/pytorch/issues/74187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-03-15T07:49:03Z +issue,74138,`torch.set_printoptions` overwrites settings of its own previous calls,https://github.com/pytorch/pytorch/issues/74138,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: printing, module: ux",scan when time allows,2022-03-14T20:26:42Z +issue,73883,Possible issue with memory allocation.,https://github.com/pytorch/pytorch/issues/73883,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: numpy",scan when time allows,2022-03-14T17:18:31Z +issue,74134,No module named 'pygame': soft_actor_critic,https://github.com/pytorch/pytorch/issues/74134,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-03-14T16:55:48Z +issue,74155,[Android Studio] DefaultCPUAllocator: not enough memory: you tried to allocate 280166432 bytes,https://github.com/pytorch/pytorch/issues/74155,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: memory usage",scan when time allows,2022-03-14T12:46:06Z +issue,74152,All-caps names for pages are confusing,https://github.com/pytorch/pytorch/issues/74152,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2022-03-14T12:45:18Z +issue,74146,RuntimeError: Connection reset by peer when backened by NCCL,https://github.com/pytorch/pytorch/issues/74146,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nccl",scan when time allows,2022-03-14T12:42:22Z +issue,73870,max_pool1d() returns when given invalid large `kernel_size` inputs,https://github.com/pytorch/pytorch/issues/73870,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: edge cases, module: nn",scan when time allows,2022-03-14T09:37:22Z +issue,73924,torch.cuda.amp: Remove SPMD DDP doc portion,https://github.com/pytorch/pytorch/issues/73924,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: docs, oncall: distributed",scan when time allows,2022-03-14T09:28:34Z +issue,74101,C++ Context::setDeterministicAlgorithms default 2nd arg not defined in header,https://github.com/pytorch/pytorch/issues/74101,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: determinism",scan when time allows,2022-03-14T07:28:55Z +issue,72545,Split up torch.distributions docs into multiple pages,https://github.com/pytorch/pytorch/issues/72545,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs",scan when time allows,2022-03-13T23:38:30Z +issue,45943,Improve distributed documentation for NCCL_BLOCKING_WAIT,https://github.com/pytorch/pytorch/issues/45943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-03-12T03:50:09Z +issue,73916,Invalid code inrandom_ kernel,https://github.com/pytorch/pytorch/issues/73916,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-03-11T15:28:32Z +issue,73989,Invalid example tensor: detectron2_maskrcnn,https://github.com/pytorch/pytorch/issues/73989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-03-11T14:58:22Z +issue,73994,Lazy: add support for index ops,https://github.com/pytorch/pytorch/issues/73994,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-03-11T14:58:10Z +issue,73985,Release note bot is not sending messages on PRs anymore,https://github.com/pytorch/pytorch/issues/73985,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2022-03-11T14:58:00Z +issue,6257,"[feature request] adding a nonzero element ""in-place"" in sparse tensor",https://github.com/pytorch/pytorch/issues/6257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-03-11T09:10:47Z +issue,50097,Cannot print 32-bit complex tensors,https://github.com/pytorch/pytorch/issues/50097,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: docs, module: half, module: printing",scan when time allows,2022-03-11T07:13:24Z +issue,66858,CUDAgraph error while capturing,https://github.com/pytorch/pytorch/issues/66858,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda graphs,scan when time allows,2022-03-09T20:03:29Z +issue,45760,Complex Number support for distributed,https://github.com/pytorch/pytorch/issues/45760,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: c10d, module: complex, oncall: distributed",scan when time allows,2022-03-09T18:59:59Z +issue,73860,tensor subclass and `__torch_function__` performance issues,https://github.com/pytorch/pytorch/issues/73860,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2022-03-09T13:58:15Z +issue,73600,Add a section in DDP tutorial to explain why DDP sometimes is slower than local training and how to improve it,https://github.com/pytorch/pytorch/issues/73600,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-03-08T22:03:17Z +issue,73683,test_broadcast_coalesced_nccl fails on A100 GPUs,https://github.com/pytorch/pytorch/issues/73683,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-03-08T22:03:14Z +issue,73721,Audit exception rewrapping to ensure stack traces are preserved,https://github.com/pytorch/pytorch/issues/73721,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2022-03-08T22:03:09Z +issue,73722,Why is distributed RPC using the default pickler?,https://github.com/pytorch/pytorch/issues/73722,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2022-03-08T22:03:04Z +issue,65022,"[AWS EC2 P3DN, EFA is enabled] Torch RPC tensorpipe/common/ibv.h:172 """": Operation not supported",https://github.com/pytorch/pytorch/issues/65022,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe, oncall: distributed",scan when time allows,2022-03-08T22:02:44Z +issue,73825,TypeError: cannot pickle 'torch._C._distributed_c10d._ProcessGroupGloo' object,https://github.com/pytorch/pytorch/issues/73825,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: ddp, oncall: distributed",scan when time allows,2022-03-08T21:59:42Z +issue,73070,`enumerate_support` for continuous distributions,https://github.com/pytorch/pytorch/issues/73070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-03-08T21:54:27Z +issue,67463,`test_forward_mode_AD` hangs for `nn.functional.cosine_embedding_loss`,https://github.com/pytorch/pytorch/issues/67463,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: deadlock, module: forward ad, module: nn",scan when time allows,2022-03-08T17:04:11Z +issue,73826,Better Error report in torch.distribution.*.sample (when passing a non-iterable),https://github.com/pytorch/pytorch/issues/73826,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: error checking",scan when time allows,2022-03-08T14:23:46Z +issue,73840,"max_pool1d(): argument 'dilation' must be tuple of ints, but found element of type int at pos 1",https://github.com/pytorch/pytorch/issues/73840,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: edge cases, module: pooling",scan when time allows,2022-03-07T20:34:48Z +issue,62973,Audit use of `C10_UNUSED`,https://github.com/pytorch/pytorch/issues/62973,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-07T18:19:44Z +issue,73841,max_pool1d(): `RuntimeError: [enforce fail at CPUAllocator.cpp:68] . DefaultCPUAllocator: can't allocate memory`,https://github.com/pytorch/pytorch/issues/73841,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2022-03-07T14:54:35Z +issue,73848,torch.distributions.multinomial.Multinomial (an example mistake of docs)?,https://github.com/pytorch/pytorch/issues/73848,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs",scan when time allows,2022-03-07T14:52:52Z +issue,73832,Legacy sparse tensor constructor (e.g. torch.cuda.sparse.FloatTensor) silently ignores device kwarg,https://github.com/pytorch/pytorch/issues/73832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2022-03-07T14:50:54Z +issue,73699,How to get tolerance override in OpInfo-based test?,https://github.com/pytorch/pytorch/issues/73699,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: testing",scan when time allows,2022-03-07T14:42:39Z +issue,46802,"DataLoader gives ""Broken pipe"" error on Linux platform",https://github.com/pytorch/pytorch/issues/46802,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2022-03-07T02:54:02Z +issue,73176,"Cuda lacks checking of ""out of bound""",https://github.com/pytorch/pytorch/issues/73176,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking, module: loss, module: nn",scan when time allows,2022-03-06T22:38:35Z +issue,73711,"Refactor/Cleanup LazyTensor, LTCTensorImpl, Data",https://github.com/pytorch/pytorch/issues/73711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-03-05T00:12:08Z +issue,73755,CI: Bake as many dependencies as we can in the AMI (windows),https://github.com/pytorch/pytorch/issues/73755,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-03-04T19:23:55Z +issue,64048,"Be able to use pytest to run a ""core"" set of tests",https://github.com/pytorch/pytorch/issues/64048,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-03-04T19:09:52Z +issue,73083,Improve test_test_history.py,https://github.com/pytorch/pytorch/issues/73083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-03-04T19:01:27Z +issue,72555,Enforce quotas on CI users,https://github.com/pytorch/pytorch/issues/72555,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-03-04T19:00:06Z +issue,45242,Functional interface for optimizers,https://github.com/pytorch/pytorch/issues/45242,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: optimizer",scan when time allows,2022-03-04T16:20:11Z +issue,73640,[Feature Request] Allow torch.Generator to be passed to torch.nn.init functions,https://github.com/pytorch/pytorch/issues/73640,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-03-03T21:45:06Z +issue,73709,DropIT: Dropping Intermediate Tensors for Memory-Efficient DNN Training,https://github.com/pytorch/pytorch/issues/73709,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2022-03-03T18:36:41Z +issue,24205,deprecate cuda arch 3.5/3.7 in nightlies,https://github.com/pytorch/pytorch/issues/24205,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-03-03T17:37:36Z +issue,71951,Multivariate normal defined by eigendecomposition,https://github.com/pytorch/pytorch/issues/71951,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-03-03T11:23:04Z +issue,60858,Sparse CSR layout CPU backend tracking issue,https://github.com/pytorch/pytorch/issues/60858,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkl, module: sparse",scan when time allows,2022-03-03T08:28:54Z +issue,37384,Error running trace on Pytorch Crowd Counting model,https://github.com/pytorch/pytorch/issues/37384,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-03-03T07:53:32Z +issue,21645,Batch Dataloader and Dataset,https://github.com/pytorch/pytorch/issues/21645,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-03-03T02:23:45Z +issue,73475,different results on each batch_size in torch==1.10.2+cu113 on RTX 3080,https://github.com/pytorch/pytorch/issues/73475,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility,scan when time allows,2022-03-02T13:55:11Z +issue,69359,"[feature request] More masked reductions: amin/amax, argmin/argmax, quantile, mean/var/std/std_mean/var_mean",https://github.com/pytorch/pytorch/issues/69359,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: masked operators, module: sparse",scan when time allows,2022-03-02T11:46:37Z +issue,285,Keyword arguments passed to module's __call__ aren't forwarded to the hooks,https://github.com/pytorch/pytorch/issues/285,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-03-01T23:47:49Z +issue,1178,Batched sparse QR factorizations and solves with cusolver,https://github.com/pytorch/pytorch/issues/1178,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: sparse",scan when time allows,2022-03-01T21:02:45Z +issue,73585,Add optional log_scale argument for torch.distributions.Normal,https://github.com/pytorch/pytorch/issues/73585,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-03-01T19:24:19Z +issue,73424,"""munmap_chunk(): invalid pointer"" interaction error with pytorch (< 1.10), pybind, and cv_bridge",https://github.com/pytorch/pytorch/issues/73424,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lts,scan when time allows,2022-03-01T18:13:07Z +issue,842,BCELoss doesn't accept LongTensor targets,https://github.com/pytorch/pytorch/issues/842,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2022-03-01T18:02:40Z +issue,711,Feature Request: Easier to extend base RNN implementation,https://github.com/pytorch/pytorch/issues/711,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-01T17:08:45Z +issue,630,Add Peephole connections for LSTMs?,https://github.com/pytorch/pytorch/issues/630,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-03-01T17:08:43Z +issue,88,expose backend selection and cudnn settings to the end user,https://github.com/pytorch/pytorch/issues/88,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2022-03-01T17:08:38Z +issue,73137,`create_graph=True` results in grad_fn error for differentiable functions,https://github.com/pytorch/pytorch/issues/73137,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-03-01T16:56:33Z +issue,65253,TestCppExtensionJit's test_crash_handler fails for force_on_cpu job on GHA,https://github.com/pytorch/pytorch/issues/65253,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-03-01T15:24:20Z +issue,73196,`torch.pow` errors out on specific input,https://github.com/pytorch/pytorch/issues/73196,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-03-01T14:33:59Z +issue,73150,How do we handle metadata-modifying in-place operators (like `squeeze_`) with `__torch_dispatch__`?,https://github.com/pytorch/pytorch/issues/73150,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-03-01T14:31:19Z +issue,73268,Add detection of interned submodule of externed module in PackageExporter,https://github.com/pytorch/pytorch/issues/73268,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deploy,scan when time allows,2022-03-01T14:30:55Z +issue,73218,"This is not completely a bug, but something that will be amazing if can be taken care of regarding nn.DataParallel.",https://github.com/pytorch/pytorch/issues/73218,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2022-03-01T14:27:29Z +issue,73425,ufunc codegen support for dtypes that are supported on CUDA but not CPU,https://github.com/pytorch/pytorch/issues/73425,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-03-01T14:22:07Z +issue,73505,Figure out what to do with functions that take both Tensor and TensorOptions,https://github.com/pytorch/pytorch/issues/73505,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-03-01T14:07:08Z +issue,73517,Teach tools.codegen.api.translate about IValues,https://github.com/pytorch/pytorch/issues/73517,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: codegen",scan when time allows,2022-03-01T14:05:51Z +issue,73506,Idiom for extensible string printing for TensorImpl subclasses,https://github.com/pytorch/pytorch/issues/73506,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: printing",scan when time allows,2022-03-01T14:05:12Z +issue,44726,Pytorch Installation from source fails (GCC-8.4/CUDA-10.2/RHEL7) and (GCC-7.5/CUDA-10.2/RHEL7),https://github.com/pytorch/pytorch/issues/44726,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-03-01T09:05:41Z +issue,29042,Support clang+cuda builds,https://github.com/pytorch/pytorch/issues/29042,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-03-01T06:18:00Z +issue,64711,Improve Docs builds to not have to push to a repository to update,https://github.com/pytorch/pytorch/issues/64711,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2022-02-28T20:30:14Z +issue,67712,"Decorate tests with a ""deadline""",https://github.com/pytorch/pytorch/issues/67712,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-02-28T20:20:30Z +issue,68017,MacOS CI result is not in /README.md,https://github.com/pytorch/pytorch/issues/68017,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: macos",scan when time allows,2022-02-28T19:47:25Z +issue,70455,Make it possible to remove all hooks on a specified module without needing the hook handles,https://github.com/pytorch/pytorch/issues/70455,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-28T18:07:42Z +issue,73413,Conv3D consumes lots of memory on Mac with Apple Silicon,https://github.com/pytorch/pytorch/issues/73413,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: macos",scan when time allows,2022-02-28T16:27:58Z +issue,73484,Rename `keep_vars` in `nn.Module.state_dict`,https://github.com/pytorch/pytorch/issues/73484,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-28T15:48:14Z +issue,70419,LazyModules `cls_to_become` field exposes implementation detail,https://github.com/pytorch/pytorch/issues/70419,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2022-02-28T09:03:00Z +issue,70500,Adding the new Phish activation function,https://github.com/pytorch/pytorch/issues/70500,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-28T08:03:06Z +issue,70447,The same code can not be reproduced on multiple GPUs with dataparallel,https://github.com/pytorch/pytorch/issues/70447,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2022-02-27T22:08:16Z +issue,73161,native_batch_norm and native_layer_norm have strange epsilon behaviors,https://github.com/pytorch/pytorch/issues/73161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2022-02-27T21:29:17Z +issue,70490,Cuda sync mode input checking is wrong for non-string/int inputs.,https://github.com/pytorch/pytorch/issues/70490,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-02-27T14:03:19Z +issue,70413,"PyTorch crashes without an error message, when running this code snippet with torch.tensor subclassing & forward hooks (Not sure what the exact cause is, but the code snippet reliably causes it)",https://github.com/pytorch/pytorch/issues/70413,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-02-26T21:02:46Z +issue,70416,libtorch cuda use too much system memory,https://github.com/pytorch/pytorch/issues/70416,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2022-02-26T21:02:45Z +issue,73461,`TestCommonCUDA.test_noncontiguous_samples_pca_lowrank_cuda_float32` fails on A100 due to TF32 operation in `svd_lowrank`,https://github.com/pytorch/pytorch/issues/73461,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: tests, module: tf32",scan when time allows,2022-02-26T19:07:30Z +issue,70386,AT_ASSERT fail with DataLoaderOptions().drop_last(),https://github.com/pytorch/pytorch/issues/70386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-02-26T19:02:51Z +issue,70388,RuntimeError: tensor has too many (>25) dims when permuting tensor with GPU backend,https://github.com/pytorch/pytorch/issues/70388,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: assert failure,scan when time allows,2022-02-26T17:02:56Z +issue,70391,linalg.lstsq INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/70391,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: mkl",scan when time allows,2022-02-26T17:02:55Z +issue,65774,Non-persistent modules,https://github.com/pytorch/pytorch/issues/65774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization",scan when time allows,2022-02-26T15:59:50Z +issue,73417,MaybeEncodingError: Error sending result,https://github.com/pytorch/pytorch/issues/73417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2022-02-26T00:41:44Z +issue,72525,KL divergence between two Continuous Bernoulli is negative,https://github.com/pytorch/pytorch/issues/72525,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-02-25T15:15:06Z +issue,73412,Build from source failed,https://github.com/pytorch/pytorch/issues/73412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-02-25T11:10:08Z +issue,73355,Review and refactor the way libcublas static libraries are linked,https://github.com/pytorch/pytorch/issues/73355,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cublas, oncall: releng",scan when time allows,2022-02-25T04:38:15Z +issue,73380,elastic/rendezvous: _matches_machine_hostname doesn't resolve hostnames fully,https://github.com/pytorch/pytorch/issues/73380,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: r2p",scan when time allows,2022-02-25T04:36:10Z +issue,73394,Make it easier to figure out if packages need to be interned/mocked/externed,https://github.com/pytorch/pytorch/issues/73394,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-02-25T04:34:53Z +issue,72788,Docs bug: type annotations for linspace (and logspace) start and end arguments is wrong,https://github.com/pytorch/pytorch/issues/72788,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: tensor creation",scan when time allows,2022-02-24T10:14:09Z +issue,61758,`make_tensor` tracking issue,https://github.com/pytorch/pytorch/issues/61758,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2022-02-24T09:58:38Z +issue,65640,[LTC] Memory growing up during training until OOM,https://github.com/pytorch/pytorch/issues/65640,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2022-02-23T18:57:23Z +issue,45321,Memory leak when using share_memory_ on cuda device,https://github.com/pytorch/pytorch/issues/45321,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2022-02-23T17:18:11Z +issue,72194,`TestCase.assertEqual` has `equal_nan` default to `True`,https://github.com/pytorch/pytorch/issues/72194,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: testing, module: tests",scan when time allows,2022-02-23T17:16:39Z +issue,73051,Cannot install latest pytorch into Docker on Apple M1,https://github.com/pytorch/pytorch/issues/73051,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries, module: macos",scan when time allows,2022-02-22T20:20:33Z +issue,71698,[libtorch]can not save a vector to AutogradContex->saved_data.,https://github.com/pytorch/pytorch/issues/71698,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpp",scan when time allows,2022-02-22T20:06:01Z +issue,73222,Having rrelu functional + module take a generator object to match native functions entry,https://github.com/pytorch/pytorch/issues/73222,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: random",scan when time allows,2022-02-22T18:58:08Z +issue,73134,Test functionalization doesn't run,https://github.com/pytorch/pytorch/issues/73134,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2022-02-22T14:48:39Z +issue,72240,recurrent neural network module,https://github.com/pytorch/pytorch/issues/72240,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: rnn",scan when time allows,2022-02-22T12:07:57Z +issue,72650,Some inplace ops don't raise on incompatible shapes and meta device,https://github.com/pytorch/pytorch/issues/72650,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: meta tensors, module: tests",scan when time allows,2022-02-22T11:53:32Z +issue,72537,Feature: Add tril_embed and triu_embed,https://github.com/pytorch/pytorch/issues/72537,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: tensor creation",scan when time allows,2022-02-22T11:49:56Z +issue,50098,Errors when coercing complex numbers of various sizes,https://github.com/pytorch/pytorch/issues/50098,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: half",scan when time allows,2022-02-22T11:23:41Z +issue,70346,Add hints for gradient long time overflow when using torch.cuda.amp,https://github.com/pytorch/pytorch/issues/70346,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda",scan when time allows,2022-02-21T18:02:53Z +issue,56891,Batched SVD_LOWRANK being much slower than loop implementation (both CPU and GPU),https://github.com/pytorch/pytorch/issues/56891,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2022-02-21T06:49:19Z +issue,43754,"Using Dataparallel with multi input error. Input and parameter tensors are not at the same device, found input tensor at cpu and parameter tensor at cuda:0",https://github.com/pytorch/pytorch/issues/43754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2022-02-20T23:52:44Z +issue,70171,Channels last performance problem,https://github.com/pytorch/pytorch/issues/70171,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: performance",scan when time allows,2022-02-20T07:02:51Z +issue,67457,"Reductions on zero-size dims: 1) by accepting a custom default value, 2) if tensor has another non-reduced zero-size dim",https://github.com/pytorch/pytorch/issues/67457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2022-02-19T14:26:33Z +issue,68553,[distributed] elastic/agent/server/api.py could react faster on exit / Ctrl+D / Ctrl+C,https://github.com/pytorch/pytorch/issues/68553,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: r2p,scan when time allows,2022-02-19T11:26:54Z +issue,73034,Is it intentional that PyTorch linux binaries aren't manylinux1 compliant?,https://github.com/pytorch/pytorch/issues/73034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-02-18T23:39:18Z +issue,69665,No dtype check for zero sparse tensor!,https://github.com/pytorch/pytorch/issues/69665,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-02-18T23:14:54Z +issue,73121,make c++ logger preamble meaningful,https://github.com/pytorch/pytorch/issues/73121,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: c10d,scan when time allows,2022-02-18T23:06:23Z +issue,70199,Missing Docker image for 1.10.1,https://github.com/pytorch/pytorch/issues/70199,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docker",scan when time allows,2022-02-18T23:02:50Z +issue,70158,pin_memory *still* destroys custom containers,https://github.com/pytorch/pytorch/issues/70158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-02-18T20:03:04Z +issue,73062,"Clang Compilation Error: more than one constructor applies to convert from ""ptrdiff_t"" to ""c10::Scalar""",https://github.com/pytorch/pytorch/issues/73062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: macos",scan when time allows,2022-02-18T19:48:38Z +issue,73065,`index_fill_` accepts wrong dtype for meta tensors,https://github.com/pytorch/pytorch/issues/73065,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: meta tensors, module: tests",scan when time allows,2022-02-18T19:22:37Z +issue,71843,Migrate master to main: https://github.com/pytorch/tutorials,https://github.com/pytorch/pytorch/issues/71843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-02-18T15:47:01Z +issue,17614,Document whether it is possible to train TorchScript modules,https://github.com/pytorch/pytorch/issues/17614,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs, oncall: jit",scan when time allows,2022-02-18T08:47:18Z +issue,72580,"Setting a list of Modules as an attribute does not work like setting a Module as an attribute, and throws no warning",https://github.com/pytorch/pytorch/issues/72580,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-17T19:44:44Z +issue,73007,Feature: support better rendering for ..deprecated Sphinx directive,https://github.com/pytorch/pytorch/issues/73007,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-02-17T19:26:10Z +issue,72933,Why there are 8 flavors of iOS build jobs for every commit,https://github.com/pytorch/pytorch/issues/72933,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: ios",scan when time allows,2022-02-17T03:19:34Z +issue,72909,"`torch.svd_lowrank` should set the default value of `q` as `min(6, m, n)`",https://github.com/pytorch/pytorch/issues/72909,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-02-17T03:18:05Z +issue,72911,Vectorized Jacobian and Hessian errors with ffts,https://github.com/pytorch/pytorch/issues/72911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: vmap",scan when time allows,2022-02-16T22:27:47Z +issue,72915,Is the current behavior with addcmul and integer dtypes intended?,https://github.com/pytorch/pytorch/issues/72915,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-02-16T22:22:04Z +issue,72824,Some loss functions support `dtype` broadcast but some do not,https://github.com/pytorch/pytorch/issues/72824,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn, module: type promotion",scan when time allows,2022-02-16T14:45:38Z +issue,71473,pytorch installation error with cuda!!,https://github.com/pytorch/pytorch/issues/71473,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-02-16T08:39:27Z +issue,70138,build_ios.sh prevents iOS.cmake from configuring ios deployment target correctly,https://github.com/pytorch/pytorch/issues/70138,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ios",scan when time allows,2022-02-16T02:16:28Z +issue,66565,Bazel target all_tests improperly reports failures on CPU-only (non-CUDA) build,https://github.com/pytorch/pytorch/issues/66565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: bootcamp, module: tests",scan when time allows,2022-02-15T20:33:26Z +issue,63972,bazel build warning: Artifact 'torch/csrc/api/include/torch/version.h' is duplicated,https://github.com/pytorch/pytorch/issues/63972,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: build, module: build warnings",scan when time allows,2022-02-15T20:33:26Z +issue,46631,error in bazel build //...,https://github.com/pytorch/pytorch/issues/46631,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bazel, module: build",scan when time allows,2022-02-15T20:33:25Z +issue,70100,Feature Request: dim parameter of torch.nn.functional.normalize should accept tuples,https://github.com/pytorch/pytorch/issues/70100,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2022-02-15T20:03:04Z +issue,57121,"einsum ""jk,ijkl->il"" is ~16x slower than numpy",https://github.com/pytorch/pytorch/issues/57121,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: numpy, module: performance",scan when time allows,2022-02-15T18:43:27Z +issue,72850,N-dimensional Convolutions,https://github.com/pytorch/pytorch/issues/72850,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2022-02-15T16:56:18Z +issue,70095,Add support for a `default` arg in `ModuleDict.pop`,https://github.com/pytorch/pytorch/issues/70095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-15T16:02:59Z +issue,71078,"`torch.nn.{Constant,Zero}Pad` unexpectedly fail",https://github.com/pytorch/pytorch/issues/71078,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: padding",scan when time allows,2022-02-15T02:17:40Z +issue,71059,`torch.scatter` will return random value when `input` is empty tensor,https://github.com/pytorch/pytorch/issues/71059,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: scatter & gather ops",scan when time allows,2022-02-15T02:17:15Z +issue,69348,`torch.sspaddmm` should broadcast the input tensor,https://github.com/pytorch/pytorch/issues/69348,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-02-14T23:11:20Z +issue,60234,Numerical-reproducibility issue in torch.matmul,https://github.com/pytorch/pytorch/issues/60234,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-reproducibility",scan when time allows,2022-02-14T20:39:46Z +issue,72714,Add `pct_end` parameter to `OneCycleLR`,https://github.com/pytorch/pytorch/issues/72714,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2022-02-14T16:49:37Z +issue,69984,Docs for torch.nn.MSELoss are confusing,https://github.com/pytorch/pytorch/issues/69984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: loss, module: nn",scan when time allows,2022-02-14T15:02:55Z +issue,41162,Advanced indexing gradient is extremely slow when there are many duplicate indices,https://github.com/pytorch/pytorch/issues/41162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2022-02-14T14:46:20Z +issue,69408,`torch.hstack` should raise an error when tensor is 0 dimensional,https://github.com/pytorch/pytorch/issues/69408,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-02-14T13:43:11Z +issue,72784,Torch version in docker container does not match tag,https://github.com/pytorch/pytorch/issues/72784,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2022-02-14T05:19:14Z +issue,72768,Add NCCL and MPI version printing to torch.utils.collect_env,https://github.com/pytorch/pytorch/issues/72768,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: collect_env.py, module: mpi, module: nccl",scan when time allows,2022-02-14T04:44:34Z +issue,72452,Re-raise the exception when the `forward` of a parametrization raises,https://github.com/pytorch/pytorch/issues/72452,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: nn.utils.parametrize",scan when time allows,2022-02-14T01:03:12Z +issue,69960,"FeatureAlphaDropout doesn't drop channels for (C, D, H, W)",https://github.com/pytorch/pytorch/issues/69960,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: nn",scan when time allows,2022-02-13T23:02:56Z +issue,69912,"torch.nn.functional.ctc_loss with invalid input produce NaN or infinity gradient, while the batch entries are fine",https://github.com/pytorch/pytorch/issues/69912,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: loss, module: nn",scan when time allows,2022-02-13T22:02:41Z +issue,72672,nn.functional No-batch-dim support should have OpInfo examples,https://github.com/pytorch/pytorch/issues/72672,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: tests",scan when time allows,2022-02-13T21:15:11Z +issue,69786,Implement aten::equal for sparse tensors,https://github.com/pytorch/pytorch/issues/69786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-02-13T20:02:50Z +issue,41499,Helping test example code blocks in the docs,https://github.com/pytorch/pytorch/issues/41499,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: tests",scan when time allows,2022-02-13T19:30:49Z +issue,25661,finfo operator not bound into JIT,https://github.com/pytorch/pytorch/issues/25661,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2022-02-13T18:49:13Z +issue,71211,Lack of type check in `nn.functional` APIs,https://github.com/pytorch/pytorch/issues/71211,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-13T12:13:52Z +issue,69800,Incomplete error message at tensor indexing (when indexing with set),https://github.com/pytorch/pytorch/issues/69800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2022-02-12T00:06:01Z +issue,69753,last_epoch parameter of CyclicLR and OneCycleLR is not the number of epochs,https://github.com/pytorch/pytorch/issues/69753,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2022-02-11T16:03:08Z +issue,72179,"Mechanism for Tensor subclasses to ""disable autograd""",https://github.com/pytorch/pytorch/issues/72179,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: autograd",scan when time allows,2022-02-10T21:29:17Z +issue,70678,Decouple `TensorIteratorBase` output from structured kernel outputs.,https://github.com/pytorch/pytorch/issues/70678,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2022-02-10T21:15:09Z +issue,6107,"Well documented, safe method to deserialize model parameters from untrusted sources",https://github.com/pytorch/pytorch/issues/6107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pickle, module: serialization",scan when time allows,2022-02-10T15:52:49Z +issue,43120,"When modified the model python file, the pytorch will raise the KeyError of this file",https://github.com/pytorch/pytorch/issues/43120,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2022-02-10T04:37:16Z +issue,71364,make pytorch support different hardware acceleratioin configuration,https://github.com/pytorch/pytorch/issues/71364,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2022-02-09T21:32:53Z +issue,72258,NotImplementedError in torch.distributions,https://github.com/pytorch/pytorch/issues/72258,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-02-09T14:14:22Z +issue,72558,Pin dependencies + expand the current linter,https://github.com/pytorch/pytorch/issues/72558,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: ci",scan when time allows,2022-02-08T23:14:30Z +issue,69714,Possibly out of date error in autograd codegen,https://github.com/pytorch/pytorch/issues/69714,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-02-08T22:03:16Z +issue,43567,`torch.svd()` CUDA gives incorrect results when input contains `nan`,https://github.com/pytorch/pytorch/issues/43567,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: nans and infs",scan when time allows,2022-02-08T18:17:01Z +issue,22848,[feature request] Log-determinant for symmetric positive definite matrices,https://github.com/pytorch/pytorch/issues/22848,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-02-08T15:31:55Z +issue,4358,Invoking MKL in multiprocessing with importing torch causes blocking,https://github.com/pytorch/pytorch/issues/4358,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkl, module: mkldnn, module: multiprocessing",scan when time allows,2022-02-08T15:11:28Z +issue,69744,Add ability to ignore arguments/outputs in `torch.autograd.functional.jacobian`,https://github.com/pytorch/pytorch/issues/69744,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-02-08T15:03:02Z +issue,69756,Introduction of nn.LazyRNNs,https://github.com/pytorch/pytorch/issues/69756,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2022-02-08T15:03:01Z +issue,72041,RuntimeError: Trying to backward through the graph a second time (or directly access saved tensors after they have already been freed). Saved intermediate values of the graph are freed when you call .backward() or autograd.grad(). Specify retain_graph=True if you need to backward through the graph a second time or if you need to access saved tensors after calling backward,https://github.com/pytorch/pytorch/issues/72041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn.utils.parametrize",scan when time allows,2022-02-08T14:36:07Z +issue,68331,Casting real parameter to complex during forward produces warning on backward,https://github.com/pytorch/pytorch/issues/68331,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex",scan when time allows,2022-02-08T12:54:24Z +issue,58181,torch.split_with_sizes is not documented,https://github.com/pytorch/pytorch/issues/58181,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-02-08T10:38:46Z +issue,71575,Better Engineering: test_..._mem_overlap in test_torch.py should be ported to ErrorInputs,https://github.com/pytorch/pytorch/issues/71575,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-02-08T04:43:25Z +issue,72421,Feature Request: A simpler decorator for disabling mixed precision,https://github.com/pytorch/pytorch/issues/72421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2022-02-08T03:57:21Z +issue,71472,Feature request: Add complex support to `torch.nanmean`,https://github.com/pytorch/pytorch/issues/71472,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs, module: reductions",scan when time allows,2022-02-08T03:34:43Z +issue,66924,[BUG] Inconsistent initialization on different machines (tensor.uniform_()),https://github.com/pytorch/pytorch/issues/66924,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2022-02-08T03:01:07Z +issue,29844,Fold DispatchStub into c10 dispatcher,https://github.com/pytorch/pytorch/issues/29844,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2022-02-08T02:53:19Z +issue,69676,"[LazyTensor] model after to(device), traced IR changed and lose type info",https://github.com/pytorch/pytorch/issues/69676,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-02-08T00:06:58Z +issue,16737,torch.save overwrite,https://github.com/pytorch/pytorch/issues/16737,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-02-07T19:38:24Z +issue,16717,Performance regression on CPU from 0.4.1 to 1.0.0 on ResNet inference,https://github.com/pytorch/pytorch/issues/16717,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: single threaded",scan when time allows,2022-02-07T19:36:36Z +issue,4930,[feature request] Type-1 Multi-layer bidirectional RNN,https://github.com/pytorch/pytorch/issues/4930,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: rnn",scan when time allows,2022-02-07T19:34:23Z +issue,10298,[feature request] Runtime warning for inappropriate labels (among others),https://github.com/pytorch/pytorch/issues/10298,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: loss",scan when time allows,2022-02-07T18:35:11Z +issue,3667,Exposing CuDNN benchmark strategy selection,https://github.com/pytorch/pytorch/issues/3667,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: cudnn",scan when time allows,2022-02-07T18:29:56Z +issue,72341,InstanceNorm doesn't preserve memory format,https://github.com/pytorch/pytorch/issues/72341,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: norms and normalization",scan when time allows,2022-02-07T18:29:23Z +issue,69530,Extend composite compliant testing to backward formulas,https://github.com/pytorch/pytorch/issues/69530,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2022-02-06T22:02:51Z +issue,29372,torch.std() returns nan for single item tensors.,https://github.com/pytorch/pytorch/issues/29372,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2022-02-06T21:40:12Z +issue,69360,1.10.0 failed to build due to missing TensorBody.h,https://github.com/pytorch/pytorch/issues/69360,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-02-06T14:45:03Z +issue,69531,"`(svd|pca)_lowrank`: backward is unstable when for a matrix `A`, the parameter `q` is set to a value `q > rank(A)`.",https://github.com/pytorch/pytorch/issues/69531,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra, module: numerical-stability",scan when time allows,2022-02-05T22:03:30Z +issue,69516,`torch.sparse.softmax` and `torch.sparse.log_softmax` do not support negative dim.,https://github.com/pytorch/pytorch/issues/69516,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-02-05T20:02:43Z +issue,69442,The formula for KL-divloss is wrong in the document,https://github.com/pytorch/pytorch/issues/69442,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2022-02-04T22:02:54Z +issue,72361,Add ZeroTensor fastpath for torch.baddbmm,https://github.com/pytorch/pytorch/issues/72361,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-02-04T20:31:23Z +issue,72358,Add ZeroTensor fastpath for torch.addbmm,https://github.com/pytorch/pytorch/issues/72358,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-02-04T20:28:14Z +issue,72347,Add ZeroTensor fastpath for torch.addcmul,https://github.com/pytorch/pytorch/issues/72347,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-02-04T18:37:39Z +issue,72332,Ability to assign to `tensor.require_grad` might lead to bugs,https://github.com/pytorch/pytorch/issues/72332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: ux",scan when time allows,2022-02-04T16:47:29Z +issue,69921,Context manager to enable/disabled TensorFloat32 on demand,https://github.com/pytorch/pytorch/issues/69921,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tf32,scan when time allows,2022-02-03T19:08:27Z +issue,62812,Support `torch.linalg.einsum`,https://github.com/pytorch/pytorch/issues/62812,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-02-02T19:00:26Z +issue,71359,"Torchvision Installation Logic Python vs C++ ""Mismatch""",https://github.com/pytorch/pytorch/issues/71359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-02-02T17:15:13Z +issue,71697,Libtorch dlls delayed loading,https://github.com/pytorch/pytorch/issues/71697,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: static linking",scan when time allows,2022-02-02T16:12:24Z +issue,32030,Dockerfile for people to quick-start contributing,https://github.com/pytorch/pytorch/issues/32030,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2022-02-02T14:07:53Z +issue,71918,Support mixed python scalar/tensor types for torch.clamp's min/max args,https://github.com/pytorch/pytorch/issues/71918,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2022-02-02T14:06:34Z +issue,64645,Compilation from source fails,https://github.com/pytorch/pytorch/issues/64645,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: build, module: protobuf",scan when time allows,2022-02-01T22:25:32Z +issue,71495,Memory Leak in PyTorch 1.10.1,https://github.com/pytorch/pytorch/issues/71495,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2022-02-01T19:27:20Z +issue,71340,locally installed PyTorch upgraded or superseded on windows/macos conda tests,https://github.com/pytorch/pytorch/issues/71340,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lts,scan when time allows,2022-02-01T19:03:32Z +issue,71877,Add `output_size` argument to `Upsample` forward method (just like for `ConvTranspose` Modules),https://github.com/pytorch/pytorch/issues/71877,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-01T17:59:49Z +issue,69325,`nn.functional.fractional_max_pool2d` and `nn.functional.fractional_max_pool3d` produce incorrect output on non-contiguous inputs,https://github.com/pytorch/pytorch/issues/69325,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-02-01T16:03:13Z +issue,14907,Missing dilation from several pooling modules (AvgPool),https://github.com/pytorch/pytorch/issues/14907,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2022-02-01T15:03:11Z +issue,69286,"subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '32']' returned non-zero exit status 1.",https://github.com/pytorch/pytorch/issues/69286,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-02-01T02:14:34Z +issue,72046,Building PyTorch with Vulkan backend don't work,https://github.com/pytorch/pytorch/issues/72046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vulkan,scan when time allows,2022-02-01T01:24:27Z +issue,72055,index_add : Inconsistent between CPU and CUDA,https://github.com/pytorch/pytorch/issues/72055,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda",scan when time allows,2022-02-01T01:20:50Z +issue,71671,Many APIs `INTERNAL ASSERT FAILED` when promoting `complex32` dtype,https://github.com/pytorch/pytorch/issues/71671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2022-02-01T01:16:51Z +issue,71635,"`torch.nn.functional.{instance, batch}_norm` trigger INTERNAL ASSERT FAILED when input is empty tensor with `complex32`",https://github.com/pytorch/pytorch/issues/71635,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: half",scan when time allows,2022-01-31T20:43:29Z +issue,69287,`out=` variant for `conj` unary operator,https://github.com/pytorch/pytorch/issues/69287,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2022-01-31T20:02:57Z +issue,70654,[RFC] UCC integration in ProcessGroupNCCL,https://github.com/pytorch/pytorch/issues/70654,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-01-31T19:29:40Z +issue,69281,Implement reparameterized sampling for LKJCholesky distribution w.r.t. concentration parameter,https://github.com/pytorch/pytorch/issues/69281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-01-31T18:02:48Z +issue,69290,Deprecation warnings generated when including header files in C++17 code,https://github.com/pytorch/pytorch/issues/69290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2022-01-31T18:02:46Z +issue,69288,Inplace and `out` variants for `positive` operator in PyTorch,https://github.com/pytorch/pytorch/issues/69288,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2022-01-31T16:03:54Z +issue,72061,"`svd_backward`: does not handle inputs of rank `r < min(m, n)`.",https://github.com/pytorch/pytorch/issues/72061,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra",scan when time allows,2022-01-31T15:49:56Z +issue,54047,RuntimeError: cuDNN error: CUDNN_STATUS_NOT_INITIALIZED in pytorch lightning,https://github.com/pytorch/pytorch/issues/54047,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2022-01-31T05:11:18Z +issue,71911,Functions depending on SVD are broken for inputs with non-finite values with MKL 2022+ and OpenBLAS 0.3.15+,https://github.com/pytorch/pytorch/issues/71911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: mkl, module: openblas",scan when time allows,2022-01-31T03:43:21Z +issue,72022,[PT-D] To make the ShardedTensor Reshard more generic,https://github.com/pytorch/pytorch/issues/72022,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2022-01-31T03:34:18Z +issue,71842,Migrate master to main: https://github.com/pytorch/ignite,https://github.com/pytorch/pytorch/issues/71842,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:32:33Z +issue,71841,Migrate master to main: https://github.com/pytorch/ELF,https://github.com/pytorch/pytorch/issues/71841,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:32:33Z +issue,71815,Migrate master to main: https://github.com/pytorch/add-annotations-github-action,https://github.com/pytorch/pytorch/issues/71815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:32:03Z +issue,71814,Migrate master to main: https://github.com/pytorch/ossci-job-dsl,https://github.com/pytorch/pytorch/issues/71814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:32:02Z +issue,71813,Migrate master to main: https://github.com/pytorch/pytorch-integration-testing,https://github.com/pytorch/pytorch/issues/71813,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:32:02Z +issue,71812,Migrate master to main: https://github.com/pytorch/pytorch-ci-dockerfiles,https://github.com/pytorch/pytorch/issues/71812,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:32:01Z +issue,71810,Migrate master to main: https://github.com/pytorch/labeler-github-action,https://github.com/pytorch/pytorch/issues/71810,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:32:00Z +issue,71840,Migrate master to main: https://github.com/pytorch/captum,https://github.com/pytorch/pytorch/issues/71840,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:35Z +issue,71839,Migrate master to main: https://github.com/pytorch/glow,https://github.com/pytorch/pytorch/issues/71839,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:35Z +issue,71838,Migrate master to main: https://github.com/pytorch/serve,https://github.com/pytorch/pytorch/issues/71838,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:34Z +issue,71837,Migrate master to main: https://github.com/pytorch/xla,https://github.com/pytorch/pytorch/issues/71837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:34Z +issue,71836,Migrate master to main: https://github.com/pytorch/QNNPACK,https://github.com/pytorch/pytorch/issues/71836,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:34Z +issue,71835,Migrate master to main: https://github.com/pytorch/tnt,https://github.com/pytorch/pytorch/issues/71835,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:33Z +issue,71834,Migrate master to main: https://github.com/pytorch/hub,https://github.com/pytorch/pytorch/issues/71834,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:33Z +issue,71833,Migrate master to main: https://github.com/pytorch/extension-cpp,https://github.com/pytorch/pytorch/issues/71833,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:32Z +issue,71832,Migrate master to main: https://github.com/pytorch/android-demo-app,https://github.com/pytorch/pytorch/issues/71832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:32Z +issue,71831,Migrate master to main: https://github.com/pytorch/translate,https://github.com/pytorch/pytorch/issues/71831,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:32Z +issue,71830,Migrate master to main: https://github.com/pytorch/elastic,https://github.com/pytorch/pytorch/issues/71830,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:31Z +issue,71828,Migrate master to main: https://github.com/pytorch/tvm,https://github.com/pytorch/pytorch/issues/71828,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:31Z +issue,71826,Migrate master to main: https://github.com/pytorch/ios-demo-app,https://github.com/pytorch/pytorch/issues/71826,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:30Z +issue,71825,Migrate master to main: https://github.com/pytorch/accimage,https://github.com/pytorch/pytorch/issues/71825,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:29Z +issue,71824,Migrate master to main: https://github.com/pytorch/extension-ffi,https://github.com/pytorch/pytorch/issues/71824,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:29Z +issue,71823,Migrate master to main: https://github.com/pytorch/nestedtensor,https://github.com/pytorch/pytorch/issues/71823,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:29Z +issue,71822,Migrate master to main: https://github.com/pytorch/cppdocs,https://github.com/pytorch/pytorch/issues/71822,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:28Z +issue,71821,Migrate master to main: https://github.com/pytorch/workshops,https://github.com/pytorch/pytorch/issues/71821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:28Z +issue,71820,Migrate master to main: https://github.com/pytorch/extension-script,https://github.com/pytorch/pytorch/issues/71820,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:27Z +issue,71819,Migrate master to main: https://github.com/pytorch/java-demo,https://github.com/pytorch/pytorch/issues/71819,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:27Z +issue,71818,Migrate master to main: https://github.com/pytorch/csprng,https://github.com/pytorch/pytorch/issues/71818,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:27Z +issue,71817,Migrate master to main: https://github.com/pytorch/pytorch_sphinx_theme,https://github.com/pytorch/pytorch/issues/71817,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:26Z +issue,71816,Migrate master to main: https://github.com/pytorch/rfcs,https://github.com/pytorch/pytorch/issues/71816,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: infra,scan when time allows,2022-01-31T03:31:26Z +issue,69207,Function(s) expecting a tuple argument don't accept generators,https://github.com/pytorch/pytorch/issues/69207,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-01-30T21:02:51Z +issue,69233,[ROCM] elementwise kernel launch,https://github.com/pytorch/pytorch/issues/69233,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2022-01-30T21:02:48Z +issue,69072,Libtorch dll versioning,https://github.com/pytorch/pytorch/issues/69072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2022-01-30T11:02:49Z +issue,69086,Confusing value of `TORCH_CUDA_ARCH_LIST`,https://github.com/pytorch/pytorch/issues/69086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-01-29T22:02:56Z +issue,54309,[discussion] Support torch.matmul: strided x sparse in addition to sparse x strided,https://github.com/pytorch/pytorch/issues/54309,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-01-29T21:54:27Z +issue,56126,AttributeError: module 'torch._C' has no attribute 'ComplexDoubleStorageBase',https://github.com/pytorch/pytorch/issues/56126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-01-29T10:13:07Z +issue,69013,libtorch_cuda links against wrong libnccl.so,https://github.com/pytorch/pytorch/issues/69013,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: build, module: third_party",scan when time allows,2022-01-29T00:05:27Z +issue,71633,Negative Exponents of Int tensors result in output of zero,https://github.com/pytorch/pytorch/issues/71633,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-01-28T01:41:58Z +issue,63992,Direct inversion and linear systems solutions for small matrices,https://github.com/pytorch/pytorch/issues/63992,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",scan when time allows,2022-01-27T13:07:39Z +issue,71377,is_alias_of support for storageless tensors,https://github.com/pytorch/pytorch/issues/71377,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: lazy, module: partial aliasing",scan when time allows,2022-01-26T22:39:55Z +issue,65216,Add function that retrieves a batch from a DataLoader,https://github.com/pytorch/pytorch/issues/65216,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2022-01-26T18:56:33Z +issue,70559,Behavior of torch.nn.functional.interpolate with unchanged output size and recompute_scale_factor=False,https://github.com/pytorch/pytorch/issues/70559,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn",scan when time allows,2022-01-26T17:14:42Z +issue,66066,Finishing OpInfos: test_autograd.py,https://github.com/pytorch/pytorch/issues/66066,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: tests",scan when time allows,2022-01-26T15:07:45Z +issue,64502,[numpy] torch.nonzero is similar to np.argwhere not np.nonzero,https://github.com/pytorch/pytorch/issues/64502,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2022-01-26T06:23:10Z +issue,70954,Training grouped Conv2D is slow,https://github.com/pytorch/pytorch/issues/70954,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: performance",scan when time allows,2022-01-26T02:11:38Z +issue,71760,Adapting the citation style according to GitHub's CFF,https://github.com/pytorch/pytorch/issues/71760,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2022-01-25T20:00:18Z +issue,42861,Can not get pytorch working with tensorboard,https://github.com/pytorch/pytorch/issues/42861,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2022-01-25T18:00:42Z +issue,64635,'LSTM' object has no attribute '_flat_weights_names',https://github.com/pytorch/pytorch/issues/64635,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2022-01-25T15:16:28Z +issue,71204,`torch.diag` unexpectedly fails,https://github.com/pytorch/pytorch/issues/71204,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: viewing and reshaping",scan when time allows,2022-01-25T05:56:51Z +issue,71660,Build release binaries with USE_GLOG=ON by default,https://github.com/pytorch/pytorch/issues/71660,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: infra",scan when time allows,2022-01-25T03:33:54Z +issue,32021,[discussion] Relax optimizer constructor constraints for simplicity,https://github.com/pytorch/pytorch/issues/32021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-01-25T00:34:16Z +issue,71682,_GLIBCXX_USE_CXX11_ABI=0 does not work when building from source code,https://github.com/pytorch/pytorch/issues/71682,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-01-24T23:57:31Z +issue,71689,torch.distributions.categorical.Categorical does not work with 0 batch size,https://github.com/pytorch/pytorch/issues/71689,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-01-24T23:47:31Z +issue,71396,a lot nightly builds are canceled due to VM errors since Jan14,https://github.com/pytorch/pytorch/issues/71396,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2022-01-24T23:42:34Z +issue,64124,ciflow tracking meta issues,https://github.com/pytorch/pytorch/issues/64124,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-01-24T20:44:33Z +issue,52851,"`embedding_bag(..., include_last_offset=True)` should always error if `offsets[-1] != input.size()`",https://github.com/pytorch/pytorch/issues/52851,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: nn",scan when time allows,2022-01-24T20:13:03Z +issue,55549,`embedding_bag` has unexpected behavior when given `offsets` that are not monotonically increasing,https://github.com/pytorch/pytorch/issues/55549,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: nn",scan when time allows,2022-01-24T20:12:29Z +issue,68803,[DDP] Verify ignored parameter names in debug mode during init,https://github.com/pytorch/pytorch/issues/68803,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-01-22T10:02:39Z +issue,68701,Tensor precision manifests differently between CPU and GPU.,https://github.com/pytorch/pytorch/issues/68701,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: cuda, module: jetson, module: numerical-stability",scan when time allows,2022-01-22T09:02:47Z +issue,68696,Is it possible to use torch.linalg.cholesky for cholesky decomposition of a huge matrix that doesn't fit in memory?,https://github.com/pytorch/pytorch/issues/68696,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-01-22T08:02:57Z +issue,71084,"`torch.{inverse,cholesky}` have wrong shape check of square matrices",https://github.com/pytorch/pytorch/issues/71084,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: linear algebra",scan when time allows,2022-01-22T06:17:36Z +issue,70916,uint8 scalar tensors cannot be used for integer indexing,https://github.com/pytorch/pytorch/issues/70916,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2022-01-22T01:55:37Z +issue,68781,[docs] Clarify DDP activation checkpointing support,https://github.com/pytorch/pytorch/issues/68781,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: docs, oncall: distributed",scan when time allows,2022-01-22T00:05:44Z +issue,68655,[ux] F.binary_cross_entropy (and maybe other losses) to auto-cast bool category mask to float,https://github.com/pytorch/pytorch/issues/68655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: ux",scan when time allows,2022-01-21T22:03:06Z +issue,68746,Add kl_divergence between Normal and Laplace distribution.,https://github.com/pytorch/pytorch/issues/68746,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2022-01-21T22:03:02Z +issue,71630,More than 4 Dimensions for Channel Last Memory Format,https://github.com/pytorch/pytorch/issues/71630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2022-01-21T17:06:38Z +issue,69491,[RFC] Support MemoryView for Tensors,https://github.com/pytorch/pytorch/issues/69491,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: internals, module: numpy, module: viewing and reshaping",scan when time allows,2022-01-21T15:15:43Z +issue,71541,"`torch.sub` behaves differently with `add`, `mul`, `div`",https://github.com/pytorch/pytorch/issues/71541,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2022-01-21T14:16:57Z +issue,71471,"[pytorch1.5.0] subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '6']' returned non-zero exit status 2.",https://github.com/pytorch/pytorch/issues/71471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-01-21T11:52:43Z +issue,71595,Optimizer Overlap: Follow up features,https://github.com/pytorch/pytorch/issues/71595,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2022-01-20T22:39:51Z +issue,71209,support setting `keepdim` without setting `dim`,https://github.com/pytorch/pytorch/issues/71209,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: python array api,scan when time allows,2022-01-20T22:06:39Z +issue,70454,Building docs locally fails,https://github.com/pytorch/pytorch/issues/70454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-01-20T21:57:11Z +issue,71543,Initialize DataLoader workers in parallel,https://github.com/pytorch/pytorch/issues/71543,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2022-01-20T17:51:24Z +issue,69524,[feature request] Accepting python scalar inputs for torch.minimum/torch.maximum and friends,https://github.com/pytorch/pytorch/issues/69524,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distance functions,scan when time allows,2022-01-20T09:43:20Z +issue,34951,"[feature request] Sparse (hybrid sparse-dense) output option for topk, min, max",https://github.com/pytorch/pytorch/issues/34951,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-01-20T06:24:16Z +issue,71475,The signature of `torch.nanmedian` in the doc is wrong,https://github.com/pytorch/pytorch/issues/71475,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nans and infs,scan when time allows,2022-01-20T03:25:39Z +issue,64744,Consolidate ProcessGroup allreduce_coalesced and allreduce,https://github.com/pytorch/pytorch/issues/64744,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: c10d, oncall: distributed",scan when time allows,2022-01-19T18:35:16Z +issue,71330,[RFC] Gossip SGD (as a DDP Communication Hook),https://github.com/pytorch/pytorch/issues/71330,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: optimizer, oncall: distributed",scan when time allows,2022-01-19T05:02:31Z +issue,49795,ninja: build stopped: subcommand failed.,https://github.com/pytorch/pytorch/issues/49795,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2022-01-19T01:49:09Z +issue,71414,Gamma distribution returns some wrong extreme values,https://github.com/pytorch/pytorch/issues/71414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: distributions, module: numerical-stability",scan when time allows,2022-01-19T01:00:13Z +issue,71383,Allow to shutdown persistent workers,https://github.com/pytorch/pytorch/issues/71383,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2022-01-18T16:53:42Z +issue,71407,"Push to fork failed with cryptic ""refusing to allow a Personal Access Token to create or update workflow `.github/workflows/run_torchbench.yml` without `workflow` scope""",https://github.com/pytorch/pytorch/issues/71407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2022-01-18T15:44:58Z +issue,71366,nn.Batchnorm1d input shape notation inconsistency,https://github.com/pytorch/pytorch/issues/71366,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: norms and normalization",scan when time allows,2022-01-18T15:13:55Z +issue,71331,Change the order of checks for tensor indexing errors,https://github.com/pytorch/pytorch/issues/71331,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2022-01-18T14:59:57Z +issue,71322,Allow users to pass use_reentrant=False to checkpoint_sequential,https://github.com/pytorch/pytorch/issues/71322,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2022-01-18T14:57:39Z +issue,71305,Convert a tensor with type caffe2::TypeMeta to std::vector,https://github.com/pytorch/pytorch/issues/71305,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2022-01-18T14:55:47Z +issue,71357,[feature request] Multidim support for softmax/log_softmax/softmin,https://github.com/pytorch/pytorch/issues/71357,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2022-01-18T14:53:57Z +issue,70735,named tensor doesn't work with deepcopy,https://github.com/pytorch/pytorch/issues/70735,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2022-01-18T06:21:29Z +issue,50343,RFC: identify analogous NumPy operators when documenting PyTorch operators,https://github.com/pytorch/pytorch/issues/50343,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy",scan when time allows,2022-01-17T19:25:29Z +issue,71029,"remote failure INTERNAL ASSERT FAILED at ""../torch/csrc/distributed/rpc/rref_context.cpp"":389",https://github.com/pytorch/pytorch/issues/71029,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2022-01-16T04:37:25Z +issue,62141,Dedupe code in functional optim classes,https://github.com/pytorch/pytorch/issues/62141,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: ddp",scan when time allows,2022-01-16T03:32:46Z +issue,71317,[torch.deploy] Using zipped torch modules in torch.deploy interpreter,https://github.com/pytorch/pytorch/issues/71317,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,scan when time allows,2022-01-14T18:47:48Z +issue,63085,Split up test_nn.py,https://github.com/pytorch/pytorch/issues/63085,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2022-01-14T15:35:35Z +issue,71266,"""Memory Leak"" when creating an iterator from a tensor",https://github.com/pytorch/pytorch/issues/71266,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2022-01-14T13:39:45Z +issue,31546,[Feature Request] Make torch.solve output NaN for singular matrix,https://github.com/pytorch/pytorch/issues/31546,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2022-01-13T17:20:04Z +issue,43114,"out kwargs are sometimes inconsistent with returned named tuple field name (and in some cases, cannot be made consistent)",https://github.com/pytorch/pytorch/issues/43114,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2022-01-13T17:01:49Z +issue,71205,Possible security issue of `torch.hub.load`,https://github.com/pytorch/pytorch/issues/71205,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2022-01-13T13:10:00Z +issue,49630,Bug in CosineAnnealingWarmRestarts,https://github.com/pytorch/pytorch/issues/49630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-01-13T03:36:06Z +issue,71228,1.10.11 fails to compile libtorch_cpu with -fopenmp,https://github.com/pytorch/pytorch/issues/71228,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: undefined reference",scan when time allows,2022-01-12T19:06:26Z +issue,17525,cmake fails with -DBUILD_PYTHON=OFF -DUSE_NNPACK=OFF,https://github.com/pytorch/pytorch/issues/17525,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: releng",scan when time allows,2022-01-12T11:05:09Z +issue,53712,Poor support of `Optimizer.add_param_group`,https://github.com/pytorch/pytorch/issues/53712,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2022-01-12T08:08:36Z +issue,71152,clang format hash mismatched for linux64,https://github.com/pytorch/pytorch/issues/71152,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: lint",scan when time allows,2022-01-12T03:59:15Z +issue,65576,[LTC] Code-gen IRs and TorchScript lowerings needed by running full TorchBench,https://github.com/pytorch/pytorch/issues/65576,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lazy, oncall: jit",scan when time allows,2022-01-11T20:46:10Z +issue,70995,Rewrite tests in test_nn to not depend on LAPACK,https://github.com/pytorch/pytorch/issues/70995,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: tests",scan when time allows,2022-01-11T19:23:16Z +issue,70608,Cannot compile C++ documentation: Sphynx assertion,https://github.com/pytorch/pytorch/issues/70608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-01-11T17:00:03Z +issue,25137,Serialization does not work for quantized modules,https://github.com/pytorch/pytorch/issues/25137,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2022-01-11T16:52:13Z +issue,39411,Allow to_here to interrupt blocking wait in the case of rpc.remote timeout,https://github.com/pytorch/pytorch/issues/39411,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2022-01-11T16:24:50Z +issue,70970,argmin/argmax incorrect doc for the first form,https://github.com/pytorch/pytorch/issues/70970,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: reductions",scan when time allows,2022-01-11T15:52:23Z +issue,71058,`torch.Tensor.where` cannot work when `y` is float,https://github.com/pytorch/pytorch/issues/71058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2022-01-11T15:36:54Z +issue,71076,Error in `torch.Tensor.logit` documentation,https://github.com/pytorch/pytorch/issues/71076,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2022-01-11T15:25:16Z +issue,70901,Accuracy problem of `torch.batch_norm_gather_stats_with_counts` when `running_mean` is half tensor,https://github.com/pytorch/pytorch/issues/70901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half, module: numerical-stability",scan when time allows,2022-01-11T14:56:56Z +issue,71134,Slowdown in torch.distributed.new_group when scaling to large clusters.,https://github.com/pytorch/pytorch/issues/71134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2022-01-11T02:36:12Z +issue,71071,Feature Request: torch.special.ellipe,https://github.com/pytorch/pytorch/issues/71071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2022-01-11T02:24:45Z +issue,70191,[libtorch] Loading in Java two differente libtorch_cpu.so from different versions fails,https://github.com/pytorch/pytorch/issues/70191,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: java",scan when time allows,2022-01-10T22:58:06Z +issue,46076,NCCL watchdog thread should log warnings about long-running GPU operations instead of silently hanging,https://github.com/pytorch/pytorch/issues/46076,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: c10d",scan when time allows,2022-01-10T21:50:16Z +issue,67571,[discussion] Have default size/shape equal to a scalar,https://github.com/pytorch/pytorch/issues/67571,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2022-01-10T19:59:28Z +issue,70397,`torch.empty_strided` works when the stride is negative!,https://github.com/pytorch/pytorch/issues/70397,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: tensor creation",scan when time allows,2022-01-10T18:15:49Z +issue,70398,`torch.broadcast_to` can create tensor with negative dimension.,https://github.com/pytorch/pytorch/issues/70398,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: shape checking",scan when time allows,2022-01-10T18:14:49Z +issue,71069,C++ torch::nn::Sequential clone() method overwrites child module names,https://github.com/pytorch/pytorch/issues/71069,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2022-01-10T16:08:30Z +issue,70931,Redefinition of `cub` namespace misses Debug,https://github.com/pytorch/pytorch/issues/70931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2022-01-08T22:31:53Z +issue,30987,Remove `.data`,https://github.com/pytorch/pytorch/issues/30987,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2022-01-08T04:06:41Z +issue,70940,"ThreadLocalState::setThreadLocalState is not setting the ""enabled"" flag of SavedTensorDefaultHooks",https://github.com/pytorch/pytorch/issues/70940,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2022-01-07T04:25:48Z +issue,59566,[discussion] Make docs consistent for torch.nn Modules,https://github.com/pytorch/pytorch/issues/59566,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2022-01-06T23:25:35Z +issue,11578,Request to import pytest in test/*.py,https://github.com/pytorch/pytorch/issues/11578,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2022-01-06T23:22:34Z +issue,67653,"TSAN issue in autograd ""set_next_edges""",https://github.com/pytorch/pytorch/issues/67653,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sanitizers",scan when time allows,2022-01-06T14:42:45Z +issue,6850,[feature request] More `index_*_` functionality and/or lambda functionality,https://github.com/pytorch/pytorch/issues/6850,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2022-01-06T13:00:19Z +issue,8853,Todo functions and autograd supports for Sparse Tensor,https://github.com/pytorch/pytorch/issues/8853,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2022-01-06T12:44:49Z +issue,56836,Remove single device constraint from ProcessGroupNCCL profiling,https://github.com/pytorch/pytorch/issues/56836,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: nccl",scan when time allows,2022-01-06T09:16:56Z +issue,68179,torch.nn.Linear wrong type annotation for bias,https://github.com/pytorch/pytorch/issues/68179,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: typing",scan when time allows,2022-01-06T02:49:57Z +issue,70521,tmpxft_00008487_00000000-6_THCStorage.compute_86.cudafe1.cpp:(.text+0x60b): additional relocation overflows omitted from the output when build PyTorch 1.8.2 from source,https://github.com/pytorch/pytorch/issues/70521,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2022-01-06T02:21:26Z +issue,42919,Legacy Python2 and early Python3 leftovers,https://github.com/pytorch/pytorch/issues/42919,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2022-01-06T02:20:13Z +issue,16014,[docs] Missing argument description (value) in scatter_ function documentation,https://github.com/pytorch/pytorch/issues/16014,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: scatter & gather ops",scan when time allows,2022-01-05T20:08:41Z +issue,69688,`torch.cuda.set_per_process_memory_fraction()` does not perform VRAM isolation,https://github.com/pytorch/pytorch/issues/69688,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: memory usage",scan when time allows,2022-01-05T17:52:26Z +issue,55296,Build errors with USE_VULKAN=ON when cross-compiling for Android,https://github.com/pytorch/pytorch/issues/55296,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: build",scan when time allows,2022-01-05T16:55:16Z +issue,70242,`TensorIterator`: provide a two-argument version of `set_output`,https://github.com/pytorch/pytorch/issues/70242,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: structured kernels, module: tensoriterator",scan when time allows,2022-01-05T16:20:53Z +issue,64023,A modest proposal: delete arithmetic overloads from c10::Half,https://github.com/pytorch/pytorch/issues/64023,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2022-01-05T04:40:43Z +issue,55655,Performance debugging / warning mode,https://github.com/pytorch/pytorch/issues/55655,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2022-01-05T04:39:27Z +issue,70487,Fusion of Convolution and BatchNorm,https://github.com/pytorch/pytorch/issues/70487,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2022-01-05T03:08:01Z +issue,69772,RuntimeError: t == DeviceType::CUDAINTERNAL ASSERT FAILED when trying to calculate gradients,https://github.com/pytorch/pytorch/issues/69772,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2022-01-04T14:58:32Z +issue,58741,`torch.size()` diverges from array API specification,https://github.com/pytorch/pytorch/issues/58741,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: python array api",scan when time allows,2022-01-04T10:39:11Z +issue,70498,Multiple invalid summaries in torch.nn documentation page,https://github.com/pytorch/pytorch/issues/70498,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2022-01-03T12:43:55Z +issue,70392,third_party/breakpad/ compilation failure,https://github.com/pytorch/pytorch/issues/70392,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: third_party,scan when time allows,2022-01-01T15:34:59Z +issue,39670,Force JIT to do type inference even when mypy annotated,https://github.com/pytorch/pytorch/issues/39670,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: typing, oncall: jit",scan when time allows,2021-12-30T15:08:12Z +issue,70342,Empty or NaN data pollute gradient even if they are not involved during backward,https://github.com/pytorch/pytorch/issues/70342,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nans and infs",scan when time allows,2021-12-29T12:35:48Z +issue,32268,Numpy array functionality in torchscript.,https://github.com/pytorch/pytorch/issues/32268,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-12-28T09:08:19Z +issue,70265,"Shape parameter inconsistency in torch.Tensor.view, torch.reshape, torch.Tensor.reshape",https://github.com/pytorch/pytorch/issues/70265,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: viewing and reshaping",scan when time allows,2021-12-27T22:47:17Z +issue,42336,Audit quantization functions to ensure proper argument sizes,https://github.com/pytorch/pytorch/issues/42336,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2021-12-27T16:44:20Z +issue,39495,Misannotation of layer_norm parameters causes internal assert failure,https://github.com/pytorch/pytorch/issues/39495,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2021-12-27T16:42:29Z +issue,36426,Any reference to LPPool2d,https://github.com/pytorch/pytorch/issues/36426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-12-27T12:37:22Z +issue,33800,Can't get module gradient in autograd.Function's custom backward when DataParallel is used,https://github.com/pytorch/pytorch/issues/33800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-12-26T22:23:38Z +issue,67136,addition of loss function RMSE in the torch.nn,https://github.com/pytorch/pytorch/issues/67136,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2021-12-24T15:15:15Z +issue,70241,CPU parallelization across batch has random faulty behavior on backward,https://github.com/pytorch/pytorch/issues/70241,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2021-12-23T23:00:31Z +issue,70184,Optimization: convolution_backward doesn't always need to call .contiguous on certain inputs,https://github.com/pytorch/pytorch/issues/70184,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: performance",scan when time allows,2021-12-23T16:55:57Z +issue,70180,Can you provide the torch.trt module to directly convert the pytorch weights to tensorrt?,https://github.com/pytorch/pytorch/issues/70180,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: unknown,scan when time allows,2021-12-23T15:39:44Z +issue,70222,Grad strides do not match bucket view strides,https://github.com/pytorch/pytorch/issues/70222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ddp,scan when time allows,2021-12-23T14:40:55Z +issue,70170,EmbeddingBag allows out of index ranges.!,https://github.com/pytorch/pytorch/issues/70170,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dependency bug,scan when time allows,2021-12-23T14:12:14Z +issue,56333,Unable to pip install PyTorch on M1 Mac - Errors,https://github.com/pytorch/pytorch/issues/56333,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: macos",scan when time allows,2021-12-22T15:41:25Z +issue,69893,subclassing torch.Tensor,https://github.com/pytorch/pytorch/issues/69893,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-12-21T15:28:41Z +issue,69932,"Would pytorch like some free multi-factor authentication (MFA) tokens from Google & GitHub, via the OpenSSF?",https://github.com/pytorch/pytorch/issues/69932,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-12-20T18:27:31Z +issue,66073,[feature request] BatchNorm frozen mode in core,https://github.com/pytorch/pytorch/issues/66073,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2021-12-20T16:23:08Z +issue,69822,Implement torch.*_like tensor creation functions on sparse inputs,https://github.com/pytorch/pytorch/issues/69822,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sparse, module: tensor creation",scan when time allows,2021-12-19T18:28:31Z +issue,69892,[feature request] Support `like=` argument in tensor factory methods,https://github.com/pytorch/pytorch/issues/69892,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: tensor creation",scan when time allows,2021-12-19T11:33:49Z +issue,19953,Performance difference between 0.4.1 and 1.1.0,https://github.com/pytorch/pytorch/issues/19953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-12-18T18:46:20Z +issue,69831,Incorrect error for integer `out=` dtypes when a float is expected.,https://github.com/pytorch/pytorch/issues/69831,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: structured kernels, module: type promotion",scan when time allows,2021-12-18T02:43:06Z +issue,69433,`torch.transpose` should raise an error when indexing 0 for 0 dimensional tensor.,https://github.com/pytorch/pytorch/issues/69433,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2021-12-17T15:24:48Z +issue,70040,Error in `torch.cdist` documentation,https://github.com/pytorch/pytorch/issues/70040,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: docs",scan when time allows,2021-12-17T14:43:25Z +issue,69386,Port `normal` to structured kernel,https://github.com/pytorch/pytorch/issues/69386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2021-12-17T05:00:49Z +issue,16954,torch.multiprocessing.pool.Pool broken,https://github.com/pytorch/pytorch/issues/16954,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-12-15T16:04:00Z +issue,41081,Add a function to convert SyncBatchNorm layers back to BatchNorm Layers,https://github.com/pytorch/pytorch/issues/41081,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: nn",scan when time allows,2021-12-15T12:59:09Z +issue,55912,Support float_qparams in quantized_clone,https://github.com/pytorch/pytorch/issues/55912,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2021-12-15T05:17:23Z +issue,55230,Profile Optionals and Optional[Tensor] specifically,https://github.com/pytorch/pytorch/issues/55230,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-12-15T01:42:41Z +issue,60156,PEP-585 type annotations not supported for class-level annotations,https://github.com/pytorch/pytorch/issues/60156,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-12-14T15:52:50Z +issue,69213,torch.stack should have a pin_memory parameter,https://github.com/pytorch/pytorch/issues/69213,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-12-14T11:26:01Z +issue,53696,Support GPU/CPU communication in RPC,https://github.com/pytorch/pytorch/issues/53696,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe, oncall: distributed",scan when time allows,2021-12-14T08:52:53Z +issue,69721,Alternative to keras StringLookup in Pytorch,https://github.com/pytorch/pytorch/issues/69721,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-12-13T18:22:40Z +issue,69654,Fused AlphaFold 2 modules,https://github.com/pytorch/pytorch/issues/69654,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-12-13T17:43:01Z +issue,67761,Allow LRScheduler to take in param_groups directly without an optimizer,https://github.com/pytorch/pytorch/issues/67761,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2021-12-13T15:24:48Z +issue,68979,Support for arbitrary schedulers in SequentialLR,https://github.com/pytorch/pytorch/issues/68979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lrscheduler, module: optimizer",scan when time allows,2021-12-13T15:23:32Z +issue,59137,[autograd] `fill_` gradgradcheck raises RuntimeError,https://github.com/pytorch/pytorch/issues/59137,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-12-13T14:44:37Z +issue,51575,[doc] being able to switch version while keeping the same doc,https://github.com/pytorch/pytorch/issues/51575,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2021-12-10T20:51:12Z +issue,69627,c2r fft input generation,https://github.com/pytorch/pytorch/issues/69627,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: tests",scan when time allows,2021-12-10T20:27:34Z +issue,42843,Documentation mistake of Adam in v1.6.0?,https://github.com/pytorch/pytorch/issues/42843,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2021-12-10T16:09:26Z +issue,42353,Error while trying to install pytorch on pycharm,https://github.com/pytorch/pytorch/issues/42353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-12-10T03:18:22Z +issue,59309,CUDA memory error when using torch.device('cpu'),https://github.com/pytorch/pytorch/issues/59309,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2021-12-09T22:03:49Z +issue,68909,Bug about Dropout CUDA Kernel,https://github.com/pytorch/pytorch/issues/68909,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2021-12-09T20:30:36Z +issue,33388,Sobol engine generates out-of-bounds samples after drawing too many samples,https://github.com/pytorch/pytorch/issues/33388,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2021-12-08T21:24:04Z +issue,33086,F.max_pool*d/F.min_pool*d should support integer dtypes and bool tensors,https://github.com/pytorch/pytorch/issues/33086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2021-12-08T11:15:12Z +issue,69453,Jacobian elliptic functions,https://github.com/pytorch/pytorch/issues/69453,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2021-12-08T05:21:33Z +issue,69452,Complete elliptic integral of the first kind,https://github.com/pytorch/pytorch/issues/69452,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2021-12-08T05:21:09Z +issue,69451,Exponentially scalable modified Bessel function of the second kind,https://github.com/pytorch/pytorch/issues/69451,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2021-12-08T05:20:17Z +issue,69363,"Citation request for ""probabilities for each class"" in the doc's description on cross-entropy",https://github.com/pytorch/pytorch/issues/69363,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-12-07T22:18:37Z +issue,69467,Distribution `cross_entropy` method,https://github.com/pytorch/pytorch/issues/69467,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-12-07T19:36:27Z +issue,69469,Distribution `covariance` property,https://github.com/pytorch/pytorch/issues/69469,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-12-07T19:32:45Z +issue,46024,Reproducibility breaks down with weighted Cross Entropy loss,https://github.com/pytorch/pytorch/issues/46024,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: determinism, module: loss, module: numerical-reproducibility",scan when time allows,2021-12-07T18:13:38Z +issue,55464,Support quantized functional linear/conv operators with 32b output,https://github.com/pytorch/pytorch/issues/55464,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2021-12-07T15:44:08Z +issue,29657,Improved detection of repeated observers,https://github.com/pytorch/pytorch/issues/29657,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2021-12-07T15:19:18Z +issue,58171,"Error while converting model for NNAPI ""Unsupported node kind ('quantized::batch_norm2d')""",https://github.com/pytorch/pytorch/issues/58171,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2021-12-06T20:54:51Z +issue,47546,Whats Pytorch's policy on adding support for a wider range of hardwares for training and inference?,https://github.com/pytorch/pytorch/issues/47546,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2021-12-04T11:33:13Z +issue,69199,Protocol buffers library mismatch.,https://github.com/pytorch/pytorch/issues/69199,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: protobuf, module: tensorboard",scan when time allows,2021-12-03T13:54:18Z +issue,62933,Compilation error with gcc11 and libstdc++11 (thread_id != operator removed),https://github.com/pytorch/pytorch/issues/62933,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mkldnn,scan when time allows,2021-12-03T09:35:06Z +issue,42502,[proposal] batch mode for randperm,https://github.com/pytorch/pytorch/issues/42502,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2021-12-03T04:01:20Z +issue,60932,Is Python version of Docker image on DockerHub downgraded?,https://github.com/pytorch/pytorch/issues/60932,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2021-12-03T01:11:25Z +issue,66410,CIFAR10 doesn't work on M1 MacBook,https://github.com/pytorch/pytorch/issues/66410,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: arm,scan when time allows,2021-12-02T14:33:57Z +issue,16899,"Give clearer guidance about multithreading in PyTorch, and how to disable it",https://github.com/pytorch/pytorch/issues/16899,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: multithreading",scan when time allows,2021-12-02T13:16:38Z +issue,4247,Feature request: sparse matrix max(axis),https://github.com/pytorch/pytorch/issues/4247,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-12-01T17:45:35Z +issue,67745,A `sieve` operation for separating values of a tensor into two tensors based on a condition.,https://github.com/pytorch/pytorch/issues/67745,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ux, module: viewing and reshaping",scan when time allows,2021-12-01T07:29:53Z +issue,68931,Dead lock happened ran pytorch 1.9.0 cuda11.2 on xeon gold 6326 cpu,https://github.com/pytorch/pytorch/issues/68931,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2021-12-01T03:36:40Z +issue,59875,[ROCm] test_gather_stress_cuda is flaky,https://github.com/pytorch/pytorch/issues/59875,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: rocm",scan when time allows,2021-11-30T22:11:04Z +issue,67937,Add support for LTC ops needed by backwards/autograd/autocast,https://github.com/pytorch/pytorch/issues/67937,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-11-30T17:36:55Z +issue,68925,How to implement `bucket_by_sequence_length` with IterableDataset and DataLoader,https://github.com/pytorch/pytorch/issues/68925,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2021-11-30T15:32:48Z +issue,52743,torch.searchsorted issues,https://github.com/pytorch/pytorch/issues/52743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy, module: sorting and selection",scan when time allows,2021-11-30T12:09:08Z +issue,68935,repeat_interleave hangs in a forked subprocess,https://github.com/pytorch/pytorch/issues/68935,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-11-30T01:08:06Z +issue,68967,[feature request] torch.argminmax,https://github.com/pytorch/pytorch/issues/68967,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-29T23:29:24Z +issue,62654,Checkpointing without re-entrant autograd,https://github.com/pytorch/pytorch/issues/62654,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-11-29T21:05:34Z +issue,68752,`softmin` and `softmax` operators support different input dtypes based on whether the `dtype` kwarg is passed,https://github.com/pytorch/pytorch/issues/68752,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: ux",scan when time allows,2021-11-29T18:20:48Z +issue,45412,Autocast but there are no kernels registered for this dispatch key. The operator is p�8ߎU Aborted (core dumped),https://github.com/pytorch/pytorch/issues/45412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2021-11-29T08:53:14Z +issue,27300,"Tensorboard add image with boxes, labels, and confidence scores.",https://github.com/pytorch/pytorch/issues/27300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2021-11-28T15:53:06Z +issue,29044,Add Wishart and inverse-Wishart distributions,https://github.com/pytorch/pytorch/issues/29044,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-11-28T01:05:21Z +issue,66083,[feature request] Allow broadcasting for F.huber_loss and potentially other losses,https://github.com/pytorch/pytorch/issues/66083,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: shape checking",scan when time allows,2021-11-26T00:20:12Z +issue,67751,Remove logic for constructing symbolic shapes in Profiler,https://github.com/pytorch/pytorch/issues/67751,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-25T00:23:05Z +issue,67687,"Trying to quantize and save a pre-trained deberta model, where i get a runtime error",https://github.com/pytorch/pytorch/issues/67687,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-23T18:32:36Z +issue,52747,[proposal] Make torchvision and other libraries' docs searchable via main docs search,https://github.com/pytorch/pytorch/issues/52747,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2021-11-23T17:19:27Z +issue,68622,"torch.{tensor, Tensor, LongTensor, ...} isn't captured under `enable_python_mode()`",https://github.com/pytorch/pytorch/issues/68622,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2021-11-22T14:27:45Z +issue,68422,There is a problem with GPU memory reclamation,https://github.com/pytorch/pytorch/issues/68422,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2021-11-21T04:01:03Z +issue,18796,[Feature Request] Flattened indices option for max pooling,https://github.com/pytorch/pytorch/issues/18796,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2021-11-19T23:25:57Z +issue,17538,ONNX->TensorRT parser library duplicated,https://github.com/pytorch/pytorch/issues/17538,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-19T23:10:14Z +issue,55374,[RFC] Extend Autocast to CPU/CUDA with BF16 data type,https://github.com/pytorch/pytorch/issues/55374,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: bfloat16",scan when time allows,2021-11-19T18:50:10Z +issue,47311,Quantized modules should properly implement __getstate__ and __setstate__ (copy.deepcopy doesn't work on quantized model),https://github.com/pytorch/pytorch/issues/47311,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2021-11-18T23:25:12Z +issue,68610,`torch.unique_consecutive`: passing positional optional arguments results in empty tensors,https://github.com/pytorch/pytorch/issues/68610,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2021-11-18T20:54:50Z +issue,68575,[ux] Print tensor's dtype by default,https://github.com/pytorch/pytorch/issues/68575,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2021-11-18T20:00:05Z +issue,68512,'replicate' padding in convolution is 77 times slower on cpu than 'zeros',https://github.com/pytorch/pytorch/issues/68512,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding, module: performance",scan when time allows,2021-11-18T18:08:13Z +issue,68420,`nn.functional.max_unpool(2|3)d`: failing shape check for correct inputs (with `dilation > 1`) with specified `output_size`,https://github.com/pytorch/pytorch/issues/68420,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2021-11-18T15:24:57Z +issue,68519,dataloader will miss batch data when num worker>0,https://github.com/pytorch/pytorch/issues/68519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-11-18T14:59:42Z +issue,68507,Multiprocessing hang and queue short circuiting,https://github.com/pytorch/pytorch/issues/68507,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: multiprocessing",scan when time allows,2021-11-18T09:05:49Z +issue,9624,Request: Pycuda interoperability,https://github.com/pytorch/pytorch/issues/9624,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-17T08:39:29Z +issue,68430,Strides issue on unsqueezed channels last tensor,https://github.com/pytorch/pytorch/issues/68430,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-11-17T08:14:47Z +issue,68498,Review disabling tests workflow,https://github.com/pytorch/pytorch/issues/68498,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-17T01:23:49Z +issue,62926,cudnn_convolution_add_relu fails under basic conditions,https://github.com/pytorch/pytorch/issues/62926,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: dependency bug",scan when time allows,2021-11-17T00:02:30Z +issue,68423,`torch.utils.checkpoint.checkpoint_sequential` is not optimal,https://github.com/pytorch/pytorch/issues/68423,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing,scan when time allows,2021-11-17T00:01:32Z +issue,68474,[LTC][BE] Fix all compile warnings,https://github.com/pytorch/pytorch/issues/68474,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-16T23:54:43Z +issue,68475,[LTC][BE] Make the compile flags more strict,https://github.com/pytorch/pytorch/issues/68475,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-16T23:54:33Z +issue,68337,"`nn.functional.max_unpool{n}d`: shape checks fail with `output_size=(C, ...)`.",https://github.com/pytorch/pytorch/issues/68337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2021-11-16T18:04:30Z +issue,47351,no member named 'beta1' in 'torch::optim::AdamOptions',https://github.com/pytorch/pytorch/issues/47351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2021-11-16T16:03:14Z +issue,67535,[LTC] `UncachedCompile` numbers increases for several iterations until fixed,https://github.com/pytorch/pytorch/issues/67535,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-11-16T04:54:54Z +issue,68315,[docs] Strange signatures for torch.autocast,https://github.com/pytorch/pytorch/issues/68315,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: docs",scan when time allows,2021-11-15T22:14:35Z +issue,68288,List of issues to unblock dynamic shapes in Lazy Tensor,https://github.com/pytorch/pytorch/issues/68288,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-15T22:09:59Z +issue,68105,Some type combinations of cublas gemm are not supported when they should,https://github.com/pytorch/pytorch/issues/68105,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: type promotion",scan when time allows,2021-11-15T17:14:44Z +issue,53879,Linear algebra GPU library function bug tracking issue [magma/cusolver/cublas],https://github.com/pytorch/pytorch/issues/53879,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra",scan when time allows,2021-11-15T16:16:25Z +issue,67968,"unresolved external symbol ""__declspec(dllimport) struct _object * __cdecl THPVariable_Wrap",https://github.com/pytorch/pytorch/issues/67968,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-11-15T14:21:06Z +issue,63479,Cannot include extension.h under Windows - linker error THPVariable_Wrap,https://github.com/pytorch/pytorch/issues/63479,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cpp-extensions",scan when time allows,2021-11-15T00:12:58Z +issue,49394,Apparent Memory Leak with torch.as_tensor,https://github.com/pytorch/pytorch/issues/49394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: numpy, module: tensor creation",scan when time allows,2021-11-14T23:55:38Z +issue,32358,torch.bartlett_window not jitable,https://github.com/pytorch/pytorch/issues/32358,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:51:49Z +issue,32365,[jit] Calling `torch.jit.script` on `@staticmethod`s which in turn call other `@staticmethod`s results in a `TypeError`,https://github.com/pytorch/pytorch/issues/32365,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:51:45Z +issue,32039,Internal assert failure for user IValue unwrapping,https://github.com/pytorch/pytorch/issues/32039,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:51:43Z +issue,19628,[jit] nn.LSTM errors in nn.ScriptModule,https://github.com/pytorch/pytorch/issues/19628,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:51:40Z +issue,25591,IValue pickle does not work properly if an empty tensor table is not provided,https://github.com/pytorch/pytorch/issues/25591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:51:38Z +issue,21275,weight_norm is not supported in TorchScript,https://github.com/pytorch/pytorch/issues/21275,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:51:36Z +issue,20591,@ignore annotation for user defined type,https://github.com/pytorch/pytorch/issues/20591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:51:34Z +issue,35074,"[JIT] If a python function type comment is referring to a wrong type, JIT frontend gives a not helpful error message",https://github.com/pytorch/pytorch/issues/35074,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:50:55Z +issue,32300,[jit] Various problems calling `@staticmethod`s in 1.4.0,https://github.com/pytorch/pytorch/issues/32300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:50:53Z +issue,35520,TorchScript can't use lists in conditionals,https://github.com/pytorch/pytorch/issues/35520,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-14T21:50:52Z +issue,56110,caffe2/utils/signal_handler.cc is failing to symbolize libtorch_cpu.so,https://github.com/pytorch/pytorch/issues/56110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-11-14T21:50:47Z +issue,56397,Symbolicate crashes from releases,https://github.com/pytorch/pytorch/issues/56397,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:45Z +issue,55952,Compiler warnings tracking,https://github.com/pytorch/pytorch/issues/55952,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2021-11-14T21:50:43Z +issue,57404,[lint] Make sure that all CMake options have a corresponding output in the summary,https://github.com/pytorch/pytorch/issues/57404,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:38Z +issue,57987,`type: ignore` everything,https://github.com/pytorch/pytorch/issues/57987,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:36Z +issue,58930,[local lint] Shellcheck quicklint file filter is incorrect,https://github.com/pytorch/pytorch/issues/58930,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:35Z +issue,59682,Add a lint rule to recommend TORCH_CHECK(false) over throw std::runtime_error,https://github.com/pytorch/pytorch/issues/59682,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:32Z +issue,60924,[rfc][local lint] Create lint runner as a setup.py target,https://github.com/pytorch/pytorch/issues/60924,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:28Z +issue,60999,Merge seemethere/upload-artifact-s3 and driazati/upload-artifact-s3,https://github.com/pytorch/pytorch/issues/60999,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:26Z +issue,61079,"[clang-tidy] Add a custom rule for AT_ERROR -> TORCH_CHECK(false, ...)",https://github.com/pytorch/pytorch/issues/61079,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:24Z +issue,61384,Merge fork of upload-artifact-s3 to a single commit,https://github.com/pytorch/pytorch/issues/61384,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:20Z +issue,62673,Make printing to stdout / stderr in tests in CI an error,https://github.com/pytorch/pytorch/issues/62673,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-14T21:50:18Z +issue,36922,Multiprocess DataLoader with DLPack conversion sometimes corrupts memory,https://github.com/pytorch/pytorch/issues/36922,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: serialization, module: xnnpack",scan when time allows,2021-11-13T17:24:18Z +issue,55752,"[question] Influence of divisibility of B, C, T by 16 on Conv1d (and Conv2d perf) with CuDNN, including presence of padding",https://github.com/pytorch/pytorch/issues/55752,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: performance",scan when time allows,2021-11-13T04:33:29Z +issue,68298,[Lazy] Add torch.autograd.Function wrappers for the following ops in order to support dynamic shapes,https://github.com/pytorch/pytorch/issues/68298,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-12T23:27:59Z +issue,68287,[Lazy] Add torch.autograd.Function wrappers for the following ops in order to support dynamic shapes,https://github.com/pytorch/pytorch/issues/68287,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-12T22:16:43Z +issue,68204,Feature Request: Scriptable Python Source-of-Truth for operator shape functions,https://github.com/pytorch/pytorch/issues/68204,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-12T22:03:28Z +issue,68193,clang-tidy shows false positive failure on PRs that rename files,https://github.com/pytorch/pytorch/issues/68193,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: lint",scan when time allows,2021-11-11T22:24:54Z +issue,68165,Upgrade NCCL2.11.4 in the new PyTorch,https://github.com/pytorch/pytorch/issues/68165,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-11-11T15:51:10Z +issue,68163,wsl2 install failed from source code,https://github.com/pytorch/pytorch/issues/68163,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkl",scan when time allows,2021-11-11T15:50:35Z +issue,65950,"Storage `is` should work if the storages ""are the same""",https://github.com/pytorch/pytorch/issues/65950,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-11-10T16:24:56Z +issue,67333,Feature request: Complex Number Support for Special Functions,https://github.com/pytorch/pytorch/issues/67333,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: special",scan when time allows,2021-11-10T12:24:17Z +issue,67594,Feature request: set_grad_enabled(*mods_or_params) as a safe context manager,https://github.com/pytorch/pytorch/issues/67594,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2021-11-10T11:44:49Z +issue,67482,Improve docs coverage testing,https://github.com/pytorch/pytorch/issues/67482,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-11-10T09:52:02Z +issue,63004,ignore_index does not work as described with nn.CrossEntropyLoss,https://github.com/pytorch/pytorch/issues/63004,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-11-10T05:47:34Z +issue,65657,Transpose channels when exporting to ONNX,https://github.com/pytorch/pytorch/issues/65657,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: onnx,scan when time allows,2021-11-09T23:00:25Z +issue,13300,Use a dill-based multiprocessing library and serialization,https://github.com/pytorch/pytorch/issues/13300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-11-09T17:05:25Z +issue,68046,"register_ Hook causes the CUDA out of memory, and remove() is useless",https://github.com/pytorch/pytorch/issues/68046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-11-09T16:50:58Z +issue,64897,Either support torch.mean on BoolTensors or fix the error message,https://github.com/pytorch/pytorch/issues/64897,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: error checking, module: reductions",scan when time allows,2021-11-09T10:47:48Z +issue,51535,Why isn't there a pip-installable PyTorch for Raspberry Pi?,https://github.com/pytorch/pytorch/issues/51535,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries",scan when time allows,2021-11-09T08:55:11Z +issue,47433,unbalanced gpu memory when using DistributedDataParallel,https://github.com/pytorch/pytorch/issues/47433,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-11-09T05:27:15Z +issue,62034,cudnn_batch_norm_backward is extremely imprecise for some input shapes,https://github.com/pytorch/pytorch/issues/62034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2021-11-09T05:14:27Z +issue,67989,the abnormal display of bottleneck of resnet in tensorboard,https://github.com/pytorch/pytorch/issues/67989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2021-11-08T21:23:15Z +issue,45823,"Traceback for ""Warning: Mixed memory format inputs detected while calling the operator.""",https://github.com/pytorch/pytorch/issues/45823,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-11-08T02:59:09Z +issue,66775,Feature request: FFT operations on Metal,https://github.com/pytorch/pytorch/issues/66775,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: fft, module: macos, oncall: mobile",scan when time allows,2021-11-07T13:06:58Z +issue,67879,AMP does not improve performance,https://github.com/pytorch/pytorch/issues/67879,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2021-11-05T20:14:49Z +issue,65702,Indicate support for more general use of `pos_weight` in binary cross entropy,https://github.com/pytorch/pytorch/issues/65702,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-11-05T17:35:57Z +issue,56838,Comprehensively test NCCL's `get_future()` API,https://github.com/pytorch/pytorch/issues/56838,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-11-05T17:33:17Z +issue,66874,Error using _stateless version of Module,https://github.com/pytorch/pytorch/issues/66874,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-11-05T17:32:13Z +issue,31994,Trying to disable cuda to run torch on OSX run it on CPU,https://github.com/pytorch/pytorch/issues/31994,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-11-05T12:56:36Z +issue,61636,Enhance Distributed `get_future()` docs,https://github.com/pytorch/pytorch/issues/61636,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-11-04T19:19:32Z +issue,67758,Building libtorch from source requires too much RAM,https://github.com/pytorch/pytorch/issues/67758,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-11-04T13:44:30Z +issue,67772,Functions are rendered incorrectly,https://github.com/pytorch/pytorch/issues/67772,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2021-11-03T16:21:57Z +issue,45042,"torch.distributed launch.py is hanged. (pid, sts) = os.waitpid(self.pid, wait_flags)",https://github.com/pytorch/pytorch/issues/45042,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-11-03T14:26:55Z +issue,67008,kwonly arguments without defaults don't work with test_overrides.py,https://github.com/pytorch/pytorch/issues/67008,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-11-02T12:26:22Z +issue,67589,Expose the `OptState` in `torch.cuda.amp.GradScaler`,https://github.com/pytorch/pytorch/issues/67589,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2021-11-02T11:45:09Z +issue,67642,Is there a pre-built pytorch package for CUDA 10.1?,https://github.com/pytorch/pytorch/issues/67642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-11-02T11:34:22Z +issue,32822,Error tracing custom autograd.Function,https://github.com/pytorch/pytorch/issues/32822,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-11-01T20:49:24Z +issue,67481,Batchnorm2D does behaves differently with different shapes when batch_size=1,https://github.com/pytorch/pytorch/issues/67481,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2021-11-01T17:50:24Z +issue,67540,LTC asynchronous executor is not synced with python code and ignores context manager.,https://github.com/pytorch/pytorch/issues/67540,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-10-31T14:05:13Z +issue,67493,XLA not being tested in TestAutogradDeviceType,https://github.com/pytorch/pytorch/issues/67493,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: tests, module: xla",scan when time allows,2021-10-29T23:00:54Z +issue,66284,Memory leak issue with pytorch_java_only 1.9.0 and libtorch 1.9.0+cpu,https://github.com/pytorch/pytorch/issues/66284,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, module: memory usage, oncall: java, oncall: mobile",scan when time allows,2021-10-29T15:58:21Z +issue,54630,torch-tensor-repr gdb macro doesn't seem to work in opt mode,https://github.com/pytorch/pytorch/issues/54630,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-29T07:32:36Z +issue,67312,ProcessGroupNCCLTest.testSequenceNumInit is failing on 4xlarge,https://github.com/pytorch/pytorch/issues/67312,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2021-10-28T18:17:34Z +issue,67439,Exception in thread when using dataloader,https://github.com/pytorch/pytorch/issues/67439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multithreading",scan when time allows,2021-10-28T18:04:25Z +issue,61030,[Docker] Incompatible torchvision for 1.9.0-cuda10.2-cudnn7-runtime tag.,https://github.com/pytorch/pytorch/issues/61030,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, module: vision",scan when time allows,2021-10-28T05:34:36Z +issue,67377,[LTC] Replace mv with addmv in ts_native_functions.yaml,https://github.com/pytorch/pytorch/issues/67377,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-27T22:43:58Z +issue,67390,Missing doc for torch.distributions functions,https://github.com/pytorch/pytorch/issues/67390,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs",scan when time allows,2021-10-27T22:43:35Z +issue,67388,Missing doc for torch.cuda functions,https://github.com/pytorch/pytorch/issues/67388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2021-10-27T22:39:50Z +issue,67387,Missing doc for torch.autograd functions,https://github.com/pytorch/pytorch/issues/67387,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2021-10-27T22:36:36Z +issue,63802,[torch.distributed.launch|run] Hangs on SIGINT when using a TCPStore backed rdzv_backend,https://github.com/pytorch/pytorch/issues/63802,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2021-10-27T21:17:06Z +issue,67279,register_dispatch_key should provide a way to tell if a native function has a no-op meta kernel,https://github.com/pytorch/pytorch/issues/67279,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: lazy",scan when time allows,2021-10-27T13:37:39Z +issue,67284,Failed to build with master.,https://github.com/pytorch/pytorch/issues/67284,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkl",scan when time allows,2021-10-27T13:30:57Z +issue,67208,TestForeachCUDA.test_binary_op_tensorlists_fastpath__foreach_add_cuda_bool and TestForeachCUDA.test_pointwise_op_fastpath__foreach_addcmul_cuda_uint8 fail intermittently,https://github.com/pytorch/pytorch/issues/67208,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: mta, module: tests",scan when time allows,2021-10-27T13:29:55Z +issue,63846,[torch.distributed] Make dynamic_rendezvous log handler configurable,https://github.com/pytorch/pytorch/issues/63846,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: r2p",scan when time allows,2021-10-27T04:46:32Z +issue,66755,Domain Transformation APIs for LibTorch and LibTorch-Lite,https://github.com/pytorch/pytorch/issues/66755,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: data, module: vision",scan when time allows,2021-10-26T19:44:07Z +issue,67071,Forward results vary depending on batch size on A100 machine,https://github.com/pytorch/pytorch/issues/67071,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: numerical-reproducibility",scan when time allows,2021-10-26T16:22:01Z +issue,67153,Build not working with cuda 11.5,https://github.com/pytorch/pytorch/issues/67153,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-10-26T15:44:26Z +issue,67168,Multiply two named tensor causes RuntimeError,https://github.com/pytorch/pytorch/issues/67168,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2021-10-26T15:35:03Z +issue,58172,Mobile Android: Could not run 'aten::quantize_per_tensor' with arguments from the 'Vulkan' backend,https://github.com/pytorch/pytorch/issues/58172,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2021-10-25T22:29:42Z +issue,58548,Pixel shuffle support for PyTorch Mobile NN-API,https://github.com/pytorch/pytorch/issues/58548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2021-10-25T22:28:51Z +issue,63223,[bug] torch.distributed.elastic logging: Failed to print the first info statement,https://github.com/pytorch/pytorch/issues/63223,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: elastic, oncall: r2p",scan when time allows,2021-10-25T17:57:27Z +issue,64636,[docs or feature request] torchelastic: OOM recovery / skipping batches (e.g. if inf loss or nan gradients),https://github.com/pytorch/pytorch/issues/64636,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: docs, oncall: r2p",scan when time allows,2021-10-25T17:55:40Z +issue,66992,Bug? :Run torch.unique twice get different running time?,https://github.com/pytorch/pytorch/issues/66992,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: benchmark, module: performance, module: sorting and selection",scan when time allows,2021-10-25T16:12:27Z +issue,67187,Expose `isDifferentiableType` to python,https://github.com/pytorch/pytorch/issues/67187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-10-25T15:24:46Z +issue,66963,Significantly difference in execution time when convolution is run as nn.Conv2d and as nn.Sequential,https://github.com/pytorch/pytorch/issues/66963,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: performance",scan when time allows,2021-10-21T14:04:22Z +issue,32695,torch.tensordot has inconsistent signature with torch script,https://github.com/pytorch/pytorch/issues/32695,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-10-21T12:11:42Z +issue,66902,Modify Dr. CI so it could detect runner disconnection failures,https://github.com/pytorch/pytorch/issues/66902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2021-10-20T18:09:18Z +issue,66921,readme not update,https://github.com/pytorch/pytorch/issues/66921,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-10-20T15:21:15Z +issue,66907,torch.chunk return type may not be documented correctly,https://github.com/pytorch/pytorch/issues/66907,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-10-20T15:16:08Z +issue,66936,Lifetime issues when recording external CUDA streams with the caching allocator,https://github.com/pytorch/pytorch/issues/66936,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-10-20T14:53:19Z +issue,66868,torch.triu behaves differently when diagonal out of range,https://github.com/pytorch/pytorch/issues/66868,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2021-10-19T16:55:22Z +issue,66875,"Make the evaluated value of function f(x) accessible from `torch.autograd.functional.jacobian(f,x)`",https://github.com/pytorch/pytorch/issues/66875,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-10-19T16:53:40Z +issue,66824,Migrate C++ tests to Python runner,https://github.com/pytorch/pytorch/issues/66824,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-10-19T16:45:02Z +issue,66751,torch.nn.functional.embedding behave differently in two cases of cpu and cuda,https://github.com/pytorch/pytorch/issues/66751,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: error checking",scan when time allows,2021-10-19T14:40:05Z +issue,65526,[docs] Better explain signature of toch.randn,https://github.com/pytorch/pytorch/issues/65526,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-10-18T17:44:39Z +issue,66640,Static Build of Libtorch not linking correctly,https://github.com/pytorch/pytorch/issues/66640,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-10-18T15:43:44Z +issue,66750,torch.nn.EmbeddingBag behave differently in two cases of cpu and cuda,https://github.com/pytorch/pytorch/issues/66750,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: error checking",scan when time allows,2021-10-18T15:42:53Z +issue,20056,Creation of too big multidimensional array returns empty tensor.,https://github.com/pytorch/pytorch/issues/20056,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: tensor creation",scan when time allows,2021-10-17T22:18:41Z +issue,44511,"LR scheduler throws warning when using scaler.step instead of optimizer.step, and when saving optimizer state",https://github.com/pytorch/pytorch/issues/44511,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2021-10-17T13:06:54Z +issue,53462,Second order derivatives for F.grid_sample(),https://github.com/pytorch/pytorch/issues/53462,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-16T08:30:13Z +issue,56707,[TorchScript JIT] `dtype` propagation pass (meta-issue),https://github.com/pytorch/pytorch/issues/56707,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-14T20:24:06Z +issue,43127,`onnxifi_load` symbol undefined in `libtorch_cpu.a` library built from source,https://github.com/pytorch/pytorch/issues/43127,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-14T19:18:00Z +issue,65837,NVFuser Examples to Help with Integration with Lazy Tensor Core,https://github.com/pytorch/pytorch/issues/65837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-10-14T02:20:08Z +issue,66563,Give a better error message when REGISTER_DISPATCH is used in improper context,https://github.com/pytorch/pytorch/issues/66563,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: internals",scan when time allows,2021-10-13T23:10:10Z +issue,66489,The uniform operator param names in the C++ impl use python keywords,https://github.com/pytorch/pytorch/issues/66489,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-10-13T22:41:22Z +issue,66460,Inconsistencies in LSTM outputs when processing sequence stepwise on CPU,https://github.com/pytorch/pytorch/issues/66460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: rnn",scan when time allows,2021-10-13T22:05:44Z +issue,66223,Data Loader tests hang when run in ASAN test job,https://github.com/pytorch/pytorch/issues/66223,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: tests",scan when time allows,2021-10-13T18:31:51Z +issue,66366,Feature Request: Add constant padding_mode for grid_sample,https://github.com/pytorch/pytorch/issues/66366,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn, module: padding",scan when time allows,2021-10-13T16:07:11Z +issue,66403,[opinfo] Confusing interface for `ops` decorator,https://github.com/pytorch/pytorch/issues/66403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-10-13T07:48:38Z +issue,51965,Feature Request:Fast Cosine Transform,https://github.com/pytorch/pytorch/issues/51965,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2021-10-13T07:04:49Z +issue,66476,"0INTERNAL ASSERT FAILED, We don't have an op for aten::eq but it isn't a special case",https://github.com/pytorch/pytorch/issues/66476,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-10-12T16:35:17Z +issue,66473,nccl comm init needs a global barrier.,https://github.com/pytorch/pytorch/issues/66473,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nccl",scan when time allows,2021-10-12T16:33:43Z +issue,65301,Anomaly detection: Error detected in CudnnRnnBackward0,https://github.com/pytorch/pytorch/issues/65301,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: rnn",scan when time allows,2021-10-12T15:52:23Z +issue,66341,Invalid handling of out args with type Tensor[] in native_functions.yaml,https://github.com/pytorch/pytorch/issues/66341,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: internals",scan when time allows,2021-10-12T13:54:58Z +issue,55340,[RFC] Plan to deduplicate test_c10d and distributed_test,https://github.com/pytorch/pytorch/issues/55340,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-10-11T23:53:48Z +issue,66359,question on installing GPU-enabled Pytorch,https://github.com/pytorch/pytorch/issues/66359,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: lts",scan when time allows,2021-10-11T21:14:38Z +issue,66196,Specifying USE_VULKAN=0 in launching build_android.sh does not disable Vulkan,https://github.com/pytorch/pytorch/issues/66196,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vulkan",scan when time allows,2021-10-11T20:13:18Z +issue,66305,Modify convolution kernels for ops,https://github.com/pytorch/pytorch/issues/66305,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2021-10-11T19:04:50Z +issue,66208,Add LTC-TS support for ops used in SGD optimizer,https://github.com/pytorch/pytorch/issues/66208,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-10-11T14:36:59Z +issue,66224,[feature request][quant] Support FakeQuant qconfigs in `test_module_init`,https://github.com/pytorch/pytorch/issues/66224,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-10-11T14:36:53Z +issue,66249,Latency issue with torch.sin,https://github.com/pytorch/pytorch/issues/66249,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-10-11T14:23:19Z +issue,66364,Hope to obtain authorization for Chinese translation of official tutorials.,https://github.com/pytorch/pytorch/issues/66364,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-10-11T14:18:18Z +issue,66251,Should the drop_last parameter of Dataloader be mutually exclusive with the batch samplers?,https://github.com/pytorch/pytorch/issues/66251,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-10-11T14:05:18Z +issue,65949,torch.unique signature is not descriptive,https://github.com/pytorch/pytorch/issues/65949,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-10-11T01:49:41Z +issue,65038,"[pytorch v1.9.0] for torch sync.pipeline in a single GPU, passing tensor layer by layer is better or using the skip-connection API is better?",https://github.com/pytorch/pytorch/issues/65038,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-10-09T06:46:57Z +issue,38471,Immediate mode API (with functional flavor) for optimizers,https://github.com/pytorch/pytorch/issues/38471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-10-09T03:25:18Z +issue,48287,"[doc] Error ""You've reached a dead end"" when opening torch.__config__ in the docs",https://github.com/pytorch/pytorch/issues/48287,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2021-10-08T22:57:58Z +issue,64766,Ability to explicitly close/dispose a DataLoader,https://github.com/pytorch/pytorch/issues/64766,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader, module: memory usage",scan when time allows,2021-10-08T02:50:06Z +issue,58139,[docs] Strange torch.unique function signature,https://github.com/pytorch/pytorch/issues/58139,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: docs",scan when time allows,2021-10-07T17:07:48Z +issue,65984,libtorch: collate_fn equivalent,https://github.com/pytorch/pytorch/issues/65984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: dataloader",scan when time allows,2021-10-07T13:57:30Z +issue,64874,I am unable to detect torch.Tensor._version change during __torch_dispatch__ call,https://github.com/pytorch/pytorch/issues/64874,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2021-10-06T15:53:18Z +issue,66199,Move the `queue_callback()` API out of `Variable._execution_engine` and into a public API,https://github.com/pytorch/pytorch/issues/66199,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-10-06T14:18:56Z +issue,45769,F.conv2d() causes RuntimeError: cuDNN error: CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/45769,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2021-10-06T14:14:56Z +issue,65811,"Beta distribution yields wrong results for a,b >1 and x=0 or x=1",https://github.com/pytorch/pytorch/issues/65811,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-10-06T12:15:04Z +issue,65910,Deprecation: Remove nn.functional.sigmoid,https://github.com/pytorch/pytorch/issues/65910,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: nn",scan when time allows,2021-10-06T04:38:15Z +issue,66026,`__torch_dispatch__` can result in returning `None` for an op that should return Tensors.,https://github.com/pytorch/pytorch/issues/66026,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: pybind",scan when time allows,2021-10-06T03:52:40Z +issue,64186,Complex recurrent layers produce NaN as grad,https://github.com/pytorch/pytorch/issues/64186,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs, module: nn",scan when time allows,2021-10-05T20:39:26Z +issue,65306,[numpy] Add `iscomplexobj` and `isrealobj`,https://github.com/pytorch/pytorch/issues/65306,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy",scan when time allows,2021-10-05T19:05:23Z +issue,66116,Removal of BufferedShuffleDataset,https://github.com/pytorch/pytorch/issues/66116,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: docs",scan when time allows,2021-10-05T13:05:50Z +issue,65739,Sobol state API,https://github.com/pytorch/pytorch/issues/65739,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: random",scan when time allows,2021-10-05T12:40:28Z +issue,65753,[POC] Convolution-relu fusion via torch function and one-step lazy evaluation,https://github.com/pytorch/pytorch/issues/65753,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: lazy",scan when time allows,2021-10-04T23:10:13Z +issue,65229,at::parallel_for created max_threads for inputs larger than GRAIN_SIZE,https://github.com/pytorch/pytorch/issues/65229,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: openmp, module: performance",scan when time allows,2021-10-04T17:35:35Z +issue,65426,Add a `__torch_pre_dispatch__` that will live at the top of the dispatcher,https://github.com/pytorch/pytorch/issues/65426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-10-04T17:33:36Z +issue,65694,Plain Tensor serialization does not save the content of `__dict__`.,https://github.com/pytorch/pytorch/issues/65694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-10-04T17:24:59Z +issue,65494,Feature request: add bool dtype support to CPU abs(),https://github.com/pytorch/pytorch/issues/65494,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-04T17:23:37Z +issue,65665,Pipe method for torch tensors,https://github.com/pytorch/pytorch/issues/65665,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-10-04T17:21:42Z +issue,65688,Parametrization silently disables RNN weight updates,https://github.com/pytorch/pytorch/issues/65688,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2021-10-04T17:19:41Z +issue,66076,Bad Input to torch.nn.NLLLoss causes CUDA Error,https://github.com/pytorch/pytorch/issues/66076,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: loss",scan when time allows,2021-10-04T16:32:45Z +issue,28320,Unable to install pytorch with cuda 10.0 using conda,https://github.com/pytorch/pytorch/issues/28320,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-10-04T16:04:13Z +issue,65067,[DataPipe] Add Test Cases to Ensure Correct Behaviors When IterDataPipes Reset,https://github.com/pytorch/pytorch/issues/65067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2021-10-04T14:54:36Z +issue,65339,Make it easier to accurately reflect storage/view relationships in meta/wrapper tensors,https://github.com/pytorch/pytorch/issues/65339,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: meta tensors",scan when time allows,2021-10-04T14:36:57Z +issue,65661,[BUG] ConvTranspose with out_channels=1,https://github.com/pytorch/pytorch/issues/65661,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2021-10-03T01:57:58Z +issue,20005,install error from source,https://github.com/pytorch/pytorch/issues/20005,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-10-02T01:44:04Z +issue,65788,Autocast sometimes discards changes to mutable arguments,https://github.com/pytorch/pytorch/issues/65788,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: intel",scan when time allows,2021-10-01T21:02:11Z +issue,16499,TorchConfig.cmake always sets _GLIBCXX_USE_CXX11_ABI,https://github.com/pytorch/pytorch/issues/16499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-10-01T19:25:20Z +issue,16187,Sparse matrix multiplication is too slow,https://github.com/pytorch/pytorch/issues/16187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-10-01T19:25:20Z +issue,15161,Maybe a bug when using DataParallel,https://github.com/pytorch/pytorch/issues/15161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-10-01T19:25:20Z +issue,13222,Memory inefficient in batched matmul when requiring gradients,https://github.com/pytorch/pytorch/issues/13222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2021-10-01T19:25:20Z +issue,13018,Improved performance for torch.multinomial with small batches,https://github.com/pytorch/pytorch/issues/13018,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2021-10-01T19:25:20Z +issue,10956,Better dev docs for writing native CPU kernels with Vec256,https://github.com/pytorch/pytorch/issues/10956,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vectorization,scan when time allows,2021-10-01T19:25:20Z +issue,9898,[Feature Request] Checkpoint manager,https://github.com/pytorch/pytorch/issues/9898,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-01T19:25:20Z +issue,9433,Build error: cannot find -lonnxifi_loader,https://github.com/pytorch/pytorch/issues/9433,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-10-01T19:25:20Z +issue,7066,LBFGS contribution,https://github.com/pytorch/pytorch/issues/7066,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-10-01T19:25:20Z +issue,5912,Conv-RNN combination slow in backward pass,https://github.com/pytorch/pytorch/issues/5912,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2021-10-01T19:25:20Z +issue,1505,Cannot find Intel MKL,https://github.com/pytorch/pytorch/issues/1505,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: dependency bug, module: mkl",scan when time allows,2021-10-01T19:25:20Z +issue,1462,"Change sparse_mask to take indexing mask, rather than entire sparse tensor",https://github.com/pytorch/pytorch/issues/1462,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-10-01T19:25:20Z +issue,1442,Optimizers can't be moved to a different GPU,https://github.com/pytorch/pytorch/issues/1442,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-10-01T19:25:20Z +issue,65760,Codegen issues with Tensor(a!)? arguments in native_functions.yaml schemas,https://github.com/pytorch/pytorch/issues/65760,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-10-01T18:12:22Z +issue,65622,Transformer modules should not hard-code the activation function,https://github.com/pytorch/pytorch/issues/65622,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-10-01T18:04:47Z +issue,65881,torch.hub.load can confuse external python package with local python package.,https://github.com/pytorch/pytorch/issues/65881,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2021-10-01T07:56:39Z +issue,65890,`pad_packed_sequence` is not exactly the inverse of `pack_padded_sequence`,https://github.com/pytorch/pytorch/issues/65890,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2021-10-01T01:51:36Z +issue,65868,torch.mul is not consistent with torch.multiply,https://github.com/pytorch/pytorch/issues/65868,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: numpy",scan when time allows,2021-10-01T01:46:24Z +issue,65825,Incorrect reference to previous kernel in the warning displayed when overriding a previously registered kernel for the same operator and the same dispatch,https://github.com/pytorch/pytorch/issues/65825,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-09-30T19:05:10Z +issue,65619,Integration of __torch_dispatch__ with masked tensor semantics,https://github.com/pytorch/pytorch/issues/65619,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2021-09-30T17:54:57Z +issue,42053,Function request: np.copy (alias of clone?),https://github.com/pytorch/pytorch/issues/42053,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-09-30T08:27:57Z +issue,65522,"pytorch framework tests using make_tensor hangs with pytest's boxed exec option ""--forked""",https://github.com/pytorch/pytorch/issues/65522,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-09-30T05:45:34Z +issue,62648,Feature request: Generate Sobol points directly on the GPU,https://github.com/pytorch/pytorch/issues/62648,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: distributions, module: tensor creation",scan when time allows,2021-09-30T05:39:04Z +issue,57593,[RFC] Provide an API for Structural Performance Tips in DDP,https://github.com/pytorch/pytorch/issues/57593,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-09-28T20:02:12Z +issue,65428,module 'torch.distributed' has no attribute 'optim',https://github.com/pytorch/pytorch/issues/65428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-09-28T19:31:03Z +issue,65752,Deprecate torch.cross with optional dim,https://github.com/pytorch/pytorch/issues/65752,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deprecation,scan when time allows,2021-09-28T19:04:34Z +issue,55191,"`torch.tensor(..., device='meta')` doesn't work",https://github.com/pytorch/pytorch/issues/55191,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2021-09-28T17:03:48Z +issue,61122,Make copy_ use dispatcher,https://github.com/pytorch/pytorch/issues/61122,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2021-09-28T10:13:47Z +issue,65683,`torch.nn.functional.l1_loss` fails gradgradcheck for complex inputs,https://github.com/pytorch/pytorch/issues/65683,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex, module: nn",scan when time allows,2021-09-27T23:31:39Z +issue,65650,Does ZeroRedundancyOptimizer support re-split partition on load_state_dict?,https://github.com/pytorch/pytorch/issues/65650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-09-27T22:25:39Z +issue,65637,Test ZeRO on gloo backend for GPU,https://github.com/pytorch/pytorch/issues/65637,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-09-27T21:55:33Z +issue,65670,Memory Leak in Distributed RPC nightly,https://github.com/pytorch/pytorch/issues/65670,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-09-27T20:25:59Z +issue,65646,[LTC] Building unit tests on Debian results in double defined _GLIBCXX_USE_CXX11_ABI flag,https://github.com/pytorch/pytorch/issues/65646,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-09-27T20:21:19Z +issue,65687,"repeat_interleave changes memory format from ""channels last"" to ""contiguous""",https://github.com/pytorch/pytorch/issues/65687,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-09-27T18:43:22Z +issue,65673,[libtorch]incorrect sigmoid result on arm chip(rk3326),https://github.com/pytorch/pytorch/issues/65673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: arm,scan when time allows,2021-09-27T18:32:31Z +issue,61835,Broadcasting documentation diverges from implementation,https://github.com/pytorch/pytorch/issues/61835,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: docs",scan when time allows,2021-09-27T17:46:51Z +issue,65664,C10dProcessGroupSerialization.test_process_group_as_module_member is flaky,https://github.com/pytorch/pytorch/issues/65664,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-09-27T04:54:19Z +issue,61727,```torch.distributions.Categorical``` unintended ```log_prob``` gradient w.r.t ```probs```,https://github.com/pytorch/pytorch/issues/61727,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-09-26T09:12:50Z +issue,25066,[FR][jit] torch.jit.script as a class decorator,https://github.com/pytorch/pytorch/issues/25066,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-09-25T20:10:44Z +issue,23509,Enable PyTorch Bfloat16 for CPU and add MKL-DNN bfloat16 optimization for Cooper Lake,https://github.com/pytorch/pytorch/issues/23509,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2021-09-25T01:10:57Z +issue,65628,Add torch.distributed.shard package,https://github.com/pytorch/pytorch/issues/65628,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-09-24T19:57:00Z +issue,65604,pytorch serializes entire tensor when you try to pickle a slice,https://github.com/pytorch/pytorch/issues/65604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-09-24T16:46:16Z +issue,65572,PyTorch setup.py should confirm with user if the user changes build flags,https://github.com/pytorch/pytorch/issues/65572,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-09-24T15:59:01Z +issue,65577,Support cat() for meta tensors,https://github.com/pytorch/pytorch/issues/65577,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: meta tensors,scan when time allows,2021-09-24T15:40:21Z +issue,65465,[LTC] Fail to run testcase of latest lazy_tensor_core branch,https://github.com/pytorch/pytorch/issues/65465,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-09-24T04:45:55Z +issue,65462,"Why here save `store`, rather than `prefix_store`?",https://github.com/pytorch/pytorch/issues/65462,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-09-24T04:19:11Z +issue,65261,Add an `all_paths` method to PackageExporter,https://github.com/pytorch/pytorch/issues/65261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,scan when time allows,2021-09-24T04:07:50Z +issue,65464,linking error when building on Linux,https://github.com/pytorch/pytorch/issues/65464,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-09-24T03:42:16Z +issue,65536,Pass backward flags such as retain_graph to context of custom torch.autograd.Function,https://github.com/pytorch/pytorch/issues/65536,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-09-23T21:20:21Z +issue,65520,Improve input checking for running_var of nn.functional.batch_norm on CPU/GPU,https://github.com/pytorch/pytorch/issues/65520,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-09-23T20:48:42Z +issue,65512,Performance problems of eigh operator on CPU,https://github.com/pytorch/pytorch/issues/65512,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-09-23T20:32:49Z +issue,65386,[LTC] Setup a subset of TorchBench for the current implementation,https://github.com/pytorch/pytorch/issues/65386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-09-23T20:27:16Z +issue,65569,Support 'meta' device for 'shard_parameter' API,https://github.com/pytorch/pytorch/issues/65569,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-09-23T19:56:55Z +issue,18317,Add a RandomBatchSampler ?,https://github.com/pytorch/pytorch/issues/18317,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-09-23T18:43:52Z +issue,65475,building pytorch from source without conda,https://github.com/pytorch/pytorch/issues/65475,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-09-23T17:32:39Z +issue,36175,New dtype ComplexPolarFloat (phasor),https://github.com/pytorch/pytorch/issues/36175,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2021-09-22T16:47:31Z +issue,65329,How can I re-weight a sample based on both class weights and instance weights?,https://github.com/pytorch/pytorch/issues/65329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2021-09-22T16:31:49Z +issue,65446,libtorch compile problem. How to get the correct protobuf version? what PROTOBUF_VERSION <3011000 and 3011004 ,https://github.com/pytorch/pytorch/issues/63645,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-09-17T17:06:16Z +issue,64690,_disabled_torch_function_impl is unimplementable (must be special cased),https://github.com/pytorch/pytorch/issues/64690,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2021-09-17T16:33:03Z +issue,65049,Setting a random seed or passing a generator to Modules for deterministic sampling.,https://github.com/pytorch/pytorch/issues/65049,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: nn, module: random",scan when time allows,2021-09-17T15:43:05Z +issue,65100,[Bug] [CUDA IPC] CUDA IPC memory cost,https://github.com/pytorch/pytorch/issues/65100,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing, module: performance",scan when time allows,2021-09-17T12:17:15Z +issue,65132,How to reference a tensor variable from a superclass of `torch.Tensor`?,https://github.com/pytorch/pytorch/issues/65132,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xla,scan when time allows,2021-09-16T21:35:37Z +issue,65131,Adding a device variable to a `torch.Tensor` subclass fails,https://github.com/pytorch/pytorch/issues/65131,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xla,scan when time allows,2021-09-16T21:31:39Z +issue,65166,Review the Tensor and Operator Basics Onboarding page,https://github.com/pytorch/pytorch/issues/65166,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-09-16T20:52:59Z +issue,39671,TestTorchDeviceTypeCPU.test_float_to_int_conversion_finite_cpu_uint8 is broken on PowerPC,https://github.com/pytorch/pytorch/issues/39671,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: power, module: tests",scan when time allows,2021-09-16T09:34:25Z +issue,18434,improve jit error message for legacy constructor,https://github.com/pytorch/pytorch/issues/18434,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-09-16T01:04:32Z +issue,50962,Add option for Convolutions to operate over NTC and NHWC tensors,https://github.com/pytorch/pytorch/issues/50962,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-09-15T22:21:28Z +issue,65069,torch.linalg.cholesky does not raise a RuntimeError when the matrix is not symmetric,https://github.com/pytorch/pytorch/issues/65069,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: linear algebra",scan when time allows,2021-09-15T21:43:08Z +issue,65050,torch.cross precision problem,https://github.com/pytorch/pytorch/issues/65050,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2021-09-15T18:14:48Z +issue,64743,Named Tensors to support custom unify functions,https://github.com/pytorch/pytorch/issues/64743,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2021-09-15T12:24:43Z +issue,65002,Traceback tensor mode,https://github.com/pytorch/pytorch/issues/65002,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-09-15T00:01:19Z +issue,16589,Make sure `data_ptr` for non-zero-size input tensors stays the same after the VariableType dispatch,https://github.com/pytorch/pytorch/issues/16589,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: autograd, module: molly-guard",scan when time allows,2021-09-14T23:05:01Z +issue,55192,"Consider adding meta:1, meta:2 devices",https://github.com/pytorch/pytorch/issues/55192,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: meta tensors, module: structured kernels",scan when time allows,2021-09-14T20:46:06Z +issue,19978,[JIT] traced model with optimization shows no performance improvement,https://github.com/pytorch/pytorch/issues/19978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-09-14T17:18:26Z +issue,18157,JIT does not batch linear layers in an ensemble,https://github.com/pytorch/pytorch/issues/18157,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-09-14T17:17:19Z +issue,17655,TracedModule 'to' attribute doesn't work for tensors created on forward.,https://github.com/pytorch/pytorch/issues/17655,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-09-14T17:17:08Z +issue,40507,SyncBatchNorm for JIT and a list of not supported operations,https://github.com/pytorch/pytorch/issues/40507,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-09-14T17:16:12Z +issue,30754,"JIT, nn.utils.weight_norm and {save,load}_state_dict produce wrong results",https://github.com/pytorch/pytorch/issues/30754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-09-14T17:15:10Z +issue,22155,[jit] Optional type refinement on non-named expressions,https://github.com/pytorch/pytorch/issues/22155,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-09-14T17:13:36Z +issue,64939,BC CI error message should link to some information about how to squash the warning,https://github.com/pytorch/pytorch/issues/64939,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2021-09-13T17:38:17Z +issue,58380,Clarity of error message in einsum regressed in performance improvements,https://github.com/pytorch/pytorch/issues/58380,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: linear algebra",scan when time allows,2021-09-13T16:08:36Z +issue,41099,max_pool2d always compute indices even when it's not required,https://github.com/pytorch/pytorch/issues/41099,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2021-09-13T16:07:14Z +issue,40570,Generalized CPU vector reductions,https://github.com/pytorch/pytorch/issues/40570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: reductions",scan when time allows,2021-09-13T16:07:05Z +issue,64881,Add support for GPU based data loading and transformation,https://github.com/pytorch/pytorch/issues/64881,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-09-13T14:55:41Z +issue,17966,torch.cuda.is_available() returns misleading value,https://github.com/pytorch/pytorch/issues/17966,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: ux",scan when time allows,2021-09-13T14:02:16Z +issue,64812,PyTorch inference on tensors of a particular size cause Illegal Instruction (core dumped) on Jetson Nano,https://github.com/pytorch/pytorch/issues/64812,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: jetson",scan when time allows,2021-09-13T08:16:28Z +issue,23067,creation of a tensor from a numba.cuda array,https://github.com/pytorch/pytorch/issues/23067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numba,scan when time allows,2021-09-13T05:07:58Z +issue,64818,Inconsistent NaN handling by cholesky between CPU and CUDA,https://github.com/pytorch/pytorch/issues/64818,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: nans and infs",scan when time allows,2021-09-11T20:41:37Z +issue,30873,ReduceLROnPlateau detects a plateau during a steady decrease after a spike,https://github.com/pytorch/pytorch/issues/30873,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-09-11T07:45:12Z +issue,64845,Track Reliability of Test Reporting and Uploading,https://github.com/pytorch/pytorch/issues/64845,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-09-11T01:30:57Z +issue,17849,Tensor::options() returns false for requires_grad when it is true,https://github.com/pytorch/pytorch/issues/17849,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2021-09-11T01:25:33Z +issue,17879,C++ nn::Sequential push_back() copies module if the module is concrete type,https://github.com/pytorch/pytorch/issues/17879,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2021-09-11T01:23:35Z +issue,64868,CMake outputs lots of warnings in CI,https://github.com/pytorch/pytorch/issues/64868,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2021-09-11T00:14:47Z +issue,64311,issues flutter in android studio,https://github.com/pytorch/pytorch/issues/64311,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2021-09-10T20:09:10Z +issue,64407,Create a boxed/templated ADInplaceOrView kernel,https://github.com/pytorch/pytorch/issues/64407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-09-09T19:39:01Z +issue,63041,[RFC] Should DDP support custom reduction logic for registered module buffers?,https://github.com/pytorch/pytorch/issues/63041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-09-09T19:02:05Z +issue,64745,Consolidate ProcessGroup allgather_coalesced and allgather,https://github.com/pytorch/pytorch/issues/64745,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: c10d, oncall: distributed",scan when time allows,2021-09-09T14:57:50Z +issue,64637,Evaluating on single GPU (DDP),https://github.com/pytorch/pytorch/issues/64637,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ddp,scan when time allows,2021-09-09T00:42:52Z +issue,64644,Improve meta tensor testing,https://github.com/pytorch/pytorch/issues/64644,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: structured kernels, module: tests",scan when time allows,2021-09-09T00:41:57Z +issue,64709,Add out= resize warning to cat and maybe_native_stack,https://github.com/pytorch/pytorch/issues/64709,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2021-09-08T22:59:05Z +issue,17150,Conv2d layers should accept 4-tuple for padding argument,https://github.com/pytorch/pytorch/issues/17150,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",scan when time allows,2021-09-08T18:50:53Z +issue,64544,[Feature Request] Any plan to add 'Sparse Convolution' as default nn module?,https://github.com/pytorch/pytorch/issues/64544,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: sparse",scan when time allows,2021-09-08T14:41:52Z +issue,61468,Version is set to 1.8.0 in the torch tag v1.8.1,https://github.com/pytorch/pytorch/issues/61468,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: releng",scan when time allows,2021-09-08T02:20:09Z +issue,64539,[LTC][Documentation] Python examples in API_GUIDE.md don't work,https://github.com/pytorch/pytorch/issues/64539,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: lazy",scan when time allows,2021-09-07T22:42:54Z +issue,64527,"Asking for ""git submodule update"" when building a specific version",https://github.com/pytorch/pytorch/issues/64527,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-09-07T20:36:26Z +issue,64548,Interpreter initialized state changed by torch dispatch,https://github.com/pytorch/pytorch/issues/64548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__, module: deploy",scan when time allows,2021-09-07T20:32:43Z +issue,64541,[docs] Version switcher to not reset home page unconditionally,https://github.com/pytorch/pytorch/issues/64541,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2021-09-07T20:28:39Z +issue,64458,Rename C++ Serialization APIs,https://github.com/pytorch/pytorch/issues/64458,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-09-07T03:07:18Z +issue,64292,torch.empty_like not taking None as layout?,https://github.com/pytorch/pytorch/issues/64292,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: tensor creation",scan when time allows,2021-09-06T05:57:59Z +issue,64107,as_tensor and negative strided np arrays,https://github.com/pytorch/pytorch/issues/64107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: tensor creation",scan when time allows,2021-09-06T05:30:22Z +issue,64106,modulefinder_determinator is incompatible with imports into the test/ directory,https://github.com/pytorch/pytorch/issues/64106,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-09-05T03:30:55Z +issue,17612,The documentation to Module._version isn't visible.,https://github.com/pytorch/pytorch/issues/17612,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: serialization",scan when time allows,2021-09-05T03:13:31Z +issue,26456,[libtorch]Same model in CUDA and CPU got different result?,https://github.com/pytorch/pytorch/issues/26456,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2021-09-03T01:58:27Z +issue,64426,Add support for ROCm MIOpen miopenConvolutionForwardBias and miopenConvolutionBackwardBias,https://github.com/pytorch/pytorch/issues/64426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rocm,scan when time allows,2021-09-02T16:57:49Z +issue,17644,"Not depend on third_party submodules, but self-built libraries?",https://github.com/pytorch/pytorch/issues/17644,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-09-02T15:04:49Z +issue,62032,"Create a boxed fallback / template recipe for Autograd that forwards, but errors on backwards",https://github.com/pytorch/pytorch/issues/62032,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: dispatch, module: internals",scan when time allows,2021-09-02T04:30:02Z +issue,14112,[sparse] add descriptions and examples for methods at torch.sparse doc page,https://github.com/pytorch/pytorch/issues/14112,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-09-01T22:33:10Z +issue,64308,Building from source results in no member named 'cerr' in namespace 'std',https://github.com/pytorch/pytorch/issues/64308,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-09-01T20:22:56Z +issue,64384,[doc] incomplete `get_num_interop_threads` doc,https://github.com/pytorch/pytorch/issues/64384,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-09-01T19:01:31Z +issue,64306,GradGrad of max_pool2d fails with empty batch dimension,https://github.com/pytorch/pytorch/issues/64306,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2021-09-01T17:54:28Z +issue,63297,Named tensor in tracer,https://github.com/pytorch/pytorch/issues/63297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2021-09-01T14:16:46Z +issue,51558,torch.reshape fails to keep the memory format,https://github.com/pytorch/pytorch/issues/51558,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: viewing and reshaping",scan when time allows,2021-09-01T12:20:41Z +issue,48483,Vulkan Api Backend Build Error with GCC,https://github.com/pytorch/pytorch/issues/48483,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-09-01T07:22:31Z +issue,63870,torch median / nanmedian w/ nans speed,https://github.com/pytorch/pytorch/issues/63870,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: performance, module: reductions, module: sorting and selection",scan when time allows,2021-09-01T00:06:39Z +issue,62788,PyTorch(with mkldnn) in inference mode using non-optimal prop_kind for convolution,https://github.com/pytorch/pytorch/issues/62788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mkldnn,scan when time allows,2021-08-31T19:44:23Z +issue,62473,Sparse updates to logits in distributions.Categorical,https://github.com/pytorch/pytorch/issues/62473,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: distributions, module: memory usage, module: sparse",scan when time allows,2021-08-31T17:40:30Z +issue,64284,FR: Record results of OpInfo reference tests and detect when numerics of an operator change,https://github.com/pytorch/pytorch/issues/64284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-08-31T17:00:36Z +issue,64206,Document how to generate Pybind bindings for C++ Autograd,https://github.com/pytorch/pytorch/issues/64206,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-08-31T14:20:23Z +issue,38185,Feature Request: Sampled softmax loss,https://github.com/pytorch/pytorch/issues/38185,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2021-08-31T07:19:46Z +issue,38205,RESOLVED: Disable zero-dim CUDA tensors interacting with CUDA tensors on other devices,https://github.com/pytorch/pytorch/issues/38205,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: cuda",scan when time allows,2021-08-31T07:15:20Z +issue,64138,Longer int and normal Floats conversion to FloatTensor is strange,https://github.com/pytorch/pytorch/issues/64138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-31T06:45:25Z +issue,57447,`torch._dirichlet_grad` returns `nan` value on cuda device,https://github.com/pytorch/pytorch/issues/57447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda, module: nans and infs",scan when time allows,2021-08-31T00:47:52Z +issue,51906,torch.nn.functional.binary_cross_entropy(_with_logits) outputs NaN,https://github.com/pytorch/pytorch/issues/51906,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: nn",scan when time allows,2021-08-30T21:14:39Z +issue,51714,Inconsistent result between NumPy and PyTorch for 0^(z) when Re(z) < 0,https://github.com/pytorch/pytorch/issues/51714,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",scan when time allows,2021-08-30T21:13:57Z +issue,48000,Inconsistent complex results with NumPy when computing non-positive power of 0,https://github.com/pytorch/pytorch/issues/48000,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs, module: numpy",scan when time allows,2021-08-30T21:06:40Z +issue,60825,Where OpInfo doesn't handle cases where one of the inputs is a scalar,https://github.com/pytorch/pytorch/issues/60825,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection, module: tests",scan when time allows,2021-08-30T20:34:56Z +issue,64133,nn.MultiheadAttention does not belong to activation function,https://github.com/pytorch/pytorch/issues/64133,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-08-30T17:58:34Z +issue,64038,torch.nn.functional.softmax _stacklevel undocumented,https://github.com/pytorch/pytorch/issues/64038,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-08-30T17:32:31Z +issue,64187,Uniformly test that defaulted Tensor arguments appropriately handle __torch_function__,https://github.com/pytorch/pytorch/issues/64187,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: tests",scan when time allows,2021-08-30T17:29:18Z +issue,63970,[LTC] Investigate a testing utility to make sure correct lowering has happened,https://github.com/pytorch/pytorch/issues/63970,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-08-30T17:00:16Z +issue,63113,Implement half_to_float flag for both _log_softmax and _softmax operators,https://github.com/pytorch/pytorch/issues/63113,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-08-30T17:00:15Z +issue,62614,LazyTensor recomputes intermediate tensor in fragmented computation,https://github.com/pytorch/pytorch/issues/62614,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lazy,scan when time allows,2021-08-30T17:00:13Z +issue,64162,[feature request] torch.Generator constructor to accept seed directly,https://github.com/pytorch/pytorch/issues/64162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2021-08-30T15:58:23Z +issue,63920,quicklint clang-tidy step fails with many unrelated errors,https://github.com/pytorch/pytorch/issues/63920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2021-08-30T15:48:24Z +issue,63919,quicklint mypy step fails with assertion,https://github.com/pytorch/pytorch/issues/63919,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: lint, module: typing",scan when time allows,2021-08-30T15:48:01Z +issue,17554,Add ASSERT for calling accessor on a GPU tensor,https://github.com/pytorch/pytorch/issues/17554,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: molly-guard",scan when time allows,2021-08-30T15:39:31Z +issue,57272,Consistent treatment of non-differentiability in linear algebra operations,https://github.com/pytorch/pytorch/issues/57272,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: linear algebra",scan when time allows,2021-08-30T15:25:42Z +issue,17425,"improved assert message in the case of ""CUDA error: device-side assert triggered""",https://github.com/pytorch/pytorch/issues/17425,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: cuda, module: molly-guard",scan when time allows,2021-08-30T15:23:18Z +issue,63293,Support `torch.linalg.outer`,https://github.com/pytorch/pytorch/issues/63293,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: python array api",scan when time allows,2021-08-30T12:53:29Z +issue,56235,`test_cholesky_solve` gradgradcheck fails sometimes,https://github.com/pytorch/pytorch/issues/56235,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra, module: magma, module: tests",scan when time allows,2021-08-30T08:29:01Z +issue,58015,Support a dist.group_like API,https://github.com/pytorch/pytorch/issues/58015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: c10d, oncall: distributed",scan when time allows,2021-08-30T07:13:01Z +issue,8577,Update tests to no longer spew debug info,https://github.com/pytorch/pytorch/issues/8577,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-08-30T03:11:59Z +issue,11496,"Tests with ""."" in the name cannot be run standalone",https://github.com/pytorch/pytorch/issues/11496,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-08-30T03:09:43Z +issue,52738,[OpInfo] Improvements for sparse ops tests,https://github.com/pytorch/pytorch/issues/52738,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sparse, module: tests",scan when time allows,2021-08-30T03:08:01Z +issue,52984,Tracker: pytest-related test improvements,https://github.com/pytorch/pytorch/issues/52984,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-08-30T03:07:16Z +issue,53352,OpInfo mechanism to test for nondeterminism,https://github.com/pytorch/pytorch/issues/53352,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: tests",scan when time allows,2021-08-30T03:06:57Z +issue,53510,Improve memory format testing,https://github.com/pytorch/pytorch/issues/53510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: tests",scan when time allows,2021-08-30T03:05:03Z +issue,54006,enable parallel test execution for GPU tests,https://github.com/pytorch/pytorch/issues/54006,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",scan when time allows,2021-08-30T03:03:28Z +issue,55159,Better syntax for OpInfo,https://github.com/pytorch/pytorch/issues/55159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-08-30T03:03:08Z +issue,54346,Missing tests for gradcheck,https://github.com/pytorch/pytorch/issues/54346,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: tests",scan when time allows,2021-08-30T03:03:00Z +issue,55907,test_variant_consistency_eager_addbmm fails on both cpu and cuda,https://github.com/pytorch/pytorch/issues/55907,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: tests",scan when time allows,2021-08-30T03:00:23Z +issue,56202,Tolerance for non-determinism operators in gradcheck,https://github.com/pytorch/pytorch/issues/56202,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: determinism, module: tests",scan when time allows,2021-08-30T02:59:46Z +issue,57301,Avoid code repeat in create sample inputs for sort/msort,https://github.com/pytorch/pytorch/issues/57301,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection, module: tests",scan when time allows,2021-08-30T02:57:16Z +issue,56275,cumprod gradgradcheck fails in fast_mode=True,https://github.com/pytorch/pytorch/issues/56275,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: tests",scan when time allows,2021-08-30T02:56:35Z +issue,60748,"Record file/line number when creating test data, and then report it in backtraces associated with this data",https://github.com/pytorch/pytorch/issues/60748,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-08-30T02:52:43Z +issue,60916,Scipy 1.7.0 may cause some test failures,https://github.com/pytorch/pytorch/issues/60916,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: tests",scan when time allows,2021-08-30T02:50:40Z +issue,62925,[Testing] memory_format decorator,https://github.com/pytorch/pytorch/issues/62925,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: tests",scan when time allows,2021-08-30T02:45:34Z +issue,63837,`torch.scatter` doesn't fail correctly on CUDA (memory overlap),https://github.com/pytorch/pytorch/issues/63837,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: scatter & gather ops, module: structured kernels, module: tests",scan when time allows,2021-08-30T02:39:38Z +issue,49909,Test functionality to detect extra cross device synchronizations,https://github.com/pytorch/pytorch/issues/49909,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2021-08-29T23:01:55Z +issue,64079,torch.equal does not support sparse tensors,https://github.com/pytorch/pytorch/issues/64079,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-08-27T05:16:46Z +issue,63555,pytorch 1.8.1+cu111 used much more CPU RAM than pytorch 1.8.1 after run `import torch`,https://github.com/pytorch/pytorch/issues/63555,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpu, module: cuda, module: memory usage",scan when time allows,2021-08-27T03:28:17Z +issue,61890,The matrix multiplication operator can't get correct results on 3090 !!,https://github.com/pytorch/pytorch/issues/61890,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tf32",scan when time allows,2021-08-26T20:40:53Z +issue,52052,common_jit.check_against_reference() should check gradients w.r.t. weights,https://github.com/pytorch/pytorch/issues/52052,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: jit",scan when time allows,2021-08-26T16:09:57Z +issue,58720,`pickle_save`/`pickle_load` are undocumented,https://github.com/pytorch/pytorch/issues/58720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-08-26T16:09:07Z +issue,54394,Half Normal Log_Prob not defined for 0,https://github.com/pytorch/pytorch/issues/54394,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: half, module: numerical-stability, module: random",scan when time allows,2021-08-26T14:55:50Z +issue,63781,[rfc] Hardcoded Target Determination,https://github.com/pytorch/pytorch/issues/63781,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-08-26T06:01:09Z +issue,63998,cat out= variant memory_format inconsistent for cpu and cuda,https://github.com/pytorch/pytorch/issues/63998,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory format",scan when time allows,2021-08-26T03:47:46Z +issue,16458,TensorRTOpTest.test_vgg19 is flaky,https://github.com/pytorch/pytorch/issues/16458,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests,scan when time allows,2021-08-25T23:50:41Z +issue,62849,"slice_embed, select_embed like diag_embed",https://github.com/pytorch/pytorch/issues/62849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: viewing and reshaping",scan when time allows,2021-08-25T17:53:13Z +issue,63720,Torch.save with zip serialization causes memory bloat,https://github.com/pytorch/pytorch/issues/63720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-08-25T16:43:54Z +issue,63475,C++ Inference error causes Identity Module,https://github.com/pytorch/pytorch/issues/63475,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-08-25T08:50:20Z +issue,54743,[Feature request] Add batched matrix support for torch.diag,https://github.com/pytorch/pytorch/issues/54743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: linear algebra, module: numpy",scan when time allows,2021-08-25T05:55:19Z +issue,31571,Bug in ForkingPickler for multiprocessing spawn context for shared storages on Linux,https://github.com/pytorch/pytorch/issues/31571,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-08-25T04:50:42Z +issue,63897,Scalar construction in C++: PY`torch.tensor(7)` !~= C++`at::tensor(7)`,https://github.com/pytorch/pytorch/issues/63897,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: tensor creation",scan when time allows,2021-08-24T23:27:37Z +issue,17234,[feature request] build and move distributions w/ device and/or dtype,https://github.com/pytorch/pytorch/issues/17234,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: distributions",scan when time allows,2021-08-24T19:54:31Z +issue,63849,"Installed successfully the torch-1.9 and torch_xla-1.9. RuntimeError requested XLA_GPU:0，but available devices are [CPU:0, XLA_CPU:0 ]",https://github.com/pytorch/pytorch/issues/63849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: xla",scan when time allows,2021-08-24T17:37:15Z +issue,17257,Generated `__init__.pyi` contains invalid default values,https://github.com/pytorch/pytorch/issues/17257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-08-24T15:36:00Z +issue,17268,Generic object to tensor dispatching,https://github.com/pytorch/pytorch/issues/17268,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-08-24T15:34:59Z +issue,17286,Use standard docker image for XLA build,https://github.com/pytorch/pytorch/issues/17286,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, module: xla, oncall: releng",scan when time allows,2021-08-24T15:33:48Z +issue,20117,[FR] [RFC] add Sequential.append & .extend,https://github.com/pytorch/pytorch/issues/20117,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-08-24T13:43:30Z +issue,62588,static library kineto_LIBRARY-NOTFOUND not found.,https://github.com/pytorch/pytorch/issues/62588,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2021-08-24T02:11:19Z +issue,7343,[memory leak] [PyTorch] .backward(create_graph=True),https://github.com/pytorch/pytorch/issues/7343,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",scan when time allows,2021-08-23T23:15:42Z +issue,62162,Clarify sparse COO tensor coalesce behavior wrt overflow + how to binarize a sparse tensor,https://github.com/pytorch/pytorch/issues/62162,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sparse",scan when time allows,2021-08-23T14:53:45Z +issue,63644,LeakyReLU and Elu use more VRAM than needed,https://github.com/pytorch/pytorch/issues/63644,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2021-08-23T14:14:14Z +issue,59263,Ubuntu 20.04 GeForce RTX 3070 / NVIDIA-SMI 460.73.01 Driver Version: 460.73.01 / GPU Visibility Issue,https://github.com/pytorch/pytorch/issues/59263,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2021-08-22T05:38:28Z +issue,61182,"Feature Request: torch.functional.interpolate to quietly ignore ""align_corners"" when mode is set to ""nearest""",https://github.com/pytorch/pytorch/issues/61182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn",scan when time allows,2021-08-21T23:52:14Z +issue,62352,[feature request] Expand a given dim,https://github.com/pytorch/pytorch/issues/62352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2021-08-21T16:21:18Z +issue,63618,C++ torch::cuda::synchronize speeds up training,https://github.com/pytorch/pytorch/issues/63618,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-08-21T13:29:31Z +issue,40427,"Can we make torch.inverse FP16? - RuntimeError: ""inverse_cuda"" not implemented for 'Half'",https://github.com/pytorch/pytorch/issues/40427,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: linear algebra",scan when time allows,2021-08-20T16:18:47Z +issue,63624,Raw saved tensors can survive the deletion of the underlying SavedVariable object,https://github.com/pytorch/pytorch/issues/63624,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-08-20T14:12:30Z +issue,63559,"Functional grid_sample: example of padding_mode=""reflection"" description error",https://github.com/pytorch/pytorch/issues/63559,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: interpolation, module: nn",scan when time allows,2021-08-19T21:52:19Z +issue,63574,compile_commands.json doesn't contain nvcc invocations,https://github.com/pytorch/pytorch/issues/63574,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-08-19T21:42:37Z +issue,63485,Preserve tensor subclasses when unpacking a SavedTensor,https://github.com/pytorch/pytorch/issues/63485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-08-19T14:54:29Z +issue,18106,copy.copy not working for ScriptModule,https://github.com/pytorch/pytorch/issues/18106,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-08-18T23:39:42Z +issue,59526,Documentation for torch.finfo doesn't match implementation,https://github.com/pytorch/pytorch/issues/59526,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy",scan when time allows,2021-08-18T21:06:57Z +issue,38642,Synchronization problem in torch.distributed with MPI on CUDA,https://github.com/pytorch/pytorch/issues/38642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-08-18T16:00:07Z +issue,63477,Supporting Pytorch for Custom Compiler Backend,https://github.com/pytorch/pytorch/issues/63477,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: dispatch",scan when time allows,2021-08-18T13:30:33Z +issue,63441,OpenCV causes backpropagation to get stuck,https://github.com/pytorch/pytorch/issues/63441,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: multithreading",scan when time allows,2021-08-18T04:00:31Z +issue,57610,[perf] 10x improvement when doing `x.sum(-1)` manually,https://github.com/pytorch/pytorch/issues/57610,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: reductions",scan when time allows,2021-08-17T22:00:08Z +issue,63290,DistributedDataParallel documentation,https://github.com/pytorch/pytorch/issues/63290,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-08-17T19:24:45Z +issue,16896,`nn.Linear` allows 1d input tensors,https://github.com/pytorch/pytorch/issues/16896,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-17T17:59:49Z +issue,62955,Make axes selection keyword arguments in torch.diagonal and torch.transpose consistent,https://github.com/pytorch/pytorch/issues/62955,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: numpy, module: python array api",scan when time allows,2021-08-17T08:32:19Z +issue,62265,`periodic-pytorch-linux-xenial-cuda11.3-cudnn8-py3.6-gcc7-build` started to fail,https://github.com/pytorch/pytorch/issues/62265,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-08-16T22:08:31Z +issue,63226,"cuda.is_available returns True in pycharms python console, False in code",https://github.com/pytorch/pytorch/issues/63226,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-08-16T21:56:19Z +issue,63217,Torch.nn.functional.interpolate() function memory release problem,https://github.com/pytorch/pytorch/issues/63217,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: memory usage",scan when time allows,2021-08-16T21:54:46Z +issue,63313,Automatic mixed precision works worse for 3D neural networks.,https://github.com/pytorch/pytorch/issues/63313,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: performance",scan when time allows,2021-08-16T21:54:25Z +issue,63295,pytorch 1.8.2 cuda test errors,https://github.com/pytorch/pytorch/issues/63295,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: build, module: cuda, module: linear algebra",scan when time allows,2021-08-16T21:53:39Z +issue,63356,[python codegen] correctly plumb TensorOptions defaults through the python binding layer.,https://github.com/pytorch/pytorch/issues/63356,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-08-16T20:52:02Z +issue,63299,Channels last mode doesn't work with inputs with 0 batch dimension,https://github.com/pytorch/pytorch/issues/63299,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: nn",scan when time allows,2021-08-16T17:39:32Z +issue,63023,[feature request] Type promotions for Boolean tensors with sub operation + Numpy compatability,https://github.com/pytorch/pytorch/issues/63023,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: numpy, module: type promotion",scan when time allows,2021-08-16T17:17:56Z +issue,63282,Serialising `torch.bool` generates a warning about `np.bool` being deprecated.,https://github.com/pytorch/pytorch/issues/63282,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: numpy, module: serialization",scan when time allows,2021-08-16T01:13:05Z +issue,63257,`help(torch.finfo)` doesn't give useful information even though class has documentation,https://github.com/pytorch/pytorch/issues/63257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-15T16:19:45Z +issue,63211,Memory leak in `ddp_zero_hook.py` hooks,https://github.com/pytorch/pytorch/issues/63211,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: memory usage",scan when time allows,2021-08-15T16:14:59Z +issue,63207,"document our compatibility with upstream packages (e.g. languages, compilers, tools)",https://github.com/pytorch/pytorch/issues/63207,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-15T16:14:26Z +issue,63205,split up test/test_autograd.py,https://github.com/pytorch/pytorch/issues/63205,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-08-13T13:16:20Z +issue,63016,grid_sample should not require grid to be the same dimension as input.,https://github.com/pytorch/pytorch/issues/63016,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn",scan when time allows,2021-08-13T02:54:32Z +issue,7020,[Caffe2] LT and GT cannot be exported to ONNX,https://github.com/pytorch/pytorch/issues/7020,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-08-12T23:25:07Z +issue,6505,[Caffe2] Onnx to Caffe2 Blob error,https://github.com/pytorch/pytorch/issues/6505,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-08-12T23:13:25Z +issue,63069,CUDA error: invalid configuration argument for torch.sparse tensor backward,https://github.com/pytorch/pytorch/issues/63069,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2021-08-12T22:47:30Z +issue,63057,Find out why TestSparseCPU.test_softmax_cpu_float64 grad and input dtypes are different,https://github.com/pytorch/pytorch/issues/63057,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2021-08-12T22:46:30Z +issue,60933,jit.load error when changed Folder Name,https://github.com/pytorch/pytorch/issues/60933,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-08-12T22:39:54Z +issue,62606,Guidance on where to add documentation,https://github.com/pytorch/pytorch/issues/62606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-12T21:50:31Z +issue,63082,script/build_android nn::Module support,https://github.com/pytorch/pytorch/issues/63082,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2021-08-12T20:51:33Z +issue,62989,Doc page for functions with overload is not properly indented,https://github.com/pytorch/pytorch/issues/62989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-12T19:31:30Z +issue,55609,requires_grad does not get propagated properly when using the JIT compiler,https://github.com/pytorch/pytorch/issues/55609,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-08-12T18:59:25Z +issue,63078,torch.cuda.amp fails with torch.sparse.softmax,https://github.com/pytorch/pytorch/issues/63078,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: sparse",scan when time allows,2021-08-12T18:55:16Z +issue,38964,[JIT] jit can not recognize the imported function,https://github.com/pytorch/pytorch/issues/38964,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-08-12T06:05:49Z +issue,60724,bilinear interpolate is very slow under mixed precision training mode.,https://github.com/pytorch/pytorch/issues/60724,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: interpolation, module: nn",scan when time allows,2021-08-12T04:16:20Z +issue,57815,copy.deepcopy fails on spectral-norm layers after the first forward,https://github.com/pytorch/pytorch/issues/57815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-08-12T03:38:13Z +issue,40208,C++ API for torch.autograd.functional.jacobian,https://github.com/pytorch/pytorch/issues/40208,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpp",scan when time allows,2021-08-12T00:21:07Z +issue,60164,Training goes wrong with amp and no_grad since pytorch 1.8,https://github.com/pytorch/pytorch/issues/60164,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda",scan when time allows,2021-08-11T15:12:10Z +issue,20359,[feature request] Run examples from docs as tests,https://github.com/pytorch/pytorch/issues/20359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-11T15:03:20Z +issue,42427,./scripts/build_android.sh need support nn::module nn::Functional nn::Linear,https://github.com/pytorch/pytorch/issues/42427,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2021-08-11T14:16:55Z +issue,35723,LibTorch API on Mobile,https://github.com/pytorch/pytorch/issues/35723,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: mobile",scan when time allows,2021-08-11T13:46:28Z +issue,45073,Compilation errors on power-pc,https://github.com/pytorch/pytorch/issues/45073,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2021-08-11T02:04:22Z +issue,14095,[feature request] bincount along specified dimension(s),https://github.com/pytorch/pytorch/issues/14095,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: sorting and selection",scan when time allows,2021-08-10T22:25:25Z +issue,62888,Add a mechanism to not universally dispatch to python for all for all operations when using __torch_dispatch__,https://github.com/pytorch/pytorch/issues/62888,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_dispatch__,scan when time allows,2021-08-10T18:19:13Z +issue,62171,[feature request] torch.nn.Conv3d on tensors having more than 5 dimensions,https://github.com/pytorch/pytorch/issues/62171,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",scan when time allows,2021-08-09T21:05:20Z +issue,62931,GPU compatibility check without initializing CUDA,https://github.com/pytorch/pytorch/issues/62931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-08-09T16:20:49Z +issue,16873,Allow building C++ custom ops that imports another custom ops,https://github.com/pytorch/pytorch/issues/16873,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2021-08-09T15:47:43Z +issue,62934,`tools/autograd/derivatives.yaml` doesn't support methods on optional tensor,https://github.com/pytorch/pytorch/issues/62934,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: codegen",scan when time allows,2021-08-09T15:25:38Z +issue,45203,Staged backend boxed fallback (per-operator precomputation / precompute),https://github.com/pytorch/pytorch/issues/45203,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-08-09T14:49:11Z +issue,62845,allow specification of variadic arguments in native_functions.yaml,https://github.com/pytorch/pytorch/issues/62845,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-08-09T13:49:25Z +issue,23151,Deterministic mode for scatter_add operation,https://github.com/pytorch/pytorch/issues/23151,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: scatter & gather ops",scan when time allows,2021-08-08T10:08:40Z +issue,58979,gradient layout contract,https://github.com/pytorch/pytorch/issues/58979,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-08-07T07:01:29Z +issue,62865,Multiplication of non-int torch.tensor with string gives unexpected TypeError,https://github.com/pytorch/pytorch/issues/62865,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2021-08-06T15:43:13Z +issue,62862,Convert unset variables in `Reducer::Timer` to use c10::optional,https://github.com/pytorch/pytorch/issues/62862,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ddp,scan when time allows,2021-08-06T14:39:30Z +issue,62884,Conjugate fallback dispatch key should be per-backend,https://github.com/pytorch/pytorch/issues/62884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: dispatch",scan when time allows,2021-08-06T14:36:45Z +issue,16708,Better include path when compiling mkldnn,https://github.com/pytorch/pytorch/issues/16708,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-08-06T14:35:21Z +issue,16700,[feature request] Store accumulated gradients in separate GPU or on CPU memory,https://github.com/pytorch/pytorch/issues/16700,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-08-06T14:33:59Z +issue,62784,Runtime error when passing dim as None in torch.squeeze (and Tensor.squeeze),https://github.com/pytorch/pytorch/issues/62784,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-08-06T06:51:10Z +issue,48714,RTX 3090 setup vs 2x RTX 2080TI setup slower? Help..,https://github.com/pytorch/pytorch/issues/48714,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: performance",scan when time allows,2021-08-06T06:04:08Z +issue,62790,Remove c10d::kDefaultFirstBucketBytes,https://github.com/pytorch/pytorch/issues/62790,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: ddp, oncall: distributed",scan when time allows,2021-08-05T01:11:23Z +issue,62742,Add a way to edit CI env variables from PRs,https://github.com/pytorch/pytorch/issues/62742,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-08-04T18:52:48Z +issue,56038,[Meta] PyTorch features build/test matrix,https://github.com/pytorch/pytorch/issues/56038,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2021-08-04T18:35:44Z +issue,16695,Check whether _cudnn_rnn_flatten_weight can avoid changing the TensorImpl or Storage pointer of tensors in `weight_arr`,https://github.com/pytorch/pytorch/issues/16695,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: cudnn",scan when time allows,2021-08-04T17:28:10Z +issue,16668,Reduce fragmentation with CUDA caching allocator when using many streams,https://github.com/pytorch/pytorch/issues/16668,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2021-08-04T17:27:49Z +issue,16556,"KeyError: 'No translator registered for layer: name: ""res3b_relu_norm""\ntype: ""Normalize""\nbottom: ""res3b""\ntop: ""res3b_relu_norm""\nnorm_param",https://github.com/pytorch/pytorch/issues/16556,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-08-04T17:15:58Z +issue,16523,"The word ""ints"" is not rendered correclty in docs.",https://github.com/pytorch/pytorch/issues/16523,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-08-04T17:09:25Z +issue,16432,A bug in parallel.data_parallel when module_kwargs is not None,https://github.com/pytorch/pytorch/issues/16432,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: cuda, module: data parallel, module: error checking",scan when time allows,2021-08-04T17:06:14Z +issue,62500,meta tensor `set_` seems fishy,https://github.com/pytorch/pytorch/issues/62500,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: meta tensors",scan when time allows,2021-08-04T16:11:04Z +issue,62448,Dynamic tensor rematerialization,https://github.com/pytorch/pytorch/issues/62448,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: internals",scan when time allows,2021-08-04T16:06:57Z +issue,62699,Can not parse the caffe2 pretrain model. google.protobuf.message.DecodeError: Error parsing message,https://github.com/pytorch/pytorch/issues/62699,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-08-04T14:32:00Z +issue,61730,"Can't specify compiler, leading to SUPPORT_GLIBCXX_USE_C99 failed",https://github.com/pytorch/pytorch/issues/61730,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-08-04T09:40:36Z +issue,62591,Unnecessary python bindings and documentation for internal functions/ops,https://github.com/pytorch/pytorch/issues/62591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-08-04T05:31:24Z +issue,40147,SyncBatchNorm doesn't work when I set track_running_stats False,https://github.com/pytorch/pytorch/issues/40147,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-08-04T05:17:14Z +issue,62655,Make KernelFunction::makeFromUnboxedFunctor infer KernelFunctor from input argument,https://github.com/pytorch/pytorch/issues/62655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: dispatch",scan when time allows,2021-08-03T23:12:07Z +issue,62656,"Building pytorch from source results in ""an object with that name is already defined"" error on import",https://github.com/pytorch/pytorch/issues/62656,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: pybind",scan when time allows,2021-08-03T22:50:59Z +issue,62650,Error while importing torch: libtorch_python.so: undefined symbol: PyThread_tss_alloc,https://github.com/pytorch/pytorch/issues/62650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: abi,scan when time allows,2021-08-03T15:49:36Z +issue,57136,Intermittent segmentation fault when using rpc_async/rpc_sync with CUDA tensors,https://github.com/pytorch/pytorch/issues/57136,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-08-03T14:27:14Z +issue,62602,"""test_events_wait"" is flaky on ROCm",https://github.com/pytorch/pytorch/issues/62602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: rocm",scan when time allows,2021-08-03T14:18:21Z +issue,62446,[docs] Can't get header links to h3 on docs pages,https://github.com/pytorch/pytorch/issues/62446,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2021-08-03T08:06:39Z +issue,52236,CosineAnnealingWarmRestarts LR scheduler fails when lash_epoch != -1,https://github.com/pytorch/pytorch/issues/52236,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-08-03T07:39:34Z +issue,62566,`c10::optional` operators should delegate to corresponding operators on `T`,https://github.com/pytorch/pytorch/issues/62566,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-08-02T19:48:52Z +issue,62451,torch.Generator ignores the device argument?,https://github.com/pytorch/pytorch/issues/62451,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2021-08-02T15:28:59Z +issue,62449,Review replacing test/test_namedtuple_return_api.py with an OpInfo-based test,https://github.com/pytorch/pytorch/issues/62449,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-07-30T03:53:44Z +issue,61591,Add `map_location` arg for c10d collective comms,https://github.com/pytorch/pytorch/issues/61591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-29T21:06:39Z +issue,62399,Understatement about how running_var is computed in BatchNorm,https://github.com/pytorch/pytorch/issues/62399,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: norms and normalization",scan when time allows,2021-07-29T19:15:09Z +issue,61659,Python program locks up when using some innocuous code inside mp.spawn(),https://github.com/pytorch/pytorch/issues/61659,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-28T23:47:44Z +issue,62343,maybe_wrap_dim should have an example attached to it,https://github.com/pytorch/pytorch/issues/62343,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-07-28T18:38:06Z +issue,62147,Tensor subclassing does not support bool input,https://github.com/pytorch/pytorch/issues/62147,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: tensor creation",scan when time allows,2021-07-28T03:30:35Z +issue,16328,Should torch.arange take a layout parameter?,https://github.com/pytorch/pytorch/issues/16328,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sparse, module: tensor creation, module: ux",scan when time allows,2021-07-27T19:26:00Z +issue,62273,Serialization map_location silently ignores xla/mlc/meta (any serialization mechanism that skips storage),https://github.com/pytorch/pytorch/issues/62273,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-07-27T18:47:57Z +issue,62271,Refactor serialization tests to use device parametrization,https://github.com/pytorch/pytorch/issues/62271,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: serialization, module: tests",scan when time allows,2021-07-27T18:47:33Z +issue,38995,[FR] [distributed] coalesced primitives,https://github.com/pytorch/pytorch/issues/38995,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-27T17:41:53Z +issue,37063,Misleading documentation in torch.nn.functional.fold,https://github.com/pytorch/pytorch/issues/37063,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-07-27T05:37:52Z +issue,62026,Should we be upcasting integral types to int64 in torch.sum and torch.prod?,https://github.com/pytorch/pytorch/issues/62026,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python array api, module: reductions",scan when time allows,2021-07-26T22:10:24Z +issue,61662,`super().__init__()` usage for `_Joinable`,https://github.com/pytorch/pytorch/issues/61662,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-26T19:01:40Z +issue,61661,Return `Future` instead of `Work` in `notify_join_context()`,https://github.com/pytorch/pytorch/issues/61661,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-26T19:01:31Z +issue,28794,torch.from_file not documented,https://github.com/pytorch/pytorch/issues/28794,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-07-26T02:45:13Z +issue,62130,"nn.DataParallel not working on A100 with nvidia-driver 470.42.01, Cuda 11.1, Debian 10",https://github.com/pytorch/pytorch/issues/62130,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: multi-gpu",scan when time allows,2021-07-26T02:28:35Z +issue,32083,build pytorch from source fauled: undefined reference to `cusparseGetErrorString(cusparseStatus_t)',https://github.com/pytorch/pytorch/issues/32083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-24T12:46:18Z +issue,62009,multi-input and multi-output support for nn.Module,https://github.com/pytorch/pytorch/issues/62009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-07-23T02:36:11Z +issue,35041,Make operator registrations truly commutative using priority,https://github.com/pytorch/pytorch/issues/35041,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-07-22T15:57:03Z +issue,62031,Caffe2 tensor to ATen conversion doesn't initialize PyTorch CUDA state,https://github.com/pytorch/pytorch/issues/62031,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: internals",scan when time allows,2021-07-22T15:36:19Z +issue,62016,ppc64le build fail: invalid conversion from Bfloat16 in functional_base.h:180,https://github.com/pytorch/pytorch/issues/62016,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: power",scan when time allows,2021-07-22T15:32:26Z +issue,61772,Can I set the algorithm of torch.nn.Conv to CUDNN_CONVOLUTION FWD_ALGO_GEMM by myself?,https://github.com/pytorch/pytorch/issues/61772,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn",scan when time allows,2021-07-21T23:44:01Z +issue,57406,nn.Module custom setattr leads very confusing behaviors,https://github.com/pytorch/pytorch/issues/57406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-07-21T06:47:35Z +issue,61930,NNAPI delegate library cannot be built on MacOS,https://github.com/pytorch/pytorch/issues/61930,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2021-07-20T21:30:38Z +issue,61923,Calling `torch.save()` twice writes a corrupted file,https://github.com/pytorch/pytorch/issues/61923,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-07-20T19:29:52Z +issue,61919,RuntimeError: !ref.requires_grad()INTERNAL ASSERT FAILED,https://github.com/pytorch/pytorch/issues/61919,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-07-20T19:14:53Z +issue,60426,"Output of non-inplace, non-CompositeImplicitAutograd op has TensorImpl > 1 or StorageImpl use_count != 1",https://github.com/pytorch/pytorch/issues/60426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-07-20T15:46:27Z +issue,58093,"TrilinearBackward takes 98.4% of total computational time, is this to be expected?",https://github.com/pytorch/pytorch/issues/58093,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: performance",scan when time allows,2021-07-20T15:36:07Z +issue,16148,CategoricalCrossEntropy Loss runs with wrong tag,https://github.com/pytorch/pytorch/issues/16148,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2021-07-20T13:17:07Z +issue,61839,died with ,https://github.com/pytorch/pytorch/issues/61839,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-20T08:36:58Z +issue,16090,build failure with NNAPI enabled (Caffe2 path but under pytorch umbrella),https://github.com/pytorch/pytorch/issues/16090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-19T21:36:36Z +issue,61827,Bail out if `-std=c++` setting is detected during build,https://github.com/pytorch/pytorch/issues/61827,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-19T21:35:49Z +issue,61843,Investigate fairscale's use of storage APIs and find alternatives,https://github.com/pytorch/pytorch/issues/61843,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-07-19T21:34:19Z +issue,60956,Naming convention for variants of operations that never short circuit,https://github.com/pytorch/pytorch/issues/60956,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2021-07-19T19:06:34Z +issue,61695,miss header file while using python setup.py install to install pytorch with anaconda env,https://github.com/pytorch/pytorch/issues/61695,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-19T17:31:25Z +issue,61538,The use of epsilon in torch.nn.functional.pairwise_distance leads to incorrect minima of the distances,https://github.com/pytorch/pytorch/issues/61538,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: distance functions, module: nn",scan when time allows,2021-07-19T17:13:47Z +issue,61513,Adding 'differentiable' section to each function docstring,https://github.com/pytorch/pytorch/issues/61513,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2021-07-19T17:10:54Z +issue,61825,Improved OpInfo dtype testing,https://github.com/pytorch/pytorch/issues/61825,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-07-19T05:46:54Z +issue,57534,`torch.view_as_complex()` does not work when `storage_offset` is odd,https://github.com/pytorch/pytorch/issues/57534,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2021-07-19T03:07:05Z +issue,48832,cudnn cannot be pickled by cloudpickle,https://github.com/pytorch/pytorch/issues/48832,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: serialization",scan when time allows,2021-07-16T17:59:42Z +issue,61665,Default Collate doesn't work for subtypes of ndarray,https://github.com/pytorch/pytorch/issues/61665,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-07-16T16:21:33Z +issue,61718,ATen codegen is always re-running for Makefile generator,https://github.com/pytorch/pytorch/issues/61718,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-15T19:13:54Z +issue,61658,Move `torch.cholesky_solve` into `torch.linalg`.,https://github.com/pytorch/pytorch/issues/61658,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2021-07-15T14:26:45Z +issue,61410,requires grad get lost during transform.,https://github.com/pytorch/pytorch/issues/61410,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: deprecation, module: tensor creation",scan when time allows,2021-07-15T06:24:51Z +issue,61645,`channel_shuffle` output is sometimes aliased with its input,https://github.com/pytorch/pytorch/issues/61645,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2021-07-14T15:46:49Z +issue,61619,"When 'trapezoid' is called with an empty tensor input, it does not produce an output with requires_grad",https://github.com/pytorch/pytorch/issues/61619,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-07-14T15:20:07Z +issue,61614,Possibly flaky test: TestMultiprocessing.test_fd_sharing,https://github.com/pytorch/pytorch/issues/61614,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests,scan when time allows,2021-07-14T15:18:41Z +issue,61650,"Move `{svd,pca}_lowrank` into `torch.linalg`",https://github.com/pytorch/pytorch/issues/61650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2021-07-14T15:10:17Z +issue,51014,[pipeline] gpu util + peak mem reporting to tune partitions and chunks,https://github.com/pytorch/pytorch/issues/51014,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-13T19:16:53Z +issue,61149,`pad(mode='reflect')` beyond `input.size`,https://github.com/pytorch/pytorch/issues/61149,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: padding",scan when time allows,2021-07-13T16:38:21Z +issue,61486,Some reduction operators have double signatures,https://github.com/pytorch/pytorch/issues/61486,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: python array api, module: reductions",scan when time allows,2021-07-13T15:06:12Z +issue,61544,`__torch_function__` always records default kwargs as kwargs,https://github.com/pytorch/pytorch/issues/61544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2021-07-13T14:25:56Z +issue,61503,UCI Data Sets,https://github.com/pytorch/pytorch/issues/61503,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-07-13T13:39:46Z +issue,31943,error when building pytorch 1.1.0 from source,https://github.com/pytorch/pytorch/issues/31943,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-13T12:12:50Z +issue,61452,New module: Split log softmax with loss,https://github.com/pytorch/pytorch/issues/61452,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-07-13T08:01:44Z +issue,61516,MemoryError with pip Install even with --no-cache-dir option,https://github.com/pytorch/pytorch/issues/61516,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2021-07-13T04:39:59Z +issue,61523,F.nll_loss with 16-bit CUDA tensors and reduction=mean produces NaNs,https://github.com/pytorch/pytorch/issues/61523,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nans and infs, module: numerical-stability, module: reductions",scan when time allows,2021-07-12T22:49:16Z +issue,51509,Massive Performance bottlenecks in some of the Reduce operations.,https://github.com/pytorch/pytorch/issues/51509,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: reductions, module: tensoriterator",scan when time allows,2021-07-12T21:35:39Z +issue,61515,Buggy page with overlapped tabs,https://github.com/pytorch/pytorch/issues/61515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-07-12T21:29:24Z +issue,61525,"I have graphic card of P4, but torch.cuda.is_available() always shows false",https://github.com/pytorch/pytorch/issues/61525,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-07-12T20:46:24Z +issue,61528,Function Request: scipy.ndimage.map_coordinates,https://github.com/pytorch/pytorch/issues/61528,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2021-07-12T20:42:46Z +issue,61136,rename(**{}) should throw no error,https://github.com/pytorch/pytorch/issues/61136,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2021-07-12T16:10:53Z +issue,61510,What is find_package(Torch REQUIRED) doing that a manual include/glob doesnt?,https://github.com/pytorch/pytorch/issues/61510,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-12T15:39:08Z +issue,15812,Possible regression in incremental build,https://github.com/pytorch/pytorch/issues/15812,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-07-12T15:38:01Z +issue,61485,CompositeImplicitAutograd ops should not call out= variants of operators,https://github.com/pytorch/pytorch/issues/61485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-07-12T15:13:11Z +issue,15617,Feature request: transposed locally connected layer,https://github.com/pytorch/pytorch/issues/15617,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-07-12T13:32:44Z +issue,15607,documentation for adding a new type via C++ extensions,https://github.com/pytorch/pytorch/issues/15607,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: complex, module: docs",scan when time allows,2021-07-12T13:28:20Z +issue,61411,"I do have the GPU, but the print(torch.cuda.is_available()) always shows False",https://github.com/pytorch/pytorch/issues/61411,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-07-11T14:19:31Z +issue,40550,RemoteModule enhancements,https://github.com/pytorch/pytorch/issues/40550,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: rpc",scan when time allows,2021-07-10T06:45:41Z +issue,58745,"`torch.(min|max)(..., dim=...)` diverges from array API specification",https://github.com/pytorch/pytorch/issues/58745,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: python array api, module: reductions",scan when time allows,2021-07-09T22:41:39Z +issue,56926,CUDA 11 tracking issue,https://github.com/pytorch/pytorch/issues/56926,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2021-07-09T17:55:39Z +issue,61453,"Reduce with any(), all(), median() over multiple dimensions",https://github.com/pytorch/pytorch/issues/61453,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: reductions",scan when time allows,2021-07-09T17:02:56Z +issue,55442,More AVX2 vectorization support for half (float16),https://github.com/pytorch/pytorch/issues/55442,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: vectorization",scan when time allows,2021-07-09T16:15:25Z +issue,61186,possible SummaryWriter memory leak,https://github.com/pytorch/pytorch/issues/61186,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2021-07-09T15:43:17Z +issue,61289,Zero seed in DistributedSampler,https://github.com/pytorch/pytorch/issues/61289,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-07-08T23:47:58Z +issue,54082,Dispatch table for linalg_norm is fishy,https://github.com/pytorch/pytorch/issues/54082,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: norms and normalization",scan when time allows,2021-07-08T14:12:25Z +issue,51456,Optional tensor type,https://github.com/pytorch/pytorch/issues/51456,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-07-07T19:50:10Z +issue,59218,Clarify BLAS configuration option,https://github.com/pytorch/pytorch/issues/59218,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs, module: linear algebra",scan when time allows,2021-07-07T17:45:39Z +issue,61261,libtorch_cpu.so is getting too large.,https://github.com/pytorch/pytorch/issues/61261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: releng,scan when time allows,2021-07-07T04:14:29Z +issue,61260,libtorch_cpu.so: undefined symbol: _ZN3c1010ThreadPool3runESt8functionIFvvEE,https://github.com/pytorch/pytorch/issues/61260,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: build",scan when time allows,2021-07-06T21:33:54Z +issue,9678,Remove BatchNorm layers once the training is completed.,https://github.com/pytorch/pytorch/issues/9678,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2021-07-06T20:31:43Z +issue,5157,BCELoss - weight parameter shape incorrect,https://github.com/pytorch/pytorch/issues/5157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2021-07-06T20:30:21Z +issue,61213,[feature request] Low pass filtering of FFT results,https://github.com/pytorch/pytorch/issues/61213,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2021-07-06T20:18:57Z +issue,61211,[feature request] Do zero-padding in high-frequency modes in `ifft`,https://github.com/pytorch/pytorch/issues/61211,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: fft,scan when time allows,2021-07-06T20:18:48Z +issue,61233,[typing] ModuleDict ctor has wrong signature,https://github.com/pytorch/pytorch/issues/61233,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2021-07-06T19:45:39Z +issue,59099,CSR: Relaxing constraints to s_addmm_out_sparse_dense_cuda_worker,https://github.com/pytorch/pytorch/issues/59099,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-07-06T13:22:12Z +issue,49752,IndexError: Input _features.0.weight is undefined!,https://github.com/pytorch/pytorch/issues/49752,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2021-07-05T06:02:55Z +issue,56522,[cpp op] TORCH_LIBRARY schema doesn't respect keyword only,https://github.com/pytorch/pytorch/issues/56522,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cpp-extensions",scan when time allows,2021-07-04T19:10:44Z +issue,22415,Add support for serializing Mkldnn Tensor,https://github.com/pytorch/pytorch/issues/22415,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, module: serialization",scan when time allows,2021-07-04T14:22:36Z +issue,29758,[doc] Tensor.mean: dtype kwarg is not documented,https://github.com/pytorch/pytorch/issues/29758,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: reductions",scan when time allows,2021-07-02T01:33:58Z +issue,27312,torch.Tensor.mean erroneously documented as sometimes returning a tuple,https://github.com/pytorch/pytorch/issues/27312,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: reductions",scan when time allows,2021-07-02T01:33:58Z +issue,18904,[FR] torch.dist along a dimension,https://github.com/pytorch/pytorch/issues/18904,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2021-07-02T01:31:00Z +issue,56586,"Support multi-dim reductions for torch.prod, torch.all, torch.any",https://github.com/pytorch/pytorch/issues/56586,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2021-07-02T01:23:13Z +issue,60381,broken source link / missing anchors,https://github.com/pytorch/pytorch/issues/60381,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-07-01T20:05:54Z +issue,61117,flatten renames in place,https://github.com/pytorch/pytorch/issues/61117,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2021-07-01T16:31:13Z +issue,47027,Fix the way imports are done to be more correct for static type checkers,https://github.com/pytorch/pytorch/issues/47027,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2021-07-01T15:57:23Z +issue,28239,support class annotations in __init__,https://github.com/pytorch/pytorch/issues/28239,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-07-01T15:40:35Z +issue,45565,Resolving overloads and default arguments from within __torch_function__,https://github.com/pytorch/pytorch/issues/45565,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2021-07-01T14:51:16Z +issue,59936,Can't perform any operation on Vulkan device - macOS M1,https://github.com/pytorch/pytorch/issues/59936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: macos, module: vulkan",scan when time allows,2021-07-01T14:03:34Z +issue,20206,Failed to build with system protobuf,https://github.com/pytorch/pytorch/issues/20206,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: protobuf",scan when time allows,2021-07-01T12:09:32Z +issue,60541,torch.cuda.Event(blocking=True) doesn't work,https://github.com/pytorch/pytorch/issues/60541,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-06-30T14:32:39Z +issue,60997,Dispatcher doesn't handle ops with an empty list of tensors (e.g. `torch.cat()`),https://github.com/pytorch/pytorch/issues/60997,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-06-30T00:15:55Z +issue,60941,Incorrect dtype cast for binary ops w/ mixed ComplexFloat + Double operands,https://github.com/pytorch/pytorch/issues/60941,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: type promotion",scan when time allows,2021-06-30T00:01:29Z +issue,60950,Singular build setting CLANG_CXX_LANGUAGE_STANDARD has different values,https://github.com/pytorch/pytorch/issues/60950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2021-06-29T23:57:37Z +issue,60977,[package] pattern usage can potentially create broken packages,https://github.com/pytorch/pytorch/issues/60977,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: package/deploy,scan when time allows,2021-06-29T22:12:27Z +issue,60991,[NNC] APIs needed to bridge scheduling gap in reductions,https://github.com/pytorch/pytorch/issues/60991,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-29T21:26:23Z +issue,56187,AVX512 and Vec512,https://github.com/pytorch/pytorch/issues/56187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vectorization,scan when time allows,2021-06-29T03:58:13Z +issue,60911,[elastic launcher] redirects/tee support for global rank,https://github.com/pytorch/pytorch/issues/60911,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: elastic,scan when time allows,2021-06-28T23:57:33Z +issue,53658,torch.distributed with NCCL can hang if the first operation is a barrier,https://github.com/pytorch/pytorch/issues/53658,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2021-06-28T20:44:30Z +issue,54591,Various issues in derivatives.yaml,https://github.com/pytorch/pytorch/issues/54591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-06-28T17:47:20Z +issue,60802,Bug in `pytorch/benchmarks/tensorexpr/pt_engine.py`,https://github.com/pytorch/pytorch/issues/60802,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-28T16:21:30Z +issue,60796,AccumulateGrad short circuits before calling variable hooks when the incoming grad is None,https://github.com/pytorch/pytorch/issues/60796,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-06-28T16:04:32Z +issue,60843,PyTorch's last 1 year update became unfriendly due to the lack of examples,https://github.com/pytorch/pytorch/issues/60843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-28T15:38:57Z +issue,60848,Fail to build with gcc11,https://github.com/pytorch/pytorch/issues/60848,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: xnnpack",scan when time allows,2021-06-28T15:29:41Z +issue,60431,Implement simple view methods for TensorAccessor/PackedTensorAccessor,https://github.com/pytorch/pytorch/issues/60431,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2021-06-28T15:10:56Z +issue,60834,complie error when i use python setup.py build on arm computer,https://github.com/pytorch/pytorch/issues/60834,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",scan when time allows,2021-06-28T13:40:18Z +issue,46272,test_distributed_* does not work with run_test.py -i option,https://github.com/pytorch/pytorch/issues/46272,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",scan when time allows,2021-06-28T13:34:47Z +issue,47281,Error: invalid use of register,https://github.com/pytorch/pytorch/issues/47281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-28T07:40:38Z +issue,60520,[docs] torch.log_softmax is undocumented,https://github.com/pytorch/pytorch/issues/60520,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-28T03:47:08Z +issue,49810,Undefined reference to `cv::imread`,https://github.com/pytorch/pytorch/issues/49810,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-27T12:12:31Z +issue,15386,F.grid_sample doesn't respect padding_mode when height of inputs is 1,https://github.com/pytorch/pytorch/issues/15386,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: interpolation",scan when time allows,2021-06-25T21:58:12Z +issue,60766,"_foreach_maximum returns ""incompatible type"" mypy error unexpectedly",https://github.com/pytorch/pytorch/issues/60766,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2021-06-25T21:49:12Z +issue,60306,Divergent code is needed to record usage streams on different TensorImpl types,https://github.com/pytorch/pytorch/issues/60306,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse, module: vmap",scan when time allows,2021-06-25T04:54:04Z +issue,53952,[Pipe] supporting None and non-Tensors in forward's input/output,https://github.com/pytorch/pytorch/issues/53952,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-25T00:25:04Z +issue,60674,Turn on -Winfinite-recursion in OSS CI,https://github.com/pytorch/pytorch/issues/60674,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-24T19:39:31Z +issue,37751,Documentation of _CtxMethodMixin: must be tensors?,https://github.com/pytorch/pytorch/issues/37751,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2021-06-24T15:35:48Z +issue,39370,Draw quantized tensors in tensorboard,https://github.com/pytorch/pytorch/issues/39370,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2021-06-24T11:17:10Z +issue,56161,"[rfc] Add a ""core only"" build flag",https://github.com/pytorch/pytorch/issues/56161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-24T10:18:22Z +issue,60372,Docs for `Installing C++ Distributions of PyTorch` needs update,https://github.com/pytorch/pytorch/issues/60372,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-24T05:33:47Z +issue,60459,[Mkldnn] has_bf16 check only works on Linux for tests,https://github.com/pytorch/pytorch/issues/60459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, module: tests",scan when time allows,2021-06-24T05:24:56Z +issue,60288,provide hashsum for downloads,https://github.com/pytorch/pytorch/issues/60288,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2021-06-24T04:51:24Z +issue,60400,Using inbuilt function name `input` as variable names in examples,https://github.com/pytorch/pytorch/issues/60400,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-24T02:09:24Z +issue,60561,nn.CrossMapLRN2d is missing from docs,https://github.com/pytorch/pytorch/issues/60561,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-06-24T01:44:32Z +issue,60146,[torch.profiler] enhancements + corrections,https://github.com/pytorch/pytorch/issues/60146,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,scan when time allows,2021-06-24T00:26:03Z +issue,47149,Allow all torch.nn modules to accept arbitrary batch dimensions,https://github.com/pytorch/pytorch/issues/47149,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: nn, module: ux",scan when time allows,2021-06-23T21:45:42Z +issue,57157,Using register_full_backward_hook with target module for intermediate activation,https://github.com/pytorch/pytorch/issues/57157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs, module: nn",scan when time allows,2021-06-23T15:51:45Z +issue,60277,Sparse CSR tensor should not accept equal column indices in the same row,https://github.com/pytorch/pytorch/issues/60277,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: sparse",scan when time allows,2021-06-23T14:59:46Z +issue,14790,Tensor.copy_() seems to work improperly with numpy/list indices,https://github.com/pytorch/pytorch/issues/14790,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-23T11:22:24Z +issue,52575,Error: ‘str’ is not a member of ‘c10’; did you mean ‘c10::aten::str’? while using libtorch,https://github.com/pytorch/pytorch/issues/52575,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-06-23T11:15:58Z +issue,60448,Enable large file support,https://github.com/pytorch/pytorch/issues/60448,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-23T02:25:41Z +issue,57773,BFloat16 CUDA GEMM ops unsupported on Nvidia P100 (SM_60) on CUDA 11.3,https://github.com/pytorch/pytorch/issues/57773,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: cublas, module: cuda",scan when time allows,2021-06-22T19:01:45Z +issue,32288,Allow range in dim argument of reducing operations such as sum,https://github.com/pytorch/pytorch/issues/32288,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: reductions, module: ux",scan when time allows,2021-06-22T18:21:53Z +issue,21760,[FR] faster reduce sum on expanded/unfolded tensors,https://github.com/pytorch/pytorch/issues/21760,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2021-06-22T18:19:13Z +issue,60343,What changes we need to make in metrics calculation and visualization part when we use Distributed Data Parallel for distributed training,https://github.com/pytorch/pytorch/issues/60343,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-06-22T07:15:26Z +issue,37113,pip install torch==1.4.0 is broken when using CUDA 10.1,https://github.com/pytorch/pytorch/issues/37113,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-21T21:51:50Z +issue,58814,TensorIterator for sparse layouts,https://github.com/pytorch/pytorch/issues/58814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-06-21T21:25:06Z +issue,60159,Some ops throw runtime errors when called with complex tensors that require_grad but work when requires_grad=False,https://github.com/pytorch/pytorch/issues/60159,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: complex",scan when time allows,2021-06-21T16:22:08Z +issue,57158,Increasing RTT when using rpc_async in a PS architecture setup,https://github.com/pytorch/pytorch/issues/57158,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-06-21T15:35:27Z +issue,12659,Differentiation through Module parameters updates,https://github.com/pytorch/pytorch/issues/12659,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn, module: optimizer",scan when time allows,2021-06-20T20:56:02Z +issue,60096,Tutorials that require `pip install` not loading correctly in Colab,https://github.com/pytorch/pytorch/issues/60096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-20T13:46:06Z +issue,50323,running a job on multiple gpus with qsub,https://github.com/pytorch/pytorch/issues/50323,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-06-20T06:49:12Z +issue,45448,CuDNN version not found,https://github.com/pytorch/pytorch/issues/45448,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cudnn",scan when time allows,2021-06-20T01:54:07Z +issue,60264,Subtensor operations like `tril/triu` to have an option to respect the strides of the input,https://github.com/pytorch/pytorch/issues/60264,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: memory format",scan when time allows,2021-06-19T22:49:38Z +issue,41852,Difference in inference time between CUDA 10.0 & 10.2,https://github.com/pytorch/pytorch/issues/41852,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: performance",scan when time allows,2021-06-19T13:25:23Z +issue,60165,torch.load non backwards compatible on Transformer between 1.8.1 and 1.9.0,https://github.com/pytorch/pytorch/issues/60165,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-06-19T09:51:26Z +issue,44159,"[Feature Request] Add an option to run GPU tests only, and skip all CPU tests",https://github.com/pytorch/pytorch/issues/44159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-06-18T18:02:16Z +issue,15298,Momentum problem (1-momentum is correct?) in BatchNorm2d,https://github.com/pytorch/pytorch/issues/15298,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-06-18T15:26:54Z +issue,15303,CMake Error at CMakeLists.txt:10 (find_package) in C++,https://github.com/pytorch/pytorch/issues/15303,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-18T15:26:15Z +issue,59935,Error /usr/local/lib/libopenblas.so: error adding symbols: File in wrong format while building pytorch for ppc64le,https://github.com/pytorch/pytorch/issues/59935,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2021-06-18T06:21:17Z +issue,60153,[oneline docs] adding a source link for non-python functions,https://github.com/pytorch/pytorch/issues/60153,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-18T05:17:14Z +issue,15259,err:torch.nn.CrossEntropyLoss,https://github.com/pytorch/pytorch/issues/15259,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-17T22:35:54Z +issue,15260,MultiGPU for gru,https://github.com/pytorch/pytorch/issues/15260,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: rnn",scan when time allows,2021-06-17T22:34:54Z +issue,15284,index_add_ with scalar values instead of tensors,https://github.com/pytorch/pytorch/issues/15284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2021-06-17T22:32:41Z +issue,15288,possible unsafety in torch.distributions.kl_divergence for Bernoullis,https://github.com/pytorch/pytorch/issues/15288,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: nans and infs",scan when time allows,2021-06-17T22:31:36Z +issue,59855,pytorch test failed,https://github.com/pytorch/pytorch/issues/59855,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda, module: tests",scan when time allows,2021-06-17T20:18:38Z +issue,60102,Allow packaging of classes defined in a notebook,https://github.com/pytorch/pytorch/issues/60102,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-17T05:02:16Z +issue,45428,Use libtorch for online inference: an illegal memory access was encountered,https://github.com/pytorch/pytorch/issues/45428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-06-17T03:32:59Z +issue,59925,Implicit module registration permits silent programming errors in torch.nn.Module,https://github.com/pytorch/pytorch/issues/59925,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-06-16T21:23:10Z +issue,60063,Replace native NHWC BN kernel,https://github.com/pytorch/pytorch/issues/60063,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-06-16T19:17:42Z +issue,60092,`memory_format` flag for `tensor.new_empty()`.,https://github.com/pytorch/pytorch/issues/60092,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-06-16T18:53:07Z +issue,60083,caffe_translator.py Unable to convert Power Layer,https://github.com/pytorch/pytorch/issues/60083,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-16T15:23:23Z +issue,59783,[Feature Request] Deterministic implementation for AdaptiveMaxpool1d.,https://github.com/pytorch/pytorch/issues/59783,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: nn, module: pooling",scan when time allows,2021-06-16T09:06:52Z +issue,15142,Update third_party/googletest - Ability to skip tests in GTEST,https://github.com/pytorch/pytorch/issues/15142,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, module: third_party",scan when time allows,2021-06-15T22:24:05Z +issue,17688,Improve save() method in torch.jit.ScriptModule,https://github.com/pytorch/pytorch/issues/17688,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-06-15T22:06:00Z +issue,58841,[bug] torch.topk sometimes supports `float16` and sometimes doesn't,https://github.com/pytorch/pytorch/issues/58841,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: sorting and selection",scan when time allows,2021-06-14T20:10:58Z +issue,59896,Autograd engine current graph_task should be in the TLSState,https://github.com/pytorch/pytorch/issues/59896,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-06-14T18:14:52Z +issue,59854,Misreported linting messages,https://github.com/pytorch/pytorch/issues/59854,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2021-06-14T16:15:55Z +issue,36066,Add Specific Warning/Error For Unsupported GPU or Systems,https://github.com/pytorch/pytorch/issues/36066,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-06-11T23:10:13Z +issue,59778,CUDA Memory Error: PyTorch doen't allocate memory even though it's available,https://github.com/pytorch/pytorch/issues/59778,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2021-06-11T04:46:43Z +issue,47028,torch.autograd.backward() fails to sync with other stream,https://github.com/pytorch/pytorch/issues/47028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-06-11T00:10:14Z +issue,59822,TransformedDistribution's log_prob gradient is inconsistent w.r.t. cache_size,https://github.com/pytorch/pytorch/issues/59822,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-06-10T23:09:49Z +issue,56774,Support `expand_dims`,https://github.com/pytorch/pytorch/issues/56774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: python array api",scan when time allows,2021-06-10T07:55:33Z +issue,59627,Example code for is_storage missing,https://github.com/pytorch/pytorch/issues/59627,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-10T07:45:32Z +issue,59493,ctc loss bug,https://github.com/pytorch/pytorch/issues/59493,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: loss",scan when time allows,2021-06-10T00:37:50Z +issue,59743,m.fallback(torch::CppFunction::makeFromBoxedFunction<&my_fallback>) gives bad error message,https://github.com/pytorch/pytorch/issues/59743,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: cpp, module: internals",scan when time allows,2021-06-09T21:09:42Z +issue,54332,Unit testing failures when porting PyTorch wheel,https://github.com/pytorch/pytorch/issues/54332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2021-06-09T18:12:41Z +issue,23780,Construction of MultivariateNormal much slower on GPU than CPU,https://github.com/pytorch/pytorch/issues/23780,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: distributions, module: performance",scan when time allows,2021-06-09T09:20:17Z +issue,49405,PyTorch 1.7.1 on (macOS/python 3.9/conda) links libtorch_global_deps.dylib with libomp.dylib instead of libiomp5.dylib,https://github.com/pytorch/pytorch/issues/49405,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build, module: macos",scan when time allows,2021-06-08T20:05:20Z +issue,59617,torch.split() infer -1 entry from the other split sizes,https://github.com/pytorch/pytorch/issues/59617,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2021-06-08T14:41:27Z +issue,59438,"DataLoader worker (pid(s) 18056, 20540, 4512) exited unexpectedly",https://github.com/pytorch/pytorch/issues/59438,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-06-08T14:03:41Z +issue,59542,Build Pytorch with Aten?,https://github.com/pytorch/pytorch/issues/59542,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-07T20:22:51Z +issue,59531,Failed to enable libuv,https://github.com/pytorch/pytorch/issues/59531,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-07T20:20:26Z +issue,59546,Multiprocessing model evaluation gets stuck,https://github.com/pytorch/pytorch/issues/59546,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-06-07T20:18:08Z +issue,59515,Conv1d with large batch size and half precision in cuda returns incorrect result,https://github.com/pytorch/pytorch/issues/59515,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cuda",scan when time allows,2021-06-07T20:02:13Z +issue,59575,[clang-tidy] check for iteration over unordered data structures,https://github.com/pytorch/pytorch/issues/59575,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-07T19:36:28Z +issue,56440,Raise exception when torch.clamp min value underflows the input tensor's dtype,https://github.com/pytorch/pytorch/issues/56440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2021-06-07T18:37:53Z +issue,58479,Make SavedVariables aware of pyobjects,https://github.com/pytorch/pytorch/issues/58479,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-06-07T18:12:36Z +issue,59262,"Possibly alignment issues on NEON vectorized ops, Jetson platforms",https://github.com/pytorch/pytorch/issues/59262,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: jetson",scan when time allows,2021-06-07T17:16:27Z +issue,59140,No index type check of torch.tensor in array manner,https://github.com/pytorch/pytorch/issues/59140,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: error checking",scan when time allows,2021-06-07T17:11:42Z +issue,14839,Test OpenCV4 in CI,https://github.com/pytorch/pytorch/issues/14839,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2021-06-07T15:28:33Z +issue,14797,Modern interface for Storage,https://github.com/pytorch/pytorch/issues/14797,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-06-07T15:25:49Z +issue,59386,Energy-efficiency benchmarks?,https://github.com/pytorch/pytorch/issues/59386,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-07T07:59:08Z +issue,12484,CuDNN convolution on some CUDA devices will not preserve NaN weights (upstream bug),https://github.com/pytorch/pytorch/issues/12484,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: dependency bug",scan when time allows,2021-06-07T02:15:20Z +issue,58997,[discussion] Should Optimizers be also Modules?,https://github.com/pytorch/pytorch/issues/58997,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer",scan when time allows,2021-06-06T11:31:22Z +issue,58973,First-class Java API,https://github.com/pytorch/pytorch/issues/58973,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: java,scan when time allows,2021-06-05T06:17:01Z +issue,59256,ProcessGroupGloo creation crashes when world_size > 150,https://github.com/pytorch/pytorch/issues/59256,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: data parallel, oncall: distributed",scan when time allows,2021-06-04T21:05:39Z +issue,59457,Error compiling PyTorch: `/usr/bin/ld: cannot find -lmagma /usr/bin/ld: cannot find -lnvToolsExt`,https://github.com/pytorch/pytorch/issues/59457,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-06-04T18:09:52Z +issue,12532,Provide better documentation for torch.Size,https://github.com/pytorch/pytorch/issues/12532,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-06-04T14:51:35Z +issue,14702,pytorch_doc_push is racing with itself,https://github.com/pytorch/pytorch/issues/14702,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-06-04T13:41:49Z +issue,59418,Libtorch segfault in packed GRU evaluation with cuda batch_sizes,https://github.com/pytorch/pytorch/issues/59418,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn, module: rnn",scan when time allows,2021-06-04T13:39:36Z +issue,55056,[FR] Safe softmax,https://github.com/pytorch/pytorch/issues/55056,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-06-03T21:27:49Z +issue,59216,Add a BlendLoss class in torch.nn and a blend_loss function in torch.nn.functional to allow blending of a list of loss functions,https://github.com/pytorch/pytorch/issues/59216,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2021-06-02T16:23:33Z +issue,59058,Issue: support auto generation of device check for sparse tensors,https://github.com/pytorch/pytorch/issues/59058,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: sparse",scan when time allows,2021-06-02T15:37:01Z +issue,40988,'_mm256_extract_epi64' was not declared in this scope when compiling on Debian 32-bit,https://github.com/pytorch/pytorch/issues/40988,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-02T14:32:38Z +issue,59247,Support complex numbers in `at::nan_to_num`.,https://github.com/pytorch/pytorch/issues/59247,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-06-02T04:49:24Z +issue,58860,[cuDNN v8] Extend current cuDNN convolution v8 API binding to support conv-bias-activation fusion,https://github.com/pytorch/pytorch/issues/58860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2021-06-01T23:02:41Z +issue,58847,Dataloader Rerunning with num_workers=0 may give better error trace,https://github.com/pytorch/pytorch/issues/58847,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: ddp, oncall: distributed",scan when time allows,2021-06-01T22:03:02Z +issue,38639,Difference between using a python list and nn.ModuleList,https://github.com/pytorch/pytorch/issues/38639,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-06-01T21:45:19Z +issue,42689,How can I specify NumPy while building PyTorch?,https://github.com/pytorch/pytorch/issues/42689,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-01T21:40:50Z +issue,14436,Use of STL templates in cpu/ directory (compiling with different AVX settings) is silently hazardous,https://github.com/pytorch/pytorch/issues/14436,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-01T21:22:21Z +issue,14467,"Optimizer warning when parameters ""change""",https://github.com/pytorch/pytorch/issues/14467,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-06-01T21:13:49Z +issue,14477,"Add a debug mode which is -O0 for framework code, but -O for kernels",https://github.com/pytorch/pytorch/issues/14477,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-06-01T21:12:49Z +issue,59251,Ahead of time C++ Extensions should have an option to build with verbose=False,https://github.com/pytorch/pytorch/issues/59251,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions,scan when time allows,2021-06-01T20:49:56Z +issue,59101,Use of storage_offset is not needed in add_out_dense_sparse_csr_cuda,https://github.com/pytorch/pytorch/issues/59101,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: sparse",scan when time allows,2021-06-01T14:03:35Z +issue,59071,test_bottleneck_cuda fails without error message,https://github.com/pytorch/pytorch/issues/59071,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2021-06-01T13:56:13Z +issue,59104,Data access pattern in the loop in add_out_dense_sparse_csr_cuda could be pretty bad,https://github.com/pytorch/pytorch/issues/59104,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-06-01T03:06:53Z +issue,49445,[doc] how to prevent pytorch-nightly from being replaced by a released version on pip install,https://github.com/pytorch/pytorch/issues/49445,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: docs, oncall: releng",scan when time allows,2021-05-31T17:06:32Z +issue,44489,for_each doesn't support integer input to float output type promotion,https://github.com/pytorch/pytorch/issues/44489,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-31T01:07:51Z +issue,59185,Add supports_nnc metadata to OpInfos,https://github.com/pytorch/pytorch/issues/59185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-05-30T08:04:24Z +issue,56896,masked_select is x3 slower than reshaping and index_select,https://github.com/pytorch/pytorch/issues/56896,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: performance, module: viewing and reshaping",scan when time allows,2021-05-29T21:58:15Z +issue,58758,precision/consistency issue in `linspace`,https://github.com/pytorch/pytorch/issues/58758,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility, module: numpy, module: tensor creation",scan when time allows,2021-05-29T07:41:35Z +issue,32706,Torch serialization does not restore tensors properly when custom __reduce__ is defined,https://github.com/pytorch/pytorch/issues/32706,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-05-27T23:14:06Z +issue,58596,Some cuda objects take very long time to build,https://github.com/pytorch/pytorch/issues/58596,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-05-27T07:22:53Z +issue,51644,"Segmentation Fault: Garbage collector, cuda memory",https://github.com/pytorch/pytorch/issues/51644,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-05-26T22:55:04Z +issue,58765,[codegen] generated inplace/out= wrappers don't have input checks,https://github.com/pytorch/pytorch/issues/58765,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: error checking",scan when time allows,2021-05-26T17:59:44Z +issue,58962,Operating on a shared memory tensor with multiple threads hangs,https://github.com/pytorch/pytorch/issues/58962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, oncall: distributed",scan when time allows,2021-05-26T15:33:03Z +issue,58980,libtorch conflict with cxxopts and cause memery leak,https://github.com/pytorch/pytorch/issues/58980,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2021-05-26T15:29:17Z +issue,54905,Allow CUDA build without requiring a physical GPU device.,https://github.com/pytorch/pytorch/issues/54905,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-05-26T08:40:48Z +issue,58522,DDP grads dont have parity with local training when grads are undefined,https://github.com/pytorch/pytorch/issues/58522,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-05-26T02:22:24Z +issue,58876,[deploy] Enable `torch.distributed.rpc` Python bindings in deploy,https://github.com/pytorch/pytorch/issues/58876,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deploy, module: rpc, oncall: distributed",scan when time allows,2021-05-26T00:05:39Z +issue,4145,[Proposal] Consistent `batch_first` effect for RNN modules,https://github.com/pytorch/pytorch/issues/4145,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn, module: rnn",scan when time allows,2021-05-25T22:34:49Z +issue,58944,Use PYTHONHASHSEED during pytorch build to avoid nondeterminism,https://github.com/pytorch/pytorch/issues/58944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: codegen, module: determinism",scan when time allows,2021-05-25T21:31:14Z +issue,58862,[cuDNN v8] Improve cuDNN convolution v8 API error reporting,https://github.com/pytorch/pytorch/issues/58862,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2021-05-25T19:41:55Z +issue,58859,[cuDNN v8] Extend current cuDNN convolution v8 API binding to support cuDNN benchmark,https://github.com/pytorch/pytorch/issues/58859,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2021-05-25T19:41:31Z +issue,58858,[cuDNN v8] Extend current cuDNN v8 API binding to support convolution backward and transposed convolution forward,https://github.com/pytorch/pytorch/issues/58858,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2021-05-25T19:40:16Z +issue,44172,build from scratch failed because of the expansion of macro,https://github.com/pytorch/pytorch/issues/44172,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-05-25T10:07:55Z +issue,53891,Check Archive.org for Failed Dataset Downloads,https://github.com/pytorch/pytorch/issues/53891,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dependency bug,scan when time allows,2021-05-24T21:47:08Z +issue,49480,Internal symbols are leaking from torch.nn.functional,https://github.com/pytorch/pytorch/issues/49480,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-24T19:24:52Z +issue,58626,GPU 0 context created on GPU 1 worker when using pin_memory=True,https://github.com/pytorch/pytorch/issues/58626,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader",scan when time allows,2021-05-24T17:04:25Z +issue,58471,Why the same training code running on different GPUs (V100 vs P100) brings a significiant difference in outputs and thus train loss / model accuracy?,https://github.com/pytorch/pytorch/issues/58471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: determinism,scan when time allows,2021-05-24T16:53:16Z +issue,58447,TSAN issues in at::RecordFunction,https://github.com/pytorch/pytorch/issues/58447,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: profiler,scan when time allows,2021-05-24T15:26:28Z +issue,50068,Magma functions that don't have queue argument create cublas handles for each call,https://github.com/pytorch/pytorch/issues/50068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2021-05-24T02:06:58Z +issue,19267,torch.distributed.broadcast should default to current stream,https://github.com/pytorch/pytorch/issues/19267,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-05-23T22:38:32Z +issue,58820,[docs] Tensor.bernoulli_ formatting is hard to read and UX inconsistent with the function variant,https://github.com/pytorch/pytorch/issues/58820,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2021-05-23T21:13:37Z +issue,58770,MKL csr matmul issue,https://github.com/pytorch/pytorch/issues/58770,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkl, module: sparse",scan when time allows,2021-05-23T18:14:52Z +issue,55355,simple matrix multiplication yields wrong result on Ampere (3080),https://github.com/pytorch/pytorch/issues/55355,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: cublas, module: tf32",scan when time allows,2021-05-23T05:51:14Z +issue,58657,"Improvement suggestions for the error message ""Expected batch2_sizes[0] == bs && batch2_sizes[1] == contraction_size to be true, but got false""",https://github.com/pytorch/pytorch/issues/58657,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: linear algebra",scan when time allows,2021-05-23T05:09:11Z +issue,58671,torch.load with dill is unable to unserialize from buffer,https://github.com/pytorch/pytorch/issues/58671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-05-23T03:40:31Z +issue,58016,Eliminate potential double device check,https://github.com/pytorch/pytorch/issues/58016,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-22T18:54:43Z +issue,58735,CSR construction: safe_get_attr_string suppresses real errors,https://github.com/pytorch/pytorch/issues/58735,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-05-22T05:21:43Z +issue,58791,[nnc] Add more optimization for conditional stmts in loopnest transformations,https://github.com/pytorch/pytorch/issues/58791,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-22T03:38:05Z +issue,58617,[rfc] Bug developers about release notes for new APIs,https://github.com/pytorch/pytorch/issues/58617,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-20T21:26:53Z +issue,35499,Support tensor.cumsum() for 1-dim tensors,https://github.com/pytorch/pytorch/issues/35499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-05-20T19:06:04Z +issue,58672,Werror should include -Wmissing-prototypes,https://github.com/pytorch/pytorch/issues/58672,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: build warnings",scan when time allows,2021-05-20T17:17:13Z +issue,58670,Ability to enabling/disabling cuDNN and cuBLAS API logging in PyTorch API directly,https://github.com/pytorch/pytorch/issues/58670,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: cudnn",scan when time allows,2021-05-20T17:11:19Z +issue,58591,"link error on BatchNormImplBase::pretty_print when using icpc",https://github.com/pytorch/pytorch/issues/58591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: undefined reference,scan when time allows,2021-05-20T16:40:23Z +issue,58585,Calling benchmark.Timer with default `num_threads=1` disables parallelism permanently,https://github.com/pytorch/pytorch/issues/58585,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: macos, module: multithreading",scan when time allows,2021-05-20T16:33:44Z +issue,50375,Provide a set of C++ foreach APIs that will take tensor pointers as an input,https://github.com/pytorch/pytorch/issues/50375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-05-20T14:59:30Z +issue,52553,Does pytorch have the inverse activation function of nn.LeakyReLU?,https://github.com/pytorch/pytorch/issues/52553,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-20T12:29:17Z +issue,58489,Inter-category type promotion has undocumented behavior,https://github.com/pytorch/pytorch/issues/58489,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: type promotion",scan when time allows,2021-05-20T07:13:57Z +issue,58442,TSAN issues in autograd,https://github.com/pytorch/pytorch/issues/58442,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-05-20T05:56:13Z +issue,58474,nn.Upsample result mismatch in 1.1.0a0+828a6a3 and 1.9.0,https://github.com/pytorch/pytorch/issues/58474,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2021-05-20T05:55:36Z +issue,23482,Build error due to unintended include path /usr/include,https://github.com/pytorch/pytorch/issues/23482,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-05-20T04:32:07Z +issue,21028,Zero-dim Tensors (scalars) should be printed at full precision,https://github.com/pytorch/pytorch/issues/21028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: printing,scan when time allows,2021-05-19T18:14:23Z +issue,58395,[DOC] Add info about differentiability of functions,https://github.com/pytorch/pytorch/issues/58395,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-05-19T06:48:14Z +issue,58352,Broadcasting multiple tensors to all procs in distributed training,https://github.com/pytorch/pytorch/issues/58352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-05-18T22:47:54Z +issue,56843,Add clipnorm parameter to optimizers,https://github.com/pytorch/pytorch/issues/56843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-05-18T19:22:03Z +issue,58371,Tests do not pass even though build successful,https://github.com/pytorch/pytorch/issues/58371,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-05-18T18:15:56Z +issue,58212,`torch.autograd.Function` subclasses *sometimes* throw away custom subclasses,https://github.com/pytorch/pytorch/issues/58212,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-05-17T23:15:12Z +issue,58351,need sampled softmax loss function to train model to get user vector and item vector with a large dictionary,https://github.com/pytorch/pytorch/issues/58351,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2021-05-17T20:23:52Z +issue,58343,Add the features like Keras,https://github.com/pytorch/pytorch/issues/58343,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2021-05-17T20:23:12Z +issue,58386,Better shared memory allocation under Docker,https://github.com/pytorch/pytorch/issues/58386,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, module: memory usage",scan when time allows,2021-05-17T17:41:49Z +issue,58385,"A ""sanitizer"" mode for the CUDA caching allocator?",https://github.com/pytorch/pytorch/issues/58385,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: profiler",scan when time allows,2021-05-17T17:27:50Z +issue,13818,test_spectral_norm: Backward is not reentrant,https://github.com/pytorch/pytorch/issues/13818,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: data parallel, module: norms and normalization",scan when time allows,2021-05-17T16:43:27Z +issue,13993,[tracking task] FBGEMM guarding AVX2 properly,https://github.com/pytorch/pytorch/issues/13993,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2021-05-17T16:36:53Z +issue,14047,Make torch.multiprocessing.SpawnContext usable,https://github.com/pytorch/pytorch/issues/14047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-05-17T16:35:12Z +issue,20424,pos_weight argument in torch.nn.BCELoss,https://github.com/pytorch/pytorch/issues/20424,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-05-17T04:34:56Z +issue,19529,failed to load model which is saved as text format(pickle_protocol=0) instead of binary format,https://github.com/pytorch/pytorch/issues/19529,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pickle, module: serialization",scan when time allows,2021-05-14T20:57:05Z +issue,53903,[torch.vmap] Support second order grads,https://github.com/pytorch/pytorch/issues/53903,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: vmap",scan when time allows,2021-05-14T06:30:56Z +issue,47993,"build pytorch from source on ubuntu, building error from fbgemm::SparseAdaGradSignature",https://github.com/pytorch/pytorch/issues/47993,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: collect_env.py",scan when time allows,2021-05-14T02:45:31Z +issue,58238,torchdeploy doesn't prevent Obj from being used on wrong interpreter,https://github.com/pytorch/pytorch/issues/58238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deploy,scan when time allows,2021-05-13T23:32:09Z +issue,58218,"Torch.nn.DataParallel training model, the output of the model becomes list type, the number of lists is batch size",https://github.com/pytorch/pytorch/issues/58218,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-05-13T21:42:37Z +issue,48439,cudnn convolution modifies the input Tensor metadata inplace when it tries to `.resize_()` it,https://github.com/pytorch/pytorch/issues/48439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: memory format",scan when time allows,2021-05-13T10:06:32Z +issue,56356,"Resolved: Only add type promotion support to unary pwise, binary pwise, and reduction operations",https://github.com/pytorch/pytorch/issues/56356,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: type promotion, module: ux",scan when time allows,2021-05-13T04:33:16Z +issue,58167,Allow `Union` as a `Dict` key,https://github.com/pytorch/pytorch/issues/58167,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-12T19:25:45Z +issue,58128,the same input with different batchsize got different precision output,https://github.com/pytorch/pytorch/issues/58128,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: nn, module: numerical-reproducibility",scan when time allows,2021-05-12T18:55:49Z +issue,58136,scatter_add_ 6000-times slower with int64 compared to int32,https://github.com/pytorch/pytorch/issues/58136,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: performance, module: scatter & gather ops",scan when time allows,2021-05-12T17:07:52Z +issue,58127,Save checkpoint error,https://github.com/pytorch/pytorch/issues/58127,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, module: serialization",scan when time allows,2021-05-12T15:52:22Z +issue,52457,Add autograd tests to verify correctness for R -> C cases,https://github.com/pytorch/pytorch/issues/52457,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: tests",scan when time allows,2021-05-12T14:08:24Z +issue,58037,linalg.eigh and linalg.cholesky UPLO flag breaks in backward,https://github.com/pytorch/pytorch/issues/58037,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra",scan when time allows,2021-05-12T09:22:19Z +issue,58124,Python API binding code generation does not need to pack TensorOptions for `xxx_like` generators,https://github.com/pytorch/pytorch/issues/58124,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-05-12T09:18:19Z +issue,58091,Follow-up updates to inference mode python bindings,https://github.com/pytorch/pytorch/issues/58091,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-12T04:06:01Z +issue,13782,torch.linspace does not check for infinity and nan,https://github.com/pytorch/pytorch/issues/13782,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: tensor creation",scan when time allows,2021-05-12T03:58:20Z +issue,58087,TensorList upfront checks parameter types in argument parsing and throws an error unexpectedly,https://github.com/pytorch/pytorch/issues/58087,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-05-12T03:56:47Z +issue,42155,BusError memory-mapped tensor,https://github.com/pytorch/pytorch/issues/42155,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-05-11T19:50:13Z +issue,57065,Static Linking Pytorch,https://github.com/pytorch/pytorch/issues/57065,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: static linking,scan when time allows,2021-05-11T06:20:37Z +issue,57960,Use the same CUDA stream for all RPCs within the same dist autograd context,https://github.com/pytorch/pytorch/issues/57960,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2021-05-10T23:19:34Z +issue,54941,Regression in Python arg parser performance.,https://github.com/pytorch/pytorch/issues/54941,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: performance",scan when time allows,2021-05-10T20:54:05Z +issue,57957,Add a few tests to make sure new dispatch keys for backends are added properly.,https://github.com/pytorch/pytorch/issues/57957,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-05-10T19:20:09Z +issue,57787,Add CI automation tests for Infiniband for DDP,https://github.com/pytorch/pytorch/issues/57787,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-05-10T16:44:49Z +issue,56460,"I meet an error assert key in deserialized_objects when I torch.load(pthname),and pth file is trained on multi gpu",https://github.com/pytorch/pytorch/issues/56460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: serialization, oncall: distributed",scan when time allows,2021-05-10T16:32:48Z +issue,57949,[structured] Clarify if NoNamesGuard is actually needed in impls or not,https://github.com/pytorch/pytorch/issues/57949,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2021-05-10T16:22:37Z +issue,54646,Optional enhanced logging for operator calls for backend implementor debugging,https://github.com/pytorch/pytorch/issues/54646,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2021-05-10T13:04:41Z +issue,34953,Graphic tool to view the backward(Gradient Graph) and forward graph in Pytorch,https://github.com/pytorch/pytorch/issues/34953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,scan when time allows,2021-05-10T11:24:27Z +issue,54232,DistributedDataParallel: `DDP(model)` hangs on non-master node,https://github.com/pytorch/pytorch/issues/54232,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-05-10T09:00:32Z +issue,57617,complex128 autograd failures on PPC,https://github.com/pytorch/pytorch/issues/57617,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: power",scan when time allows,2021-05-10T05:24:37Z +issue,57921,Backward pass for a nn.Conv2d with half-precision on Quadro RTX 8000 leads to CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/57921,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2021-05-10T04:32:03Z +issue,57843,Torch.save doesn't make new file,https://github.com/pytorch/pytorch/issues/57843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-05-10T04:14:58Z +issue,57646,"get_future() should be documented once it is enabled for gloo, mpi backends",https://github.com/pytorch/pytorch/issues/57646,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-05-07T22:45:16Z +issue,13508,OpenCL support for smartphones,https://github.com/pytorch/pytorch/issues/13508,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-07T22:42:54Z +issue,13268,Non-Zero Padding in Convolution Module,https://github.com/pytorch/pytorch/issues/13268,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",scan when time allows,2021-05-07T22:42:29Z +issue,56583,"All thread stack trace dumps interleave output in distributed tests, giving garbled output",https://github.com/pytorch/pytorch/issues/56583,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-05-07T22:33:46Z +issue,56772,The multi-fc losses calculating in DistributedDataParallel.,https://github.com/pytorch/pytorch/issues/56772,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-05-07T18:46:16Z +issue,57631,Pruners' compute_mask returns tensors with dtypes that are not consistent with each other,https://github.com/pytorch/pytorch/issues/57631,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pruning",scan when time allows,2021-05-07T17:42:08Z +issue,57712,[Docs] nn.init._calculate_fan_in_and_fan_out(),https://github.com/pytorch/pytorch/issues/57712,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-05-07T12:20:44Z +issue,32322,deadlock when using mp.spawn multiprocessing,https://github.com/pytorch/pytorch/issues/32322,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-05-07T04:10:49Z +issue,55981,"LazyEmbedding, an embedding layer with a dynamically sized vocabulary",https://github.com/pytorch/pytorch/issues/55981,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: nn",scan when time allows,2021-05-07T01:10:34Z +issue,57759,Show commit hash for master branch docs on pytorch.org,https://github.com/pytorch/pytorch/issues/57759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2021-05-06T19:04:20Z +issue,57611,[perf] 10x improvement on element-wise operations with manual broadcast,https://github.com/pytorch/pytorch/issues/57611,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: tensoriterator",scan when time allows,2021-05-06T18:17:51Z +issue,46653,Future returned by RPC should print a warning message on destruction if it's not waited,https://github.com/pytorch/pytorch/issues/46653,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: rpc, oncall: distributed",scan when time allows,2021-05-06T18:17:25Z +issue,57742,[structured] Meta-impl split implies redundant dtype tests,https://github.com/pytorch/pytorch/issues/57742,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-06T18:04:05Z +issue,56698,Convolution2D may have a numerical error.,https://github.com/pytorch/pytorch/issues/56698,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2021-05-06T09:27:55Z +issue,57547,Add distributed testing for CUDA aware MPI,https://github.com/pytorch/pytorch/issues/57547,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2021-05-06T07:17:12Z +issue,57691,Unify `std::getenv` usages,https://github.com/pytorch/pytorch/issues/57691,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-05T23:04:18Z +issue,55661,"RuntimeError: ""max_cuda"" not implemented for 'ComplexFloat'",https://github.com/pytorch/pytorch/issues/55661,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2021-05-05T21:31:54Z +issue,57440,"I have just wasted 3 solid days trying to build PyTorch, without success",https://github.com/pytorch/pytorch/issues/57440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-05-05T03:11:18Z +issue,57533,test_grid_sample (from TestNN) fails on POWER,https://github.com/pytorch/pytorch/issues/57533,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: power, module: tests",scan when time allows,2021-05-05T03:07:21Z +issue,57415,torch 1.8.1+cuda crashes when setting breakpoints.,https://github.com/pytorch/pytorch/issues/57415,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-05T01:17:35Z +issue,57503,[new feature] Adaptive gradient clipping,https://github.com/pytorch/pytorch/issues/57503,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2021-05-04T13:52:21Z +issue,57495,Please remove this assertion - it triggers on valid use cases,https://github.com/pytorch/pytorch/issues/57495,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-04T02:06:12Z +issue,57502,[structured] Fail if CompositeExplicitAutograd is in dispatch table of structured_delegate function,https://github.com/pytorch/pytorch/issues/57502,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-05-04T02:05:57Z +issue,57507,Cleanup usage of IS_PYTHON_3_9_PLUS in autograd engine,https://github.com/pytorch/pytorch/issues/57507,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: pybind",scan when time allows,2021-05-03T22:18:09Z +issue,57475,CUDA RPC stream synchronization does not work with @rpc.functions.async_execution,https://github.com/pytorch/pytorch/issues/57475,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2021-05-03T17:39:38Z +issue,57370,gen_proto_typestubs_helper.py is slow,https://github.com/pytorch/pytorch/issues/57370,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-05-03T17:11:10Z +issue,13053,Generalized Data Class,https://github.com/pytorch/pytorch/issues/13053,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2021-05-03T17:04:24Z +issue,57380,Asserts all tensor are defined in dispatch wrapper,https://github.com/pytorch/pytorch/issues/57380,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-03T16:08:18Z +issue,57392,Simplified API for custom inplace & view kernel,https://github.com/pytorch/pytorch/issues/57392,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: custom-operators, module: dispatch",scan when time allows,2021-05-03T15:03:26Z +issue,57369,Delay errors of inference tensor to backward pass?,https://github.com/pytorch/pytorch/issues/57369,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-05-03T15:01:02Z +issue,57358,Missing complex autograd support for some operators,https://github.com/pytorch/pytorch/issues/57358,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2021-05-03T14:58:28Z +issue,57326,Renaming Autograd dispatch keys to ADCreateGraph,https://github.com/pytorch/pytorch/issues/57326,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-05-03T14:57:39Z +issue,57359,torch.return_types does not exist,https://github.com/pytorch/pytorch/issues/57359,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-05-03T14:57:03Z +issue,55736,RuntimeError: CUDA error: CUBLAS_STATUS_INTERNAL_ERROR when calling `cublasCreate(handle)`,https://github.com/pytorch/pytorch/issues/55736,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-05-03T12:32:37Z +issue,53473,RuntimeError: CUDA error: CUBLAS_STATUS_INTERNAL_ERROR when calling `cublasCreate(handle)`,https://github.com/pytorch/pytorch/issues/53473,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2021-05-03T12:32:33Z +issue,57018,Attempting to concatenate scalar tensors throws a runtime error,https://github.com/pytorch/pytorch/issues/57018,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-05-02T14:19:05Z +issue,56975,scatter does not accept scalar src=,https://github.com/pytorch/pytorch/issues/56975,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy, module: scatter & gather ops",scan when time allows,2021-05-02T04:52:15Z +issue,56329,Change `other` to `src` in torch.Tensor.scatter_add_,https://github.com/pytorch/pytorch/issues/56329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-05-01T22:23:52Z +issue,56979,"apex-0.1-py3.6-linux-x86_64.egg/apex/amp/wrap.py"", line 28, in wrapper return orig_fn(*new_args, **kwargs) RuntimeError: CUDA error: an illegal memory access was encountered",https://github.com/pytorch/pytorch/issues/56979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda",scan when time allows,2021-04-30T02:42:43Z +issue,3356,Data sampling seems to be more complicated than necessary,https://github.com/pytorch/pytorch/issues/3356,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-04-29T22:26:59Z +issue,13207,[pytorch] [feature request] Error out if the needed GPU device capability is absent in runtime,https://github.com/pytorch/pytorch/issues/13207,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",scan when time allows,2021-04-29T16:25:12Z +issue,57108,Compilation failure in caffe2/core/plan_executor_test.cc with GCC 5.4,https://github.com/pytorch/pytorch/issues/57108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-29T03:23:40Z +issue,44814,LNK2019 error after static compilation of Libtorch,https://github.com/pytorch/pytorch/issues/44814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-28T22:09:37Z +issue,57095,[FR] unflatten support empty shape if unflattened dim has size 1,https://github.com/pytorch/pytorch/issues/57095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2021-04-28T17:44:02Z +issue,12962,Come up with a better strategy for noticing BC-breaking attribute additions to serializable classes,https://github.com/pytorch/pytorch/issues/12962,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: molly-guard",scan when time allows,2021-04-28T15:13:02Z +issue,12980,Installing pytorch from source on labs.cognitiveclass.ai,https://github.com/pytorch/pytorch/issues/12980,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: power,scan when time allows,2021-04-28T15:12:19Z +issue,13162,circular module reference raises RecursionError,https://github.com/pytorch/pytorch/issues/13162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-04-28T15:03:42Z +issue,13188,cudnn explicit paths and GCC multilib suffixes prevents detection of good cudnn headers,https://github.com/pytorch/pytorch/issues/13188,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-28T15:00:19Z +issue,13034,[caffe2]How can I export init_net.pb and predict_net.pb files on my own?,https://github.com/pytorch/pytorch/issues/13034,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-28T14:57:56Z +issue,57059,Get rid of weak_intrusive_ptr,https://github.com/pytorch/pytorch/issues/57059,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-04-28T14:52:02Z +issue,57112,:attr: does not behave as expected.,https://github.com/pytorch/pytorch/issues/57112,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: linear algebra",scan when time allows,2021-04-28T14:33:06Z +issue,57123,LibTorch ships with two identical library files: libdnnl and libmlkdnn,https://github.com/pytorch/pytorch/issues/57123,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-28T14:26:13Z +issue,54906,Out variants for Convolution ops,https://github.com/pytorch/pytorch/issues/54906,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2021-04-28T05:07:09Z +issue,56777,Error when building PyTorch from source in CentOS Linux 7,https://github.com/pytorch/pytorch/issues/56777,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: tensorpipe",scan when time allows,2021-04-28T02:34:23Z +issue,56881,[FR] integral discrete distributions should support dtype=int64,https://github.com/pytorch/pytorch/issues/56881,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-04-27T23:01:23Z +issue,56809,Autograd engine worker thread initialization fails with python 3.9 debug build,https://github.com/pytorch/pytorch/issues/56809,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-27T22:58:37Z +issue,56921,torch.multiprocessing implement SyncManager,https://github.com/pytorch/pytorch/issues/56921,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-04-27T22:57:30Z +issue,57005,about torch install,https://github.com/pytorch/pytorch/issues/57005,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-27T22:44:58Z +issue,56671,{h / v / d}split methods are missing,https://github.com/pytorch/pytorch/issues/56671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-04-27T18:46:16Z +issue,54499,Caught an integer overflow or wraparound in torch.nn.MaxPool1d,https://github.com/pytorch/pytorch/issues/54499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2021-04-27T17:59:06Z +issue,27343,[jit] `ScriptFunction`s are loaded as `ScriptModule`s,https://github.com/pytorch/pytorch/issues/27343,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-04-27T17:35:30Z +issue,56464,Scatter tests missing when passing a reduction operation.,https://github.com/pytorch/pytorch/issues/56464,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: scatter & gather ops,scan when time allows,2021-04-27T12:13:27Z +issue,54282,test_randperm is failing on CPU-only build,https://github.com/pytorch/pytorch/issues/54282,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-04-27T02:55:48Z +issue,56694,PyTorch 1.8.0 / 1.8.1 CUDA memory corruption on oddly sized tesnor sections of strided tensor,https://github.com/pytorch/pytorch/issues/56694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-04-26T19:50:53Z +issue,56480,shared torch.tensor with multiprocesses using python Queue cause coredump,https://github.com/pytorch/pytorch/issues/56480,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-04-26T15:34:12Z +issue,34257,CustomFromMask pruning stores a copy of the user-provided mask,https://github.com/pytorch/pytorch/issues/34257,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pruning",scan when time allows,2021-04-26T12:07:14Z +issue,18536,Make it easier to figure out what CuDNN convolution algorithm we actually chose,https://github.com/pytorch/pytorch/issues/18536,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: logging",scan when time allows,2021-04-24T00:45:57Z +issue,51896,Can't reassign a child module instance to a lambda,https://github.com/pytorch/pytorch/issues/51896,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-04-23T19:01:13Z +issue,56805,Implement random SeedSequence,https://github.com/pytorch/pytorch/issues/56805,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2021-04-23T18:55:13Z +issue,12895,RelaxedOneHotCategorical not implementing entropy (and other abstract methods),https://github.com/pytorch/pytorch/issues/12895,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-04-23T18:41:24Z +issue,56781,Named Tensors in C++ Is Undocumented,https://github.com/pytorch/pytorch/issues/56781,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: named tensor",scan when time allows,2021-04-23T15:37:48Z +issue,56764,Change make_reduction to reflect input resizing.,https://github.com/pytorch/pytorch/issues/56764,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2021-04-23T15:31:13Z +issue,47650,DataParallel support for scripted modules in C++,https://github.com/pytorch/pytorch/issues/47650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-04-23T13:51:10Z +issue,15716,"Nn.dataparallel with multiple output, weird gradient result None",https://github.com/pytorch/pytorch/issues/15716,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-04-23T10:07:03Z +issue,56483,Add compute_residuals flag for torch.linalg.lstsq,https://github.com/pytorch/pytorch/issues/56483,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: numpy",scan when time allows,2021-04-23T05:48:34Z +issue,56710,Hard fail build when there is no CUDA but `USE_CUDA=1`,https://github.com/pytorch/pytorch/issues/56710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-23T04:47:37Z +issue,56691,[structured] at::cpu_unchecked functions,https://github.com/pytorch/pytorch/issues/56691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2021-04-23T04:41:28Z +issue,56542,CrossEntropyLoss target shape expectation is inconsistent with BCEWithLogits,https://github.com/pytorch/pytorch/issues/56542,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: shape checking,scan when time allows,2021-04-23T04:27:59Z +issue,56525,Can't deepcopy memory format objects,https://github.com/pytorch/pytorch/issues/56525,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-04-23T04:27:13Z +issue,34880,Expanded tensors don't work as registered buffers,https://github.com/pytorch/pytorch/issues/34880,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn, module: partial aliasing",scan when time allows,2021-04-22T16:48:14Z +issue,30246,CUDA masked_select uses way too much memory,https://github.com/pytorch/pytorch/issues/30246,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2021-04-22T16:38:45Z +issue,18730,Numerical instability KL divergence RelaxedOneHotCategorical,https://github.com/pytorch/pytorch/issues/18730,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: numerical-stability",scan when time allows,2021-04-22T14:09:10Z +issue,56634,[package] Module name reported in error message does not always match what is needed to extern/mock it,https://github.com/pytorch/pytorch/issues/56634,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-21T21:43:22Z +issue,49666,Use rlibm for faster and more accurate floating point operations,https://github.com/pytorch/pytorch/issues/49666,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bfloat16,scan when time allows,2021-04-21T20:41:24Z +issue,56300,torch.where with input Tensor and other Scalar raises type mismatch error,https://github.com/pytorch/pytorch/issues/56300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2021-04-21T17:28:18Z +issue,56370,"Acquiring ""is_grad_enabled()"" inside an autograd function",https://github.com/pytorch/pytorch/issues/56370,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: custom-operators",scan when time allows,2021-04-21T02:52:51Z +issue,56465,Provide a pkg-config file,https://github.com/pytorch/pytorch/issues/56465,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-20T19:56:05Z +issue,56482,unsafe_reclaim_from_nonowning is not that unsafe,https://github.com/pytorch/pytorch/issues/56482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: internals",scan when time allows,2021-04-20T19:46:54Z +issue,12715,pack_padded_sequence throws IndexError when only kwargs are specified,https://github.com/pytorch/pytorch/issues/12715,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2021-04-20T18:19:28Z +issue,12716,cdf in torch.distributions.bernoulli throws NotImplementedError,https://github.com/pytorch/pytorch/issues/12716,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-04-20T18:18:47Z +issue,12797,Port dragon4_scientific for pretty float tensor print.,https://github.com/pytorch/pytorch/issues/12797,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: printing,scan when time allows,2021-04-20T18:18:18Z +issue,12812,Pytorch BatchNorm2D Unstable,https://github.com/pytorch/pytorch/issues/12812,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2021-04-20T18:17:33Z +issue,12828,BUILD_BINARY is a lie,https://github.com/pytorch/pytorch/issues/12828,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-20T18:15:41Z +issue,12830,Test that (cd build && ninja) immediately after build is no-op in CI,https://github.com/pytorch/pytorch/issues/12830,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2021-04-20T18:14:41Z +issue,56183,"scripting LSTM module generates a script that has no .code, no .graph, and cannot be executed.",https://github.com/pytorch/pytorch/issues/56183,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-04-20T17:46:03Z +issue,12868,[CAPI] Increase of memory usage when exporting a Adam optimzer,https://github.com/pytorch/pytorch/issues/12868,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: memory usage, module: optimizer",scan when time allows,2021-04-20T15:55:42Z +issue,54655,qsize not implemented error,https://github.com/pytorch/pytorch/issues/54655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: macos, module: multiprocessing",scan when time allows,2021-04-20T01:54:21Z +issue,56100,[package] implicit externing can miss cases where the stdlib depends on further stdlib modules,https://github.com/pytorch/pytorch/issues/56100,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-20T01:45:42Z +issue,56399,Figure out what mojo FB common/process/StackTrace.h has that we don't,https://github.com/pytorch/pytorch/issues/56399,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-04-19T20:06:55Z +issue,56350,logcumsumexp (and maybe other cum* ops) has divergent CUDA and CPU out= behavior,https://github.com/pytorch/pytorch/issues/56350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: safe resize",scan when time allows,2021-04-19T18:23:44Z +issue,56330,Type checking doesn't match actual type in error message,https://github.com/pytorch/pytorch/issues/56330,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2021-04-19T15:24:43Z +issue,56328,Why does output_padding have constraints in ConvTranspose1d?,https://github.com/pytorch/pytorch/issues/56328,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-04-19T15:22:07Z +issue,56297,Audit destructors of classes bound using pybind11::class_ to see if they can block; such cases can deadlock,https://github.com/pytorch/pytorch/issues/56297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2021-04-19T14:37:39Z +issue,56357,[NNC] All NNC APIs should return bool flag,https://github.com/pytorch/pytorch/issues/56357,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-19T06:58:40Z +issue,24401,"When running model forward with large batch size, it reports the error: THCudaTensor sizes too large for THCDeviceTensor conversion",https://github.com/pytorch/pytorch/issues/24401,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-04-18T22:38:17Z +issue,27336,Allow one inferred axis in torch.split,https://github.com/pytorch/pytorch/issues/27336,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-18T08:18:55Z +issue,11213,Ubuntu 16.04 setup.py error - undefined reference to elfLink_Get_FatBinary_From_Object' /usr/lib/x86_64-linux-gnu/libcuda.so: undefined reference to elf32_section_header',https://github.com/pytorch/pytorch/issues/11213,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: undefined reference",scan when time allows,2021-04-17T09:33:09Z +issue,54871,Not possible to save dataloader in C++,https://github.com/pytorch/pytorch/issues/54871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-04-17T09:31:50Z +issue,56237,optim.Adadelta: local variable 'lr' referenced before assignment,https://github.com/pytorch/pytorch/issues/56237,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-04-17T09:13:31Z +issue,12672,Move collate_fn functionality / responsibility into Dataset object,https://github.com/pytorch/pytorch/issues/12672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-04-16T20:17:02Z +issue,12690,How can I build caffe2_gtest_main under pytorch/caffe2/test/ folder?,https://github.com/pytorch/pytorch/issues/12690,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-16T20:14:06Z +issue,12675,[feature request] ignore_index and size_average in nn.AdaptiveLogSoftMaxWithLoss,https://github.com/pytorch/pytorch/issues/12675,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-04-16T20:07:23Z +issue,55933,[NNC] Replace ComputeAt,https://github.com/pytorch/pytorch/issues/55933,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-16T20:01:17Z +issue,56285,qnnpack uses deprecated pthreadpool APIs,https://github.com/pytorch/pytorch/issues/56285,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-16T18:36:48Z +issue,56030,Add Dirichlet Multinomial to PyTorch Distributions,https://github.com/pytorch/pytorch/issues/56030,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-04-16T18:01:27Z +issue,55757,Synchronize RRef.to_here() CUDA Streams properly when the profiler is enabled,https://github.com/pytorch/pytorch/issues/55757,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2021-04-16T17:08:38Z +issue,55941,[Tracking] Remove unneeded BC duplicates from native_functions.yaml,https://github.com/pytorch/pytorch/issues/55941,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-16T16:02:56Z +issue,56263,Reflect padding_mode should be supported for Conv3d,https://github.com/pytorch/pytorch/issues/56263,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: docs",scan when time allows,2021-04-16T13:15:14Z +issue,56246,Silent incorrect running with zero padding for Conv1d,https://github.com/pytorch/pytorch/issues/56246,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: error checking, module: padding",scan when time allows,2021-04-16T13:05:17Z +issue,56231,Half precision support for torch.sparse.mm,https://github.com/pytorch/pytorch/issues/56231,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2021-04-16T12:55:36Z +issue,56224,Error message regarding Padding of Conv2d needs improving,https://github.com/pytorch/pytorch/issues/56224,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: padding",scan when time allows,2021-04-16T12:54:03Z +issue,56177,torch.gradient not throwing error when spacing to be equal to 0,https://github.com/pytorch/pytorch/issues/56177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-04-16T12:46:05Z +issue,27094,[jit] Document what types can be traced,https://github.com/pytorch/pytorch/issues/27094,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-04-15T21:19:33Z +issue,26744,[jit] Default args don't work with TorchScript classes,https://github.com/pytorch/pytorch/issues/26744,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-04-15T21:19:27Z +issue,52694,[TensorExpr] Loop Transformations: `compute_inline` should use `Let` for inlined expressions,https://github.com/pytorch/pytorch/issues/52694,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-15T20:40:50Z +issue,29378,Support out= parameters with autograd,https://github.com/pytorch/pytorch/issues/29378,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2021-04-15T17:00:06Z +issue,56090,Converting float16->bool causes an internal compiler error with llvm,https://github.com/pytorch/pytorch/issues/56090,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-15T16:54:08Z +issue,50332,torch.Tensor.random_ is divergent from NumPy's np.random.random,https://github.com/pytorch/pytorch/issues/50332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy, module: random",scan when time allows,2021-04-15T16:34:55Z +issue,39443,PyTorch multiprocessing.spawn seems slow with list of tensors,https://github.com/pytorch/pytorch/issues/39443,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: performance",scan when time allows,2021-04-15T11:01:27Z +issue,56064,Optionally include padding_idx items in the EmbeddingBag reduction,https://github.com/pytorch/pytorch/issues/56064,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: nn",scan when time allows,2021-04-15T10:29:05Z +issue,56067,A more flexible torch.hub search strategy,https://github.com/pytorch/pytorch/issues/56067,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2021-04-15T10:28:24Z +issue,12248,.cuda() changes a module's behavior when there are registered buffers with requires_grad=True,https://github.com/pytorch/pytorch/issues/12248,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nn",scan when time allows,2021-04-14T19:56:35Z +issue,56012,Failure in complex CUDA numerics tests for sigmoid,https://github.com/pytorch/pytorch/issues/56012,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2021-04-14T17:55:06Z +issue,56023,[Opinfo] Better ErrorMsg for test_out with wrong shape,https://github.com/pytorch/pytorch/issues/56023,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-04-14T17:48:28Z +issue,52310,Implementation of many complex functions is fast but inaccurate in libc++,https://github.com/pytorch/pytorch/issues/52310,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numerical-stability, module: numpy",scan when time allows,2021-04-14T15:48:51Z +issue,55756,Cuda OOM after several steps for DataParallel model,https://github.com/pytorch/pytorch/issues/55756,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2021-04-14T08:39:06Z +issue,55356,torch.allclose does not allow different types for comparison,https://github.com/pytorch/pytorch/issues/55356,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: type promotion",scan when time allows,2021-04-14T07:51:22Z +issue,55869,We should perhaps prevent linking against LibTorch DLLs with the wrong configuration (debug vs release),https://github.com/pytorch/pytorch/issues/55869,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2021-04-14T07:16:39Z +issue,52206,from_numpy function rounds float values on Jetson NX,https://github.com/pytorch/pytorch/issues/52206,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: correctness (silent), module: jetson",scan when time allows,2021-04-14T06:30:24Z +issue,55944,Sparse-sparse matrix multiplication only works with torch.sparse.mm(),https://github.com/pytorch/pytorch/issues/55944,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sparse, module: ux",scan when time allows,2021-04-14T05:44:16Z +issue,55804,Add torch.nn.Conv2D correctness test,https://github.com/pytorch/pytorch/issues/55804,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: tests",scan when time allows,2021-04-14T04:59:10Z +issue,54800,[package] Detection of dependencies introduced by torch.ops.load_library,https://github.com/pytorch/pytorch/issues/54800,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-14T00:34:12Z +issue,55784,pybind11 Tensor type caster forces reference count bump,https://github.com/pytorch/pytorch/issues/55784,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2021-04-14T00:23:47Z +issue,55643,at::globalContext().hasCUDA()'s forked tongue tells lies,https://github.com/pytorch/pytorch/issues/55643,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: multiprocessing",scan when time allows,2021-04-13T22:27:47Z +issue,55964,Finish deprecating torch.range,https://github.com/pytorch/pytorch/issues/55964,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deprecation,scan when time allows,2021-04-13T22:17:40Z +issue,55453,C++ sparse_coo_tensor ignores TensorOptions argument,https://github.com/pytorch/pytorch/issues/55453,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sparse, module: tensor creation",scan when time allows,2021-04-13T21:50:12Z +issue,55876,[rfc] Trigger callback when backwards begins for DDP with custom autograd function,https://github.com/pytorch/pytorch/issues/55876,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-04-13T20:43:27Z +issue,55951,Update linspace and logspace to throw an error when steps is not provided,https://github.com/pytorch/pytorch/issues/55951,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: tensor creation",scan when time allows,2021-04-13T20:40:43Z +issue,55095,Python dispatch key and backend fallback,https://github.com/pytorch/pytorch/issues/55095,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__,scan when time allows,2021-04-13T19:17:20Z +issue,54622,torch.pow returns incorrect value for 0^0j,https://github.com/pytorch/pytorch/issues/54622,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy",scan when time allows,2021-04-13T17:47:13Z +issue,55331,Tests should be runnable without run_test.py,https://github.com/pytorch/pytorch/issues/55331,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: testing,scan when time allows,2021-04-13T15:24:28Z +issue,12609,Request for stripped down / inference only pytorch wheels,https://github.com/pytorch/pytorch/issues/12609,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-13T14:49:15Z +issue,41177,The new version of the libtorch become slow,https://github.com/pytorch/pytorch/issues/41177,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: performance, oncall: jit",scan when time allows,2021-04-13T12:10:43Z +issue,12460,"input_device, output_device, devices_used properties",https://github.com/pytorch/pytorch/issues/12460,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-04-13T01:23:27Z +issue,12482,[Feature request]: add `LayerNormLSTMCell`,https://github.com/pytorch/pytorch/issues/12482,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2021-04-13T01:12:08Z +issue,55751,[docs] algolia docsearch in pytorch docs,https://github.com/pytorch/pytorch/issues/55751,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2021-04-13T00:58:00Z +issue,55764,clang format on OS X ssl verification failure,https://github.com/pytorch/pytorch/issues/55764,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2021-04-13T00:55:30Z +issue,55450,[DOC] Incorrect image for the confusion matrix,https://github.com/pytorch/pytorch/issues/55450,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rnn",scan when time allows,2021-04-12T23:27:12Z +issue,55286,endpoint=False for torch.linspace and torch.logspace,https://github.com/pytorch/pytorch/issues/55286,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: tensor creation",scan when time allows,2021-04-12T23:22:42Z +issue,55755,"Some float16 inputs to CPU matmul are supported, others aren't",https://github.com/pytorch/pytorch/issues/55755,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: linear algebra",scan when time allows,2021-04-12T07:27:44Z +issue,54588,Make searchsorted and bucketize API consistent,https://github.com/pytorch/pytorch/issues/54588,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: docs, module: numpy, module: ux",scan when time allows,2021-04-11T08:32:16Z +issue,54766,Add `start_dim` and `end_dim` functionality for common reduction operations.,https://github.com/pytorch/pytorch/issues/54766,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2021-04-09T21:12:28Z +issue,55629,torch.arctanh not implemented for torch.float16,https://github.com/pytorch/pytorch/issues/55629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2021-04-09T19:45:40Z +issue,55696,[TensorExpr] IRSimplifier fails to optimize useless stores,https://github.com/pytorch/pytorch/issues/55696,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-09T19:31:49Z +issue,55611,ATen/native/sparse headers not in LibTorch or pypi package,https://github.com/pytorch/pytorch/issues/55611,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2021-04-08T22:51:19Z +issue,55585,"[amp scaler] unable to prevent ""scheduler before optimizer step"" warning",https://github.com/pytorch/pytorch/issues/55585,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: optimizer",scan when time allows,2021-04-08T22:49:57Z +issue,55583,Improve test runtime of distributed tests.,https://github.com/pytorch/pytorch/issues/55583,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",scan when time allows,2021-04-08T20:59:18Z +issue,51933,Performance bugs of transpose2d on A100,https://github.com/pytorch/pytorch/issues/51933,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: performance",scan when time allows,2021-04-08T17:55:43Z +issue,55557,Gradcheck failure for `torchaudio.functional.phase_vocoder`,https://github.com/pytorch/pytorch/issues/55557,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: correctness (silent)",scan when time allows,2021-04-08T14:51:00Z +issue,35290,Why C++ version libtorch so slow,https://github.com/pytorch/pytorch/issues/35290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-04-08T14:33:14Z +issue,55571,Update TCPStore.wait() error message to be more friendly,https://github.com/pytorch/pytorch/issues/55571,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-04-07T23:32:39Z +issue,55440,Have the ability to disable `__torch_function__` dispatch for torch.nn.functional functions,https://github.com/pytorch/pytorch/issues/55440,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: __torch_function__, module: nn",scan when time allows,2021-04-07T20:03:52Z +issue,53263,Different gradients in torch.matmul depending on input shape,https://github.com/pytorch/pytorch/issues/53263,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2021-04-07T19:47:14Z +issue,4958,[Feature Request] Extract glimpses from a batch of images (as in tf.image.extract_glimpse),https://github.com/pytorch/pytorch/issues/4958,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2021-04-07T18:35:51Z +issue,55329,Split test/cpp_extensions/setup.py per orthogonal extension module and/or do Python build from ninja too,https://github.com/pytorch/pytorch/issues/55329,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp-extensions",scan when time allows,2021-04-07T16:29:53Z +issue,55330,CI not surfacing some failures (-Werror?) on PR,https://github.com/pytorch/pytorch/issues/55330,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2021-04-07T16:19:39Z +issue,55448,"Errors occur when operating broadcast with error log: Got completion with error 12, opcode 1, len 32547, vendor err 129",https://github.com/pytorch/pytorch/issues/55448,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-04-07T16:13:57Z +issue,55461,Cross-compiling for cortexa9 processor,https://github.com/pytorch/pytorch/issues/55461,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-04-07T16:03:48Z +issue,55304,nn.DataParallel should raise error when provided with list of tensors,https://github.com/pytorch/pytorch/issues/55304,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-04-07T14:59:38Z +issue,55297,SIGSEGV at at::is_vulkan_available() invocation on Android,https://github.com/pytorch/pytorch/issues/55297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: android,scan when time allows,2021-04-07T14:57:32Z +issue,55274,[Feature Request] PowerNorm,https://github.com/pytorch/pytorch/issues/55274,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-04-07T14:54:13Z +issue,55260,Cuda memory leak check is somehow unstable with repeat_test_for_types,https://github.com/pytorch/pytorch/issues/55260,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: testing",scan when time allows,2021-04-07T14:52:49Z +issue,37002,[RFC] Modularize DistributedDataParallel,https://github.com/pytorch/pytorch/issues/37002,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-04-07T08:39:41Z +issue,55122,Cusolver handle may decrease MAGMA performance on GPU,https://github.com/pytorch/pytorch/issues/55122,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: magma, module: performance",scan when time allows,2021-04-07T00:34:26Z +issue,5790,Add hookable weights,https://github.com/pytorch/pytorch/issues/5790,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-04-05T21:09:40Z +issue,28761,torch.tensor() is very slow when it is passed an h5py Dataset.,https://github.com/pytorch/pytorch/issues/28761,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: tensor creation",scan when time allows,2021-04-05T17:27:21Z +issue,55299,[feature] `torch.polygamma` : Support Tensor for argument `n`,https://github.com/pytorch/pytorch/issues/55299,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: special,scan when time allows,2021-04-05T16:59:04Z +issue,55030,Add Minimal Gated Unit (MGU),https://github.com/pytorch/pytorch/issues/55030,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-04-05T14:47:42Z +issue,54938,[tests] `cumprod` OpInfo tests take long time to run (around 1min),https://github.com/pytorch/pytorch/issues/54938,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",scan when time allows,2021-04-05T07:36:12Z +issue,54774,"RuntimeError: ""log2"" ""_vml_cpu"" not implemented for 'Half'",https://github.com/pytorch/pytorch/issues/54774,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: numpy",scan when time allows,2021-04-05T07:29:25Z +issue,55174,Isolate CPU tests from GPU tests,https://github.com/pytorch/pytorch/issues/55174,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-04-03T21:57:41Z +issue,55135,LR scheduler step() behaviour with and without epoch parameter,https://github.com/pytorch/pytorch/issues/55135,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-04-02T17:37:37Z +issue,55161,Internal assert failed: `iter.device(arg).is_cuda()`,https://github.com/pytorch/pytorch/issues/55161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2021-04-02T06:44:57Z +issue,25979,[C++] `Module::pretty_print` is broken,https://github.com/pytorch/pytorch/issues/25979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2021-04-02T02:33:55Z +issue,55156,multiprocessing function cannot pass cuda objects use when calling inside from a DDP process,https://github.com/pytorch/pytorch/issues/55156,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-04-01T23:40:32Z +issue,55143,Add suport to torch.gather for negative indices,https://github.com/pytorch/pytorch/issues/55143,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2021-04-01T21:34:17Z +issue,55104,Alternative backend fallback like mechanism which has higher precedence than operator-specific composite implementations,https://github.com/pytorch/pytorch/issues/55104,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-04-01T21:30:22Z +issue,55120,Why isn't pip finding the correct pytorch dependency?,https://github.com/pytorch/pytorch/issues/55120,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2021-04-01T21:26:46Z +issue,54030,[FR] Configurable gain value for kaiming_normal_ / kaiming_uniform_ in torch.nn.init,https://github.com/pytorch/pytorch/issues/54030,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: initialization, module: nn",scan when time allows,2021-04-01T17:54:01Z +issue,54055,Wrong initialisation gain for SNNs/SELU,https://github.com/pytorch/pytorch/issues/54055,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: initialization, module: nn",scan when time allows,2021-04-01T14:14:04Z +issue,54942,Optimizations to TORCH_CHECK change inlining behavior.,https://github.com/pytorch/pytorch/issues/54942,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: performance",scan when time allows,2021-04-01T04:20:10Z +issue,54874,Get a thread safe copy of torch::nn::Sequential object,https://github.com/pytorch/pytorch/issues/54874,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2021-03-31T21:07:47Z +issue,54104,TorchVision and TorchAudio wheels for AArch64 absent from https://download.pytorch.org/whl/torch_stable.html,https://github.com/pytorch/pytorch/issues/54104,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: vision",scan when time allows,2021-03-31T19:56:55Z +issue,55083,[JIT][Testing] Split up test_jit.py and test_jit_py3.jit,https://github.com/pytorch/pytorch/issues/55083,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-03-31T18:44:42Z +issue,55045,16 bit accumulation for Convolution,https://github.com/pytorch/pytorch/issues/55045,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2021-03-31T17:51:02Z +issue,12060,cuda test hangs if GPUs in Exclusive Process mode,https://github.com/pytorch/pytorch/issues/12060,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-03-31T17:47:48Z +issue,12069,Stop using make_intrusive directly; provide some make_tensor,https://github.com/pytorch/pytorch/issues/12069,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-03-31T17:47:17Z +issue,12159,clip_grad_norm_ does not work on grads of different types,https://github.com/pytorch/pytorch/issues/12159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-03-31T17:43:22Z +issue,12189,[Feature] Support Adaptive Max Gradient Norm / Clipping,https://github.com/pytorch/pytorch/issues/12189,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2021-03-31T17:42:15Z +issue,54564,why single pytorch process is displayed on each GPUs with memory 0MB,https://github.com/pytorch/pytorch/issues/54564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-03-31T13:50:13Z +issue,54983,Accept objects with `__float__` wherever regular `float`s are accepted,https://github.com/pytorch/pytorch/issues/54983,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: half,scan when time allows,2021-03-31T05:00:52Z +issue,54897,Default generated meta functions for inplace operations don't report errors,https://github.com/pytorch/pytorch/issues/54897,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-31T04:12:30Z +issue,54947,Results surprisingly differ on different backends,https://github.com/pytorch/pytorch/issues/54947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility,scan when time allows,2021-03-30T23:39:25Z +issue,54843,Inconsistent behaviour with `weight` argument in CrossEntropyLoss and BCEWithLogitsLoss,https://github.com/pytorch/pytorch/issues/54843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2021-03-30T23:21:31Z +issue,55005,[TensorExpr] Provide a safe API to request the N-th loop for a given Buf.,https://github.com/pytorch/pytorch/issues/55005,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2021-03-30T22:58:10Z +issue,54861,Occured error in loss.backward() when using sparse=True in Embedding layer,https://github.com/pytorch/pytorch/issues/54861,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: embedding, module: sparse",scan when time allows,2021-03-30T21:49:20Z +issue,54815,TensorExpr `LoopNest.get_loops_for` misbehaved after loop distribution transformation,https://github.com/pytorch/pytorch/issues/54815,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-03-30T21:32:52Z +issue,54829,c++ libtorch config mismatch with python version,https://github.com/pytorch/pytorch/issues/54829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-03-30T16:15:10Z +issue,54802,[nnc] simplify is sometimes broken after computeAt,https://github.com/pytorch/pytorch/issues/54802,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-30T15:56:23Z +issue,54943,Structured kernels have increased TensorIterator overhead.,https://github.com/pytorch/pytorch/issues/54943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: performance",scan when time allows,2021-03-30T07:01:00Z +issue,54677,[docs] use sphinx link/reference checking features,https://github.com/pytorch/pytorch/issues/54677,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2021-03-30T05:01:26Z +issue,54823,"Fixing pytorch distributed training (DDP) code send a SIGKILL signal on its own on most version of pytorch 1.6.0, 1.7.0, 1.7.1?",https://github.com/pytorch/pytorch/issues/54823,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-03-29T20:24:26Z +issue,54614,Dispatcher TLS to bypass loads of dispatch key from tensor arguments,https://github.com/pytorch/pytorch/issues/54614,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: performance",scan when time allows,2021-03-29T15:51:32Z +issue,46321,Add distributed examples into PyTorch CI tests,https://github.com/pytorch/pytorch/issues/46321,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: tests, oncall: distributed",scan when time allows,2021-03-29T14:16:03Z +issue,54753,torch.sqrt for negative values should either return complex tensors or clearly throw a domain error/warning,https://github.com/pytorch/pytorch/issues/54753,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy",scan when time allows,2021-03-29T10:39:56Z +issue,54713,Test test_large_cumprod_cuda_float16 gets killed due to (probably) OOM,https://github.com/pytorch/pytorch/issues/54713,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: tests",scan when time allows,2021-03-29T02:49:55Z +issue,52552,Cumulative integration?,https://github.com/pytorch/pytorch/issues/52552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-03-28T11:03:27Z +issue,54841,[testing] test_reference_numerics_extremal_clamp_cpu_bfloat16 fails on ci build with GCC5.4 and Python 3.6,https://github.com/pytorch/pytorch/issues/54841,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: tests",scan when time allows,2021-03-28T06:30:06Z +issue,49726,Wrong error is raised for property of nn.Module (again),https://github.com/pytorch/pytorch/issues/49726,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-03-27T23:40:33Z +issue,53707,[RFC] Add common device runtime abstraction,https://github.com/pytorch/pytorch/issues/53707,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-27T10:06:47Z +issue,54757,[nnc] computeAt and index flatten broken in presence of if statements,https://github.com/pytorch/pytorch/issues/54757,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-26T22:20:15Z +issue,11980,[Enhancement] Increase user-friendliness of dataset.random_split,https://github.com/pytorch/pytorch/issues/11980,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2021-03-26T18:00:48Z +issue,11974,Specify out= argument to convolution,https://github.com/pytorch/pytorch/issues/11974,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2021-03-26T16:40:13Z +issue,3877,Wrap Cephes library for mathematical special functions,https://github.com/pytorch/pytorch/issues/3877,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-03-26T04:57:15Z +issue,53642,torch.unsafe_chunk and torch.unsafe_split 's documentation does not render on the website.,https://github.com/pytorch/pytorch/issues/53642,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: viewing and reshaping",scan when time allows,2021-03-25T18:58:30Z +issue,45996,Backpropagation for sparse matrix indexing is problematic (colab provided),https://github.com/pytorch/pytorch/issues/45996,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2021-03-25T07:52:32Z +issue,54629,"""Skipped!"" tests should have more descriptive skipped reason",https://github.com/pytorch/pytorch/issues/54629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-03-24T21:52:35Z +issue,54620,"torch.pow(tensor, tensor) throws RuntimeError for dtype bool",https://github.com/pytorch/pytorch/issues/54620,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-03-24T20:36:30Z +issue,54566,Support docstring definition in torch/library.h operator definition,https://github.com/pytorch/pytorch/issues/54566,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-24T20:34:25Z +issue,53957,Is pytorch 1.8.0 incompatible with cuda 11.2 or what is the reason for this error?,https://github.com/pytorch/pytorch/issues/53957,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-03-24T14:13:31Z +issue,54546,Let backends specify a schema version when registering kernels,https://github.com/pytorch/pytorch/issues/54546,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: backend, module: codegen, module: internals",scan when time allows,2021-03-23T21:44:03Z +issue,52940,[JIT] Provide a way to access a flattened optimized graph from Profiling Executor,https://github.com/pytorch/pytorch/issues/52940,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",scan when time allows,2021-03-23T21:23:54Z +issue,54407,test_clamp fails on ppc64le,https://github.com/pytorch/pytorch/issues/54407,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: power, oncall: jit",scan when time allows,2021-03-23T17:27:33Z +issue,54445,add tests to cover DDP on mixed dense-sparse models,https://github.com/pytorch/pytorch/issues/54445,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2021-03-23T02:01:56Z +issue,54283,Split VariableTypeManual.cpp,https://github.com/pytorch/pytorch/issues/54283,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-03-22T23:33:21Z +issue,54375,[proposal] Integrate autograd graph visualizer for displaying crashed locations during detect_anomaly,https://github.com/pytorch/pytorch/issues/54375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-03-22T22:38:43Z +issue,11786,Mysterious error due to num_workers: 1,https://github.com/pytorch/pytorch/issues/11786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2021-03-22T21:01:21Z +issue,11790,set num_workers on the dataloader make the jupyter kernel crash at the almost end of the epoch,https://github.com/pytorch/pytorch/issues/11790,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-03-22T21:00:06Z +issue,11854,error: ‘array_size’ is not a class template,https://github.com/pytorch/pytorch/issues/11854,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2021-03-22T20:53:04Z +issue,53623,test/distributed/test_c10d.py::DistributedDataParallelTest only passes if run before other test cases,https://github.com/pytorch/pytorch/issues/53623,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-03-22T18:01:44Z +issue,54138,Support Array Interface (__array_interface__ attribute),https://github.com/pytorch/pytorch/issues/54138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-03-22T17:28:31Z +issue,54387,[docs] torch.lerp has unmatching signature and explained parameter names,https://github.com/pytorch/pytorch/issues/54387,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: interpolation",scan when time allows,2021-03-22T17:24:05Z +issue,54368,RMSProp documentation is confusing,https://github.com/pytorch/pytorch/issues/54368,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2021-03-22T16:30:11Z +issue,54356,Generate delegates for all non-structured kernels,https://github.com/pytorch/pytorch/issues/54356,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-03-22T16:27:58Z +issue,54351,Numpy.float64 vs native python float breaks DDP,https://github.com/pytorch/pytorch/issues/54351,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ddp,scan when time allows,2021-03-22T16:26:31Z +issue,54213,"[docs] torch.clamp / torch.clip do support None as placeholder of min/max, but this is not documented",https://github.com/pytorch/pytorch/issues/54213,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-03-22T16:23:03Z +issue,54340,RuntimeError: polar does not support automatic differentiation for outputs with complex dtype.,https://github.com/pytorch/pytorch/issues/54340,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-22T16:22:24Z +issue,47175,[feature request] Show warning if optimizer.zero_grad() was not called before optimizer.step(),https://github.com/pytorch/pytorch/issues/47175,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer, module: ux",scan when time allows,2021-03-22T02:41:42Z +issue,54307,cuDNN error upon the backward method with a sliced tensor output from nn.GRU.,https://github.com/pytorch/pytorch/issues/54307,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: rnn",scan when time allows,2021-03-20T02:36:25Z +issue,54366,[package] Heuristics for extern-ing common dependencies,https://github.com/pytorch/pytorch/issues/54366,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-20T01:06:46Z +issue,54362,[package] Buck integration for dependency analysis,https://github.com/pytorch/pytorch/issues/54362,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-20T00:54:21Z +issue,54364,[package] Clear and minimal interface for what a package provides,https://github.com/pytorch/pytorch/issues/54364,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-20T00:52:50Z +issue,11727,Semaphore leaks in dataloader,https://github.com/pytorch/pytorch/issues/11727,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-03-19T15:45:12Z +issue,54279,[Static Runtime] static_runtime_benchmark compile-time error,https://github.com/pytorch/pytorch/issues/54279,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-03-19T13:24:35Z +issue,54278,[Static Runtime] StaticRuntime.EmbeddingBag test case broken,https://github.com/pytorch/pytorch/issues/54278,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-03-19T13:23:59Z +issue,54269,SummaryWriter add_image() docs do not state that it is expecting images in a certain format,https://github.com/pytorch/pytorch/issues/54269,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",scan when time allows,2021-03-18T21:29:13Z +issue,54201,Provide half-away-from-zero rounding mode on Tensor::round,https://github.com/pytorch/pytorch/issues/54201,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-03-18T13:09:10Z +issue,54192,OSS Mobile Errors Point to fburls,https://github.com/pytorch/pytorch/issues/54192,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-18T12:50:42Z +issue,54149,OSS build doesn't hard fail on nodiscard,https://github.com/pytorch/pytorch/issues/54149,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2021-03-18T12:47:29Z +issue,54139,as_tensor returns CPU copy of a CUDA buffer object implementing CUDA Array Interface,https://github.com/pytorch/pytorch/issues/54139,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numba",scan when time allows,2021-03-18T01:56:36Z +issue,53124,Deprecate and remove torch.set_default_tensor_type,https://github.com/pytorch/pytorch/issues/53124,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: deprecation,scan when time allows,2021-03-18T00:09:28Z +issue,46187,Better support for operators that return (named) tuples of tensors,https://github.com/pytorch/pytorch/issues/46187,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy, module: ux",scan when time allows,2021-03-17T23:10:58Z +issue,49456,Impossible to run tests target with LibTorch as dependency inside a cocoapods.,https://github.com/pytorch/pytorch/issues/49456,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ios,scan when time allows,2021-03-17T20:53:42Z +issue,54116,torch.put is divergent from np.put,https://github.com/pytorch/pytorch/issues/54116,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-03-17T09:27:15Z +issue,54058,c++ convert from std::vector to c10::List>,https://github.com/pytorch/pytorch/issues/54058,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2021-03-17T06:21:43Z +issue,54062,Many advanced indexing operations have untested large tensor branches,https://github.com/pytorch/pytorch/issues/54062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: tests",scan when time allows,2021-03-17T05:49:47Z +issue,54077,[cuda extension] tutorial should mention device guard,https://github.com/pytorch/pytorch/issues/54077,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: docs",scan when time allows,2021-03-17T05:48:48Z +issue,54096,cdist skip for manual backward is kind of janky,https://github.com/pytorch/pytorch/issues/54096,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: distance functions",scan when time allows,2021-03-17T05:42:21Z +issue,22343,New Weight Scheduler Concept for Weight Decay,https://github.com/pytorch/pytorch/issues/22343,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-03-17T01:43:13Z +issue,37809,"Add Numpy-like ""order"" argument to reshape",https://github.com/pytorch/pytorch/issues/37809,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-03-17T00:10:30Z +issue,49440,[RFC] DataLoader architecture updates and TarDataset implementation,https://github.com/pytorch/pytorch/issues/49440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-03-16T02:05:24Z +issue,53982,[Feature Request] Optionally specify batch dimension in DataLoader and collate_fn,https://github.com/pytorch/pytorch/issues/53982,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-03-15T21:50:14Z +issue,54021,Support scatter/gather for NCCL backend,https://github.com/pytorch/pytorch/issues/54021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: c10d,scan when time allows,2021-03-15T20:11:08Z +issue,49481,rank-0 gpu consume too high memory,https://github.com/pytorch/pytorch/issues/49481,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-03-15T17:50:09Z +issue,53407,"torch.matmul doesn't handle zero-sized inputs in some cases, leading to batched grad failures",https://github.com/pytorch/pytorch/issues/53407,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: tests, module: vmap",scan when time allows,2021-03-15T17:05:34Z +issue,52465,[nnc][perf] Performance decrease with CPU fusion on `freeze(script(pytorch_mobilenet_v3))`,https://github.com/pytorch/pytorch/issues/52465,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-03-15T16:53:02Z +issue,53386,Default pytest is truncating stacktrace (somehow!!!),https://github.com/pytorch/pytorch/issues/53386,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-15T16:45:42Z +issue,11514,[feature request] - Allow sequences lengths to be 0 in PackSequence,https://github.com/pytorch/pytorch/issues/11514,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2021-03-15T16:02:14Z +issue,11537,"[CLEANUP] Context functions should return TypeExtendedInterface, not Type",https://github.com/pytorch/pytorch/issues/11537,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-15T15:48:32Z +issue,11516,at::Device makes it very easy to write buggy code,https://github.com/pytorch/pytorch/issues/11516,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-15T15:44:42Z +issue,11551,"CrossEntropyLoss, ignore_index does not prevent back-prop if the logits are -inf",https://github.com/pytorch/pytorch/issues/11551,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: loss, module: nn",scan when time allows,2021-03-15T15:40:11Z +issue,53983,"hope to support something like ""torch::manual_seed_for_mulit_thread""",https://github.com/pytorch/pytorch/issues/53983,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: random",scan when time allows,2021-03-15T14:19:32Z +issue,54001,[codegen] Change public C++ to accept out/mut arguments by const&,https://github.com/pytorch/pytorch/issues/54001,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-03-15T14:16:39Z +issue,30786,C++ / JIT Parity for ops in `torch/functional.py` and `torch/tensor.py`,https://github.com/pytorch/pytorch/issues/30786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-03-15T13:58:28Z +issue,11624,Add min mode to embedding bags,https://github.com/pytorch/pytorch/issues/11624,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-03-15T13:49:10Z +issue,11635,[Feature request] Advanced indexing in functions like `expand`,https://github.com/pytorch/pytorch/issues/11635,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2021-03-15T13:46:59Z +issue,11645,One GPU is more memory efficient than Multiple GPUs,https://github.com/pytorch/pytorch/issues/11645,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: multi-gpu",scan when time allows,2021-03-15T13:44:20Z +issue,53890,GoogleNet pretrained/scratch bug?,https://github.com/pytorch/pytorch/issues/53890,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2021-03-15T01:30:54Z +issue,42849,[RFC] DeepSpeed + PT Distributed Integration,https://github.com/pytorch/pytorch/issues/42849,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-03-13T22:44:24Z +issue,53900,Have the possibility to reduce a tensor with median on more than one specified dimension,https://github.com/pytorch/pytorch/issues/53900,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: reductions",scan when time allows,2021-03-13T10:55:47Z +issue,49252,Methods that solve systems of linear equations are memory-inefficient (batch-wise broadcasting),https://github.com/pytorch/pytorch/issues/49252,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: memory usage",scan when time allows,2021-03-13T07:47:37Z +issue,11439,High leverage TH operations to port to ATen,https://github.com/pytorch/pytorch/issues/11439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: porting, module: tensor creation",scan when time allows,2021-03-12T22:49:01Z +issue,11426,[Feature request] Intuitive error message when input to Linear is not cudarized,https://github.com/pytorch/pytorch/issues/11426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2021-03-12T22:13:25Z +issue,11389,[distributions] Torch distribution samplers slow on expanded parameters,https://github.com/pytorch/pytorch/issues/11389,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-03-12T22:10:32Z +issue,53910,There should be a CI build with OpenBLAS in addition to MKL,https://github.com/pytorch/pytorch/issues/53910,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2021-03-12T20:30:58Z +issue,52274,Port unary elementwise ops to structured kernels,https://github.com/pytorch/pytorch/issues/52274,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2021-03-12T16:18:58Z +issue,46822,C++ Optimizer: remove warning on Optimizer::size method,https://github.com/pytorch/pytorch/issues/46822,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-03-12T15:36:54Z +issue,53477,Show methods of the class in the right sidebar,https://github.com/pytorch/pytorch/issues/53477,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-03-12T10:51:05Z +issue,53866,Add comprehensive subgroup testing to torch.distributed,https://github.com/pytorch/pytorch/issues/53866,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: ddp, oncall: distributed",scan when time allows,2021-03-12T02:00:53Z +issue,53427,Backward compute time in DDP logging may not be accurate,https://github.com/pytorch/pytorch/issues/53427,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-03-11T21:37:58Z +issue,53666,torch.autograd.functional.jacobian/hessian allow tensor in addition to function,https://github.com/pytorch/pytorch/issues/53666,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-03-11T14:16:25Z +issue,53708,torch.load with Exception,https://github.com/pytorch/pytorch/issues/53708,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-03-11T05:56:37Z +issue,53584,How to delete Module from GPU? (libtorch C++),https://github.com/pytorch/pytorch/issues/53584,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cpp-extensions",scan when time allows,2021-03-11T03:11:09Z +issue,53491,Very slow backward speed when using gather with small-range indices,https://github.com/pytorch/pytorch/issues/53491,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-03-10T18:03:45Z +issue,53193,[nnc] Prevent user from scheduling after prepareForCodegen,https://github.com/pytorch/pytorch/issues/53193,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-10T18:00:06Z +issue,52767,[nnc] Grouped convolutions,https://github.com/pytorch/pytorch/issues/52767,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-10T17:53:50Z +issue,53463,[Feature Request] return type support for `torch.norm` and tensor iteration in for loop,https://github.com/pytorch/pytorch/issues/53463,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-10T17:51:26Z +issue,53710,Ninja recompiles certain unchanged source files in cpp extension since torch 1.8.0,https://github.com/pytorch/pytorch/issues/53710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-03-10T17:43:24Z +issue,53442,IRSimplifier performs unsafe transformation on mod: m*n%m -> 0,https://github.com/pytorch/pytorch/issues/53442,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-10T06:56:30Z +issue,47502,Hackable python-only autograd engine,https://github.com/pytorch/pytorch/issues/47502,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-03-10T01:44:27Z +issue,53622,Properly design manual_cpp_binding (make it less error prone),https://github.com/pytorch/pytorch/issues/53622,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: language binding",scan when time allows,2021-03-10T00:07:31Z +issue,53627,test_metal.py must skip when not compiled with metal support,https://github.com/pytorch/pytorch/issues/53627,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-03-10T00:03:25Z +issue,40034,Doing optimizing compilations in a separate thread,https://github.com/pytorch/pytorch/issues/40034,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-09T22:55:47Z +issue,53649,vmap gradgradcheck test fails for unfold operation,https://github.com/pytorch/pytorch/issues/53649,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: vmap",scan when time allows,2021-03-09T21:36:27Z +issue,50503,Huge file sizes for libtorch and dependencies?,https://github.com/pytorch/pytorch/issues/50503,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2021-03-09T14:30:44Z +issue,19177,8 tests in test_c10d fail when running all tests in one command,https://github.com/pytorch/pytorch/issues/19177,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",scan when time allows,2021-03-09T08:12:00Z +issue,46225,torch.mode when input has nans,https://github.com/pytorch/pytorch/issues/46225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nans and infs, module: numpy, module: reductions",scan when time allows,2021-03-08T16:58:18Z +issue,53257,Inconsistent `out=` behaviour on advanced indexing operations.,https://github.com/pytorch/pytorch/issues/53257,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: complex",scan when time allows,2021-03-08T15:47:03Z +issue,47378,torch.vmap giving INTERNAL ASSERT FAILED error,https://github.com/pytorch/pytorch/issues/47378,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vmap,scan when time allows,2021-03-08T15:33:18Z +issue,53337,Should we remind users not to use the dataset on GPU when it's the argument of DataLoader?,https://github.com/pytorch/pytorch/issues/53337,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-03-08T04:24:44Z +issue,52265,'TypeError: expected CPU (got CUDA)' when placing a CUDA tensor on a class that is inheriting torch.Tensor,https://github.com/pytorch/pytorch/issues/52265,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2021-03-07T22:14:12Z +issue,53405,When I ran the 'Get Started with Data Data Parallel' tutorial the code got 'RuntimeError '!There are no hints,https://github.com/pytorch/pytorch/issues/53405,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-03-05T21:27:41Z +issue,53391,torch.empty_strided doesn't test if strides are negative,https://github.com/pytorch/pytorch/issues/53391,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2021-03-05T19:38:58Z +issue,53360,[bug] torch.cumsum: functional and method variant promotes all ints to Long but inplace don't,https://github.com/pytorch/pytorch/issues/53360,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: type promotion",scan when time allows,2021-03-05T19:21:35Z +issue,53345,CUDA error: an illegal memory access was encountered when updating model weights using GradScaler,https://github.com/pytorch/pytorch/issues/53345,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: nn",scan when time allows,2021-03-05T18:01:05Z +issue,53090,'Tensor' object has no attribute 'astype',https://github.com/pytorch/pytorch/issues/53090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-03-05T10:34:42Z +issue,53284,Find a good namespace home for torch._assert_async,https://github.com/pytorch/pytorch/issues/53284,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: testing",scan when time allows,2021-03-05T10:32:13Z +issue,52745,Sort torch.topk() output following input index order,https://github.com/pytorch/pytorch/issues/52745,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-05T00:49:13Z +issue,53157,doxygen and pytorch documentation,https://github.com/pytorch/pytorch/issues/53157,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2021-03-04T20:46:12Z +issue,53256,`index_copy_` test fails on PyTorch/XLA,https://github.com/pytorch/pytorch/issues/53256,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, module: xla",scan when time allows,2021-03-04T18:10:57Z +issue,53185,[NIT] Unpickler persistent_load monkey patch should fall back to original persistent_load,https://github.com/pytorch/pytorch/issues/53185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2021-03-04T16:56:33Z +issue,53194,Relative performance of histc vs bincount,https://github.com/pytorch/pytorch/issues/53194,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: sorting and selection",scan when time allows,2021-03-04T16:56:00Z +issue,53130,Lint rule to forbid bare assert() in cuda,https://github.com/pytorch/pytorch/issues/53130,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: lint",scan when time allows,2021-03-04T16:44:21Z +issue,53141,Structured kernels for operators that don't have out= variants,https://github.com/pytorch/pytorch/issues/53141,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: structured kernels,scan when time allows,2021-03-04T16:43:42Z +issue,41337,Distributed tests fail with pytest,https://github.com/pytorch/pytorch/issues/41337,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-03-04T10:40:14Z +issue,41256,Add nn.functional.swish alias for nn.functional.silu,https://github.com/pytorch/pytorch/issues/41256,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-03-04T07:53:25Z +issue,52762,Basic feature missing,https://github.com/pytorch/pytorch/issues/52762,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-03-03T14:37:44Z +issue,52913,Create a standalone python execution viewer like livepython,https://github.com/pytorch/pytorch/issues/52913,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2021-03-03T11:52:54Z +issue,35534,Fail to load torch script model,https://github.com/pytorch/pytorch/issues/35534,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-03-03T07:22:57Z +issue,53094,group conv in amp too slower,https://github.com/pytorch/pytorch/issues/53094,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: performance",scan when time allows,2021-03-03T04:21:40Z +issue,52583,[TensorExpr] Add IR Verifier,https://github.com/pytorch/pytorch/issues/52583,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2021-03-02T17:40:41Z +issue,53072,[NNC] Simplify after each transformation,https://github.com/pytorch/pytorch/issues/53072,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-02T03:49:49Z +issue,11113,use_system_nccl flag does not work?,https://github.com/pytorch/pytorch/issues/11113,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-03-01T21:41:58Z +issue,11072,[...] operator for masked select does not broadcast anymore,https://github.com/pytorch/pytorch/issues/11072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2021-03-01T21:27:56Z +issue,11071,undefined reference to 'caffe2::Caffe2FlagsRegistry[abi:cxx11]()',https://github.com/pytorch/pytorch/issues/11071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-03-01T21:18:21Z +issue,11062,"non-shuffling data loaders can affect random states, thus the results of shuffling data loaders.",https://github.com/pytorch/pytorch/issues/11062,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-03-01T21:17:24Z +issue,10913,Cannot find operator schema for 'ATen' Caffe2 Ios,https://github.com/pytorch/pytorch/issues/10913,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-01T21:06:50Z +issue,52680,Add Quantized{CPU|CUDA} support to Structured Kernels,https://github.com/pytorch/pytorch/issues/52680,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: structured kernels",scan when time allows,2021-03-01T21:03:11Z +issue,53023,[codegen] Make it easier to codegen call to API,https://github.com/pytorch/pytorch/issues/53023,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-03-01T19:23:09Z +issue,52673,Add custom types to PyTorch,https://github.com/pytorch/pytorch/issues/52673,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-03-01T18:29:29Z +issue,52911,LR Scheduler load_state_dict does not properly update scaling,https://github.com/pytorch/pytorch/issues/52911,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-03-01T16:02:47Z +issue,39503,pytest suppresses stderr from Python startup by default,https://github.com/pytorch/pytorch/issues/39503,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: logging, module: tests",scan when time allows,2021-02-28T19:59:50Z +issue,14864,[discussion] Recommend a different file extension for models (.PTH is a special extension for Python),https://github.com/pytorch/pytorch/issues/14864,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-02-28T12:29:49Z +issue,32552,MultivariateNormal.rsample: use eigen-decomposition when Cholesky fails,https://github.com/pytorch/pytorch/issues/32552,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: linear algebra",scan when time allows,2021-02-26T16:17:59Z +issue,10719,size mismatch when trying to reconstruct predifined network,https://github.com/pytorch/pytorch/issues/10719,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2021-02-25T22:30:40Z +issue,52850,How to skip the images in a custom dataset and deal with None values?,https://github.com/pytorch/pytorch/issues/52850,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-02-25T22:04:56Z +issue,52800,Create CI test checker to ensure # of test ran is the same as # of test reported in xmlrunner,https://github.com/pytorch/pytorch/issues/52800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2021-02-25T21:45:40Z +issue,52845,[feature request] F.interpolate_as,https://github.com/pytorch/pytorch/issues/52845,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn",scan when time allows,2021-02-25T20:38:42Z +issue,52722,Creating torch tensor as a function of index value,https://github.com/pytorch/pytorch/issues/52722,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-02-25T17:48:05Z +issue,52005,Implementation of Normalized Euclidean Distance (NED) in pytorch,https://github.com/pytorch/pytorch/issues/52005,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: nn",scan when time allows,2021-02-25T17:44:22Z +issue,52211,Conv3D error : CUDNN_STATUS_INTERNAL_ERROR,https://github.com/pytorch/pytorch/issues/52211,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2021-02-25T05:10:18Z +issue,52688,[BE] Consolidating scripts/ and tools/ folder,https://github.com/pytorch/pytorch/issues/52688,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-02-25T02:05:56Z +issue,52712,Refactor `torch.linalg.norm`,https://github.com/pytorch/pytorch/issues/52712,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2021-02-25T02:03:27Z +issue,52717,Fix `torch.norm`'s backward to match forward for ord=+/-inf,https://github.com/pytorch/pytorch/issues/52717,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra",scan when time allows,2021-02-25T02:02:38Z +issue,52744,Gradient checkpointing support in C++ API,https://github.com/pytorch/pytorch/issues/52744,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-02-25T02:00:22Z +issue,52334,Normal_like operator,https://github.com/pytorch/pytorch/issues/52334,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-02-24T18:16:37Z +issue,3920,[docs] Tensor.new is not documented,https://github.com/pytorch/pytorch/issues/3920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-02-24T13:57:50Z +issue,52205,Allow F.pad(mode = 'reflect') when shape == pad,https://github.com/pytorch/pytorch/issues/52205,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: numpy, module: padding",scan when time allows,2021-02-24T04:07:46Z +issue,52578,version1.7.0 is ~1.3x slower than 1.4.0 on ResNet18,https://github.com/pytorch/pytorch/issues/52578,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-02-24T01:50:03Z +issue,52648,Fix `torch.linalg.vector_norm`'s backward to avoid needing `at::abs()` call in forward for ord=+/-inf,https://github.com/pytorch/pytorch/issues/52648,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2021-02-24T01:49:13Z +issue,51794,"optim.Optimizer should copy ""params"" before modifying them",https://github.com/pytorch/pytorch/issues/51794,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-02-24T01:08:54Z +issue,46886,Support deleting a parameter/buffer by name,https://github.com/pytorch/pytorch/issues/46886,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: ux",scan when time allows,2021-02-23T23:30:48Z +issue,22755,[RFC] InstanceNorm default affine value,https://github.com/pytorch/pytorch/issues/22755,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-02-23T10:24:59Z +issue,52168,CMake errors building project that uses PyTorch,https://github.com/pytorch/pytorch/issues/52168,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-23T08:49:47Z +issue,52605,Torchaudio 0.8.0.dev nightly requires torch 1.9.0.dev which is missing,https://github.com/pytorch/pytorch/issues/52605,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-23T06:36:42Z +issue,52644,[FR] torch.load's map_location supports int,https://github.com/pytorch/pytorch/issues/52644,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-02-23T02:25:29Z +issue,52598,CUDA error: no kernel image is available for execution on the device When convert cpu tensor to gpu tensor,https://github.com/pytorch/pytorch/issues/52598,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-02-23T01:11:25Z +issue,52570,"RTX 2080s performs better than RTX 3080 in Semantic Segmentation inference process(Libtorch,win10),why?",https://github.com/pytorch/pytorch/issues/52570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2021-02-22T23:07:27Z +issue,52597,Custom scatter function for DataParallel,https://github.com/pytorch/pytorch/issues/52597,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-02-22T22:06:27Z +issue,52569,Omitting mutability annotations in op schemas can lead to subtle errors,https://github.com/pytorch/pytorch/issues/52569,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-02-22T21:52:59Z +issue,51038,The pooling code is ignoring sliding windows that start within the right padding,https://github.com/pytorch/pytorch/issues/51038,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2021-02-22T19:26:19Z +issue,52208,C++ standard library confusion,https://github.com/pytorch/pytorch/issues/52208,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-22T18:31:27Z +issue,51849,.numpy() array failes to keep original storage around,https://github.com/pytorch/pytorch/issues/51849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: tensor creation",scan when time allows,2021-02-22T18:21:28Z +issue,48245,"RuntimeError: ""threshold_cpu"" not implemented for 'Half'",https://github.com/pytorch/pytorch/issues/48245,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: half",scan when time allows,2021-02-22T08:24:52Z +issue,4906,"Rebuild from no-CUDA to CUDA leads to: error: #error ""Expected GLOO_USE_CUDA to be defined""",https://github.com/pytorch/pytorch/issues/4906,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-21T18:34:36Z +issue,39634,Installation from source fails on macOS (No CUDA),https://github.com/pytorch/pytorch/issues/39634,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: macos",scan when time allows,2021-02-20T22:40:28Z +issue,52318,"TorchConfig.cmake entries of if(ON) are not accepted by CMake as ""true"", to be replaced by if(1)",https://github.com/pytorch/pytorch/issues/52318,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-20T13:27:45Z +issue,10683,Tensor.register_hook is not passing the tensor object to the hook function,https://github.com/pytorch/pytorch/issues/10683,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking,scan when time allows,2021-02-19T11:54:55Z +issue,10691,Build system doesn't prevent ATen/core from including non-core files,https://github.com/pytorch/pytorch/issues/10691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-19T11:50:02Z +issue,10685,[feature request] convtbc with group convolution,https://github.com/pytorch/pytorch/issues/10685,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2021-02-19T11:49:11Z +issue,52478,"Could you please increase the wheel compiled Linux (aarch64) GPU, thank you very much!",https://github.com/pytorch/pytorch/issues/52478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: arm,scan when time allows,2021-02-19T11:08:34Z +issue,50206,Build Fail with vulkan on ARM64 and wayland drivers,https://github.com/pytorch/pytorch/issues/50206,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-17T23:35:48Z +issue,52390,autograd.functional.vjp with function that mutates inputs in-place can cause confusion,https://github.com/pytorch/pytorch/issues/52390,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-02-17T22:38:40Z +issue,52298,[FR} Implement Skellam distribution,https://github.com/pytorch/pytorch/issues/52298,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-02-17T19:04:44Z +issue,44809,Simple functions shouldn't go through dispatcher,https://github.com/pytorch/pytorch/issues/44809,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2021-02-17T14:40:17Z +issue,52266,GPU (+CPU/RAM?) self-check script or instructions in core,https://github.com/pytorch/pytorch/issues/52266,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2021-02-17T14:29:00Z +issue,51622,"`fractional_max_pool{2, 3}d` inconsistent between CUDA and CPU",https://github.com/pytorch/pytorch/issues/51622,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2021-02-16T18:46:21Z +issue,10482,Reduce code duplication in interpolate and make it more generic,https://github.com/pytorch/pytorch/issues/10482,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: interpolation,scan when time allows,2021-02-16T01:21:48Z +issue,51777,Analytical metric package,https://github.com/pytorch/pytorch/issues/51777,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: distributions",scan when time allows,2021-02-15T20:16:57Z +issue,52011,CPU eval BatchNorm2d is not threaded,https://github.com/pytorch/pytorch/issues/52011,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization, module: performance",scan when time allows,2021-02-13T17:00:45Z +issue,32826,MKLDNN doesnt work and is slower than normal cpu mode,https://github.com/pytorch/pytorch/issues/32826,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mkldnn,scan when time allows,2021-02-13T13:32:40Z +issue,50360,error: conversion from ‘std::vector’ to non-scalar type ‘at::Tensor’ requested,https://github.com/pytorch/pytorch/issues/50360,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-02-12T13:57:32Z +issue,52096,"Unable to build: subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '4']' returned non-zero exit status 1.",https://github.com/pytorch/pytorch/issues/52096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-11T22:11:44Z +issue,52145,Trying to initialise CUDA twice in a process with no visible devices hangs the process and terminal permanently,https://github.com/pytorch/pytorch/issues/52145,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: deadlock",scan when time allows,2021-02-11T17:55:19Z +issue,52041,device (+dtype) arguments for torch.stack / torch.cat,https://github.com/pytorch/pytorch/issues/52041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2021-02-10T23:36:52Z +issue,51911,torch.nn.functional.grid_sample outputs NaN,https://github.com/pytorch/pytorch/issues/51911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nans and infs, module: nn",scan when time allows,2021-02-10T18:35:39Z +issue,51231,macOS error building 1.7.1 from source on Catalina 10.15.7,https://github.com/pytorch/pytorch/issues/51231,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2021-02-10T18:21:07Z +issue,51868,discrepancies between TorchConfigVersion.cmake to git branch,https://github.com/pytorch/pytorch/issues/51868,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-10T18:18:16Z +issue,43876,Insert at specific index/key in nn.Sequential,https://github.com/pytorch/pytorch/issues/43876,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-02-10T08:21:09Z +issue,40275,Conv3d with specific kernel size outputs inconsistent results between FP16 and FP32 in V100 GPU,https://github.com/pytorch/pytorch/issues/40275,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn, module: dependency bug",scan when time allows,2021-02-10T02:49:44Z +issue,45764,resize_(0) is very expensive,https://github.com/pytorch/pytorch/issues/45764,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-02-09T23:46:34Z +issue,49101,[RFC] RemoteTensor: Use Remote Devices as Local Devices,https://github.com/pytorch/pytorch/issues/49101,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2021-02-09T22:44:21Z +issue,44738,[tools.codegen] Remove byte-for-byte compatibility code,https://github.com/pytorch/pytorch/issues/44738,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-02-09T21:05:59Z +issue,51493,Add branch predictor hints to prefer `Context::deterministicAlgorithms() == false`,https://github.com/pytorch/pytorch/issues/51493,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: performance",scan when time allows,2021-02-09T20:29:38Z +issue,51117,TestLinalgCPU.test_cholesky_solve_autograd_cpu_float64 fails on different random seed,https://github.com/pytorch/pytorch/issues/51117,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: correctness (silent), module: linear algebra",scan when time allows,2021-02-09T20:26:21Z +issue,49426,[doc] missing torch.pdist entry,https://github.com/pytorch/pytorch/issues/49426,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: docs, module: nn",scan when time allows,2021-02-09T07:43:09Z +issue,51320,"Pytorch not working properly (I don't know how to summarize it, see below)",https://github.com/pytorch/pytorch/issues/51320,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-02-08T23:52:23Z +issue,10165,torch.utils.data.random_split() returns dataset index as tensor,https://github.com/pytorch/pytorch/issues/10165,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: docs",scan when time allows,2021-02-08T18:14:05Z +issue,51872,Tensor.nonzero tries to allocate huge amount of memory for tensors on GPU with num_elements close to INT_MAX,https://github.com/pytorch/pytorch/issues/51872,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dependency bug, module: memory usage",scan when time allows,2021-02-08T17:43:41Z +issue,10386,[feature request] batch_first option in torch.utils.data,https://github.com/pytorch/pytorch/issues/10386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-02-08T16:24:25Z +issue,51855,"Attempted to read a PyTorch file with version 3, but the maximum supported version for reading is 2. Your PyTorch installation may be too old. (init at /pytorch/caffe2/serialize/inline_container.cc:132)",https://github.com/pytorch/pytorch/issues/51855,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: assert failure,scan when time allows,2021-02-08T16:19:00Z +issue,48465,anchored direct links to functions disappeared in 1.7.0 docs,https://github.com/pytorch/pytorch/issues/48465,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2021-02-08T15:12:38Z +issue,41186,test_autograd failures on Power,https://github.com/pytorch/pytorch/issues/41186,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda, module: linear algebra, module: power",scan when time allows,2021-02-08T04:01:51Z +issue,50835,Insufficient shared memory (shm) while training,https://github.com/pytorch/pytorch/issues/50835,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2021-02-08T03:43:57Z +issue,51776,BatchNorm3d error : CUDNN_STATUS_NOT_SUPPORTED,https://github.com/pytorch/pytorch/issues/51776,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2021-02-07T00:07:44Z +issue,50382,out variant of many loss functions are not consistent with non-out variant when reduction is not none,https://github.com/pytorch/pytorch/issues/50382,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: loss, module: nn, module: reductions",scan when time allows,2021-02-05T21:12:29Z +issue,51781,Libtorch lock thread,https://github.com/pytorch/pytorch/issues/51781,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: abi,scan when time allows,2021-02-05T19:43:07Z +issue,39674,[BatchNorm] Add boolean flags to choose the stats for normalization,https://github.com/pytorch/pytorch/issues/39674,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2021-02-05T18:34:01Z +issue,51711,Weird behavior in Conv2d padding when changed to 'reflect',https://github.com/pytorch/pytorch/issues/51711,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: docs, module: nn, module: padding",scan when time allows,2021-02-05T17:26:38Z +issue,51663,Torch DataLoader fails to reraise error from sqlalchemy.,https://github.com/pytorch/pytorch/issues/51663,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-02-05T14:42:11Z +issue,49110,[RFC] Integrate DeepSpeed CUDA kernels,https://github.com/pytorch/pytorch/issues/49110,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-02-04T22:19:27Z +issue,51224,`torch.cuda.device` should look like it is inherited from `torch.device`,https://github.com/pytorch/pytorch/issues/51224,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-02-04T16:41:20Z +issue,51256,[search] duplicated entries in results,https://github.com/pytorch/pytorch/issues/51256,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-02-04T14:42:25Z +issue,31178,torch::nn::functional::interpolate crash,https://github.com/pytorch/pytorch/issues/31178,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2021-02-04T12:43:20Z +issue,40107,[RFC] Add a RPC context manager that collects/waits for all RPC futures created in scope,https://github.com/pytorch/pytorch/issues/40107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: rpc",scan when time allows,2021-02-04T02:53:46Z +issue,51197,"[docs] use `versionadded`, `versionchanged` and `deprecated` directive",https://github.com/pytorch/pytorch/issues/51197,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-02-03T20:46:54Z +issue,51471,Make TorchBind object appear in state_dict,https://github.com/pytorch/pytorch/issues/51471,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization",scan when time allows,2021-02-03T03:28:45Z +issue,51426,Different results on RTX 2060s vs GTX 1060 with gpt model,https://github.com/pytorch/pytorch/issues/51426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-reproducibility,scan when time allows,2021-02-02T23:15:34Z +issue,51451,Add support for conan package manager,https://github.com/pytorch/pytorch/issues/51451,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-02-02T18:25:32Z +issue,51418,Pytorch num_worker>0 code worked first time and then it never worked with same setting again,https://github.com/pytorch/pytorch/issues/51418,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-02-02T17:57:54Z +issue,51406,[numpy] `round` and `trunc` not supported for Integral Type while Python and NumPy supports them,https://github.com/pytorch/pytorch/issues/51406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-02-02T17:07:41Z +issue,51410,Caffe2 RNN Cell Names not rectified,https://github.com/pytorch/pytorch/issues/51410,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-02-02T17:03:15Z +issue,51407,Additon of levenberg-marquardt optimizer in TORCH.OPTIM,https://github.com/pytorch/pytorch/issues/51407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-02-02T17:02:21Z +issue,51074,error: ‘PyFrameObject {aka struct _frame}’ has no member named ‘f_lasti’,https://github.com/pytorch/pytorch/issues/51074,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-02-02T16:49:54Z +issue,28619,Channels Last (NHWC) support plan.,https://github.com/pytorch/pytorch/issues/28619,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2021-02-02T13:52:10Z +issue,51207,Initialize NCCL backend with MPI,https://github.com/pytorch/pytorch/issues/51207,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mpi, module: nccl, oncall: distributed",scan when time allows,2021-02-02T01:53:45Z +issue,51385,DataParallel copies the model onto GPUs sequentially,https://github.com/pytorch/pytorch/issues/51385,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: performance",scan when time allows,2021-02-01T18:52:56Z +issue,51440,Support unity 3d,https://github.com/pytorch/pytorch/issues/51440,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-02-01T18:28:49Z +issue,48485,Add wheels for all cuda versions on pypi,https://github.com/pytorch/pytorch/issues/48485,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2021-02-01T17:06:29Z +issue,50650,build from source fail in jeston tx2 Python3.8,https://github.com/pytorch/pytorch/issues/50650,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build",scan when time allows,2021-02-01T15:44:49Z +issue,51044,CuDNN 8 with benchmark=True takes minutes to execute for certain configurations,https://github.com/pytorch/pytorch/issues/51044,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: performance",scan when time allows,2021-01-30T22:31:29Z +issue,38193,Can't Build Pytorch0.4.1 From Source,https://github.com/pytorch/pytorch/issues/38193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-01-30T18:09:30Z +issue,37995,unsqueeze support for named tensors,https://github.com/pytorch/pytorch/issues/37995,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2021-01-30T18:02:59Z +issue,51393,max_pool2d CPU forward performance is poor,https://github.com/pytorch/pytorch/issues/51393,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: pooling",scan when time allows,2021-01-30T02:27:30Z +issue,49409,Add global gradcheck setting,https://github.com/pytorch/pytorch/issues/49409,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-01-30T00:08:21Z +issue,51302,"""make install"" skips some required libraries for static build",https://github.com/pytorch/pytorch/issues/51302,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-01-29T05:01:19Z +issue,51229,std::type_index doesn't provide reliable equality for getCustomClassTypeMap,https://github.com/pytorch/pytorch/issues/51229,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2021-01-29T05:00:35Z +issue,51282,Bind tensor.unflatten to torch.unflatten namespace (torch.flatten exists but not torch.unflatten),https://github.com/pytorch/pytorch/issues/51282,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-01-29T05:00:06Z +issue,32306,[FR] bincount along arbitrary dimension,https://github.com/pytorch/pytorch/issues/32306,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection,scan when time allows,2021-01-29T01:12:15Z +issue,43867,torch.bincount beyond 1d arrays,https://github.com/pytorch/pytorch/issues/43867,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-01-29T00:29:47Z +issue,41232,Cleanup git branches,https://github.com/pytorch/pytorch/issues/41232,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-01-28T16:13:58Z +issue,50829,pin_memory leads to GPU memory being used but it doesn't initialize CUDA,https://github.com/pytorch/pytorch/issues/50829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2021-01-28T10:34:28Z +issue,51003,Proposing new features: Developing Echo State Network layers,https://github.com/pytorch/pytorch/issues/51003,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: embedding,scan when time allows,2021-01-27T02:45:28Z +issue,17902,Conjugate gradient method,https://github.com/pytorch/pytorch/issues/17902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: derivatives,scan when time allows,2021-01-27T00:17:31Z +issue,9882,Unintuitive reduction of mini-batch loss for NLLLoss,https://github.com/pytorch/pytorch/issues/9882,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: loss, module: nn",scan when time allows,2021-01-26T23:01:46Z +issue,50852,[TorchScript Performance Deep-dive] problems discovered in TS performance deep-dive,https://github.com/pytorch/pytorch/issues/50852,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2021-01-26T22:27:45Z +issue,51084,dyld: Symbol not found: _PyBaseObject_Type Referenced from,https://github.com/pytorch/pytorch/issues/51084,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos, module: static linking",scan when time allows,2021-01-26T22:20:33Z +issue,39365,Attributes not removed by freeze module,https://github.com/pytorch/pytorch/issues/39365,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-01-26T21:36:36Z +issue,50987,Scalar/Tensor arg type for op schemas,https://github.com/pytorch/pytorch/issues/50987,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2021-01-26T18:44:44Z +issue,9753,dataloader stuck at sched_yield =0,https://github.com/pytorch/pytorch/issues/9753,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-01-26T18:13:54Z +issue,9893,interaction with FindCUDA causes spurious re-cmakes,https://github.com/pytorch/pytorch/issues/9893,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-01-26T17:58:18Z +issue,9912,Stop ifdef'ing out scatter/gather (comm) in libtorch,https://github.com/pytorch/pytorch/issues/9912,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cuda",scan when time allows,2021-01-26T17:47:17Z +issue,9950,Stop passing inplace/out arguments as (non-const) Tensor& to functions,https://github.com/pytorch/pytorch/issues/9950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: internals",scan when time allows,2021-01-26T17:14:23Z +issue,9952,[feature request] Add COCOB Optimizer,https://github.com/pytorch/pytorch/issues/9952,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-01-26T17:10:27Z +issue,47669,Pytorch 1.7.0 with cuda 11.1.1 and cudnn 8.0.5,https://github.com/pytorch/pytorch/issues/47669,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: cuda, oncall: releng",scan when time allows,2021-01-26T13:39:18Z +issue,51054,Class-based structured kernels instruction count regression,https://github.com/pytorch/pytorch/issues/51054,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: performance, module: structured kernels",scan when time allows,2021-01-25T21:52:00Z +issue,51015,/usr/local/include/google/protobuf/stubs/pbconfig.h:3:37: note: expanded from macro 'GOOGLE_PROTOBUF_HASH_MAP_H',https://github.com/pytorch/pytorch/issues/51015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2021-01-25T15:48:07Z +issue,51008,Suggestion: Assimilate activation images to PyTorch brand color,https://github.com/pytorch/pytorch/issues/51008,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-01-25T15:40:26Z +issue,50925,THCCachingAllocator::cuda_free_mutex has no effect,https://github.com/pytorch/pytorch/issues/50925,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nccl",scan when time allows,2021-01-25T15:22:04Z +issue,50888,Does pytorch support cuda 11.1,https://github.com/pytorch/pytorch/issues/50888,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-01-25T14:47:04Z +issue,50219,`requires_grad_` in `no_grad` context returns incorrect value with tensor subclasses,https://github.com/pytorch/pytorch/issues/50219,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-01-25T14:25:52Z +issue,50890,"torch.unique(x, dim=1, return_inverse=True) returns inverse for only the last sub-tensor along dim",https://github.com/pytorch/pytorch/issues/50890,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy, module: sorting and selection",scan when time allows,2021-01-25T13:51:39Z +issue,50737,Adding visualizations for indexing and other functions,https://github.com/pytorch/pytorch/issues/50737,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing, module: docs, module: scatter & gather ops",scan when time allows,2021-01-25T09:52:58Z +issue,47109,Pytorch 1.7/Cuda 11.1 binaries,https://github.com/pytorch/pytorch/issues/47109,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda",scan when time allows,2021-01-23T19:09:33Z +issue,50939,dataloader bug need help,https://github.com/pytorch/pytorch/issues/50939,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-01-23T00:19:33Z +issue,50944,libpytorch macos build: static library eigen_blas_LIBRARY-NOTFOUND not found,https://github.com/pytorch/pytorch/issues/50944,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2021-01-22T17:53:47Z +issue,4731,Consider disallowing Variables that require grad in NCCL/comm functions,https://github.com/pytorch/pytorch/issues/4731,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: nccl",scan when time allows,2021-01-22T16:40:23Z +issue,35666,[discussion] Generic solutions for too-small-epsilon in FP16 training,https://github.com/pytorch/pytorch/issues/35666,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs, module: half",scan when time allows,2021-01-22T15:57:28Z +issue,35797,Backward function causes device error in C++ when changing module's device repeatly.,https://github.com/pytorch/pytorch/issues/35797,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpp",scan when time allows,2021-01-22T15:54:01Z +issue,40480,torch.autograd.functional.* for models,https://github.com/pytorch/pytorch/issues/40480,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-01-22T15:39:50Z +issue,20058,Implement noise_shape keyword for Dropout layers,https://github.com/pytorch/pytorch/issues/20058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2021-01-21T22:03:51Z +issue,43556,FP16 gives NaN loss when using pre-trained model,https://github.com/pytorch/pytorch/issues/43556,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda",scan when time allows,2021-01-21T10:00:51Z +issue,50831,Dataloader Prefetch data to GPU by cudaMemPrefetchAsync,https://github.com/pytorch/pytorch/issues/50831,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-01-21T04:12:49Z +issue,50303,Way to solve GPU Host Thread Contention?,https://github.com/pytorch/pytorch/issues/50303,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2021-01-21T00:10:13Z +issue,50830,Pull upstream cmake's change in FindCUDA/select_compute_arch,https://github.com/pytorch/pytorch/issues/50830,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-01-20T20:42:16Z +issue,49949,"RFC: Add a ""see also"" section to operators' documentation",https://github.com/pytorch/pytorch/issues/49949,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-01-20T14:27:24Z +issue,9681,[feature request] Support for 0-length sequences in packed_sequences,https://github.com/pytorch/pytorch/issues/9681,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2021-01-20T13:28:05Z +issue,50354,for CNN in fp16 execution time depends on input scale,https://github.com/pytorch/pytorch/issues/50354,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: half, module: nn, module: performance",scan when time allows,2021-01-19T15:23:44Z +issue,14945,[Feature request] create sparse coo matrix w/o index check,https://github.com/pytorch/pytorch/issues/14945,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: sparse",scan when time allows,2021-01-19T06:20:02Z +issue,50694,CosineAnnealingWarmRestarts lacks verbose functionality,https://github.com/pytorch/pytorch/issues/50694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-01-18T23:55:08Z +issue,50468,[RFC] Make RRef proxy APIs non-blocking,https://github.com/pytorch/pytorch/issues/50468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2021-01-18T11:01:58Z +issue,9559,[feature request] Rename `Subset` -> `Resample` to reflect wider use,https://github.com/pytorch/pytorch/issues/9559,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-01-18T02:01:54Z +issue,9560,Incorrect term in _LRScheduler.,https://github.com/pytorch/pytorch/issues/9560,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-01-18T02:00:38Z +issue,50616,Use unified type for distributions.constraint API,https://github.com/pytorch/pytorch/issues/50616,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2021-01-18T01:44:51Z +issue,46644,Trying to compile PyTorch for sm_30 fails with `error: identifier “__ldg” is undefined`,https://github.com/pytorch/pytorch/issues/46644,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-01-16T03:28:19Z +issue,50613,Discourage slow gradchecks,https://github.com/pytorch/pytorch/issues/50613,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: tests",scan when time allows,2021-01-15T21:43:07Z +issue,50473,Make use of eps for numerical stability more consistent,https://github.com/pytorch/pytorch/issues/50473,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability,scan when time allows,2021-01-15T02:44:31Z +issue,50552,test_fn_grad_fft_fftn_cpu_complex128 and test_fn_grad_fft_rfftn_cpu_float64 are failing under TSAN,https://github.com/pytorch/pytorch/issues/50552,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: fft, module: sanitizers, module: tests",scan when time allows,2021-01-14T21:13:06Z +issue,47627,Allow add_embedding to have dict for metadata,https://github.com/pytorch/pytorch/issues/47627,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2021-01-14T10:15:45Z +issue,49180,torch.utils.data.DistributedSampler allow uneven inputs,https://github.com/pytorch/pytorch/issues/49180,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data, module: dataloader",scan when time allows,2021-01-14T03:51:09Z +issue,49911,[RFC] Speed up python function and arg serialization in RPC APIs,https://github.com/pytorch/pytorch/issues/49911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2021-01-13T21:48:38Z +issue,48702,provide example for distributed training with iterative dataloaders,https://github.com/pytorch/pytorch/issues/48702,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",scan when time allows,2021-01-13T14:58:22Z +issue,39680,CUDA cannot be found,https://github.com/pytorch/pytorch/issues/39680,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: build, module: cuda",scan when time allows,2021-01-13T09:55:28Z +issue,50462,Standalone OSS RPC benchmark,https://github.com/pytorch/pytorch/issues/50462,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2021-01-13T07:00:58Z +issue,9381,[doc] many losses still mention size_average in formula,https://github.com/pytorch/pytorch/issues/9381,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: docs, module: loss, module: nn",scan when time allows,2021-01-13T00:17:37Z +issue,9417,"[feature request] allow forward_pre_hook to ""preprocess"" input",https://github.com/pytorch/pytorch/issues/9417,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-01-12T22:12:25Z +issue,48605,[NNC] Bugs Exposed in Binary Op Testing,https://github.com/pytorch/pytorch/issues/48605,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2021-01-12T20:49:15Z +issue,37766,CMake Documentation Issue,https://github.com/pytorch/pytorch/issues/37766,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-01-12T08:39:35Z +issue,47476,Reciprocal and reciprocal square root instructions are too inaccurate on ARM64,https://github.com/pytorch/pytorch/issues/47476,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: correctness (silent)",scan when time allows,2021-01-12T04:55:42Z +issue,39279,Differentiable Optimizers,https://github.com/pytorch/pytorch/issues/39279,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer",scan when time allows,2021-01-11T22:30:06Z +issue,35312,Method/constructor which takes as input angle and magnitude and returns a complex tensor,https://github.com/pytorch/pytorch/issues/35312,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2021-01-11T21:50:49Z +issue,50366,Optimizer Closure: Enable skip by returning None loss.,https://github.com/pytorch/pytorch/issues/50366,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-01-11T16:46:40Z +issue,49724,cure unnecessary NaN values that arise from ±∞ arguments,https://github.com/pytorch/pytorch/issues/49724,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: bootcamp, module: nans and infs, module: numerical-stability",scan when time allows,2021-01-11T13:19:25Z +issue,7359,[feature request] [PyTorch] Dynamic Samplers.,https://github.com/pytorch/pytorch/issues/7359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-01-11T04:39:48Z +issue,50352,torch.tril_indices is incompatible with np.tril_indices,https://github.com/pytorch/pytorch/issues/50352,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: numpy, module: ux",scan when time allows,2021-01-11T02:50:49Z +issue,50268,CUDA error when using `binary_cross_entropy_with_logits`,https://github.com/pytorch/pytorch/issues/50268,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2021-01-10T18:19:57Z +issue,50333,Function Request: scipy.stats.pearsonr,https://github.com/pytorch/pytorch/issues/50333,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-01-10T11:31:34Z +issue,50336,Function Request: scipy.ndimage.zoom,https://github.com/pytorch/pytorch/issues/50336,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2021-01-10T11:28:03Z +issue,9186,[pytorch] Make dtype second positional argument of tensor factory methods,https://github.com/pytorch/pytorch/issues/9186,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: pybind, module: ux",scan when time allows,2021-01-10T10:51:02Z +issue,7745,[Caffe2] Align Element-Wise Ops Broadcasting to Numpy,https://github.com/pytorch/pytorch/issues/7745,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-01-10T08:48:32Z +issue,8062,CrossEntropyLoss mishandles weights,https://github.com/pytorch/pytorch/issues/8062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2021-01-10T08:30:07Z +issue,50340,Function request: scipy.interpolate.InterpolatedUnivariateSpline,https://github.com/pytorch/pytorch/issues/50340,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2021-01-10T07:01:08Z +issue,50338,Function request: scipy.interpolate.RegularGridInterpolator,https://github.com/pytorch/pytorch/issues/50338,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2021-01-10T06:57:42Z +issue,50337,Function request: scipy.interpolate.RectBivariateSpline,https://github.com/pytorch/pytorch/issues/50337,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: numpy",scan when time allows,2021-01-10T06:55:46Z +issue,46222,Use better tempfile creation mechanism to avoid skip windows test,https://github.com/pytorch/pytorch/issues/46222,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2021-01-10T06:44:25Z +issue,28258,[jit] TorchScript classes don't work in notebooks,https://github.com/pytorch/pytorch/issues/28258,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2021-01-08T23:29:14Z +issue,9286,Accumulate into accreal instead of real for CPU loss functions,https://github.com/pytorch/pytorch/issues/9286,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2021-01-08T22:27:17Z +issue,9250,[feature request] SSIM-based cost function as part of the standard set of loss functions,https://github.com/pytorch/pytorch/issues/9250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2021-01-08T18:28:31Z +issue,9193,[Feature Request] Additional torch.nn.LSTM functionality,https://github.com/pytorch/pytorch/issues/9193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2021-01-08T18:27:53Z +issue,49431,clip_grad_norm_ performance regression,https://github.com/pytorch/pytorch/issues/49431,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization, module: performance",scan when time allows,2021-01-08T06:07:14Z +issue,32113,Inconsistent linking flags result in error when building lib/libc10_cuda.so,https://github.com/pytorch/pytorch/issues/32113,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2021-01-08T04:15:16Z +issue,47291,ProcessGroupNCCL NCCL lib version mismatch,https://github.com/pytorch/pytorch/issues/47291,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, oncall: distributed",scan when time allows,2021-01-08T00:49:13Z +issue,50213,torch.float_power out= and inplace variant errors on non-matching output dtype instead of casting,https://github.com/pytorch/pytorch/issues/50213,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2021-01-08T00:28:35Z +issue,50185,CUDACachingAllocator is not GC-aware,https://github.com/pytorch/pytorch/issues/50185,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2021-01-07T17:59:57Z +issue,50118,torch.where scalar/tensor documentation is unclear and not formatted,https://github.com/pytorch/pytorch/issues/50118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sorting and selection",scan when time allows,2021-01-07T17:14:35Z +issue,50190,Calling emptyCache in Conv_v7.cpp causes performance degradation,https://github.com/pytorch/pytorch/issues/50190,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2021-01-07T16:54:15Z +issue,49865,torch.solve on Jetson is slower than humans,https://github.com/pytorch/pytorch/issues/49865,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: linear algebra",scan when time allows,2021-01-07T08:37:35Z +issue,50090,Torch native_layer_norm OP out-of-bounds access,https://github.com/pytorch/pytorch/issues/50090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2021-01-07T05:44:15Z +issue,50170,Automated Mixed Precision not documented to work with nn.DataParallel,https://github.com/pytorch/pytorch/issues/50170,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2021-01-07T04:48:40Z +issue,50136,Inconsistent Variable Naming in FindTorch.cmake,https://github.com/pytorch/pytorch/issues/50136,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2021-01-07T01:16:06Z +issue,49849,[dataloader] RuntimeError: Too many open files when yielding integers,https://github.com/pytorch/pytorch/issues/49849,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-01-06T21:17:40Z +issue,49992,"[docs] nn.modules pages should mention corresponding functional versions, e.g. for nn.Hardshrink",https://github.com/pytorch/pytorch/issues/49992,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2021-01-06T13:00:30Z +issue,49885,LR scheduler `get_lr()` bug,https://github.com/pytorch/pytorch/issues/49885,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2021-01-06T12:12:26Z +issue,50094,Broken LAPACK links in the documentation,https://github.com/pytorch/pytorch/issues/50094,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: linear algebra",scan when time allows,2021-01-06T04:22:36Z +issue,49207,Provide a mechanism to limit the workspace size of cudnn convolution,https://github.com/pytorch/pytorch/issues/49207,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2021-01-05T23:26:55Z +issue,50076,resize_ documentation does not match implementation when memory_format is given,https://github.com/pytorch/pytorch/issues/50076,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2021-01-05T19:36:04Z +issue,9171,Mismatch in behaviour of WeightedRandomSampler and other samplers,https://github.com/pytorch/pytorch/issues/9171,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2021-01-05T15:43:13Z +issue,9133,[Caffe2] Error running net train when running resnet50,https://github.com/pytorch/pytorch/issues/9133,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2021-01-05T15:42:34Z +issue,8821,Storages still use legacy printing,https://github.com/pytorch/pytorch/issues/8821,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: printing,scan when time allows,2021-01-04T18:34:58Z +issue,16703,split_with_sizes should accept a LongTensor as the split_sizes parameter,https://github.com/pytorch/pytorch/issues/16703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2021-01-04T18:05:52Z +issue,49673,Is there any upper-level API to disable the detection of the weight’s version,https://github.com/pytorch/pytorch/issues/49673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2021-01-04T17:00:55Z +issue,50031,Inconsistent function name between stub and implementation in `torch.optim.swa_utils`,https://github.com/pytorch/pytorch/issues/50031,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2021-01-04T16:38:51Z +issue,37713,Builtin FusedLayerNorm is slower than apex one,https://github.com/pytorch/pytorch/issues/37713,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2021-01-02T23:46:13Z +issue,34064,pip --requirement installs incorrect CUDA version,https://github.com/pytorch/pytorch/issues/34064,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: dependency bug, oncall: releng",scan when time allows,2020-12-31T18:17:37Z +issue,8874,Issues with dynamically created grad_fn for views,https://github.com/pytorch/pytorch/issues/8874,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: viewing and reshaping",scan when time allows,2020-12-30T13:13:04Z +issue,49784,Compatibility with Raspberry Pi,https://github.com/pytorch/pytorch/issues/49784,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-12-29T21:43:35Z +issue,49874,Accessibility Coding Needed,https://github.com/pytorch/pytorch/issues/49874,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2020-12-29T20:31:03Z +issue,49873,Accessibility of Website for Low Vision Users,https://github.com/pytorch/pytorch/issues/49873,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2020-12-29T20:28:52Z +issue,49476,Use METH_FASTCALL protocol in Python bindings,https://github.com/pytorch/pytorch/issues/49476,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2020-12-29T20:21:40Z +issue,49891,Support DDP find_unused_parameters=True mode when combined with Pipe,https://github.com/pytorch/pytorch/issues/49891,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: rpc, oncall: distributed",scan when time allows,2020-12-29T19:41:21Z +issue,8871,Use target_compile_options to set warning flags,https://github.com/pytorch/pytorch/issues/8871,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-12-29T18:17:44Z +issue,9032,Lint check for non-Unicode characters in diffs / Unicode characters without coding,https://github.com/pytorch/pytorch/issues/9032,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2020-12-29T18:04:25Z +issue,38684,Add Plackett-Luce distribution,https://github.com/pytorch/pytorch/issues/38684,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-12-29T00:15:40Z +issue,46971,Yolov5 detect.py(ingerence) Error:Torch.nn.modules.module.ModuleAttributeError: 'Hardswish' object has no attribute 'inplace',https://github.com/pytorch/pytorch/issues/46971,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pickle",scan when time allows,2020-12-28T18:06:31Z +issue,49416,A few years after #701 and PyTorch is still using implicit __all__ imports.,https://github.com/pytorch/pytorch/issues/49416,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: codegen, module: pybind, module: typing",scan when time allows,2020-12-28T16:20:28Z +issue,49224,Torch abs op crashes on -128 int8 tensor with ASAN,https://github.com/pytorch/pytorch/issues/49224,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2020-12-28T16:17:39Z +issue,43672,Make `torch.Generator` picklable,https://github.com/pytorch/pytorch/issues/43672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-12-28T02:49:20Z +issue,49539,Improve AssertionError for torch.nn.functional.pad 'replicate',https://github.com/pytorch/pytorch/issues/49539,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: nn",scan when time allows,2020-12-25T00:20:35Z +issue,49727,RandomSampler generator created in every iteration,https://github.com/pytorch/pytorch/issues/49727,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-12-23T21:07:08Z +issue,49723,Flattening nn.Parameters while maintaining gradients from neural network forward pass,https://github.com/pytorch/pytorch/issues/49723,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: optimizer",scan when time allows,2020-12-23T03:35:15Z +issue,49677,"File ""setup.py"", line 773:subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '8']' returned non-zero exit status 2",https://github.com/pytorch/pytorch/issues/49677,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-12-22T22:07:49Z +issue,49672,Model passed through queue hangs while loading in sub process,https://github.com/pytorch/pytorch/issues/49672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2020-12-22T19:47:09Z +issue,49716,Two independent future chains in PowerSGD cannot be kicked off asynchronously,https://github.com/pytorch/pytorch/issues/49716,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-12-22T18:54:42Z +issue,42459,[JIT] Runtime error when backpropagating through input gradient,https://github.com/pytorch/pytorch/issues/42459,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-12-22T03:27:21Z +issue,8602,Come with a better strategy for TensorArg (error reporting),https://github.com/pytorch/pytorch/issues/8602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: performance",scan when time allows,2020-12-21T20:02:28Z +issue,49642,Turn deprecation warnings into errors in CI,https://github.com/pytorch/pytorch/issues/49642,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",scan when time allows,2020-12-21T18:28:46Z +issue,30388,Increasing memory usage on CPU,https://github.com/pytorch/pytorch/issues/30388,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: memory usage, module: performance",scan when time allows,2020-12-21T17:09:24Z +issue,49398,Can we add try-except for list/slice indices even when auto_collation is True?,https://github.com/pytorch/pytorch/issues/49398,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2020-12-21T07:27:05Z +issue,49653,"torch.cuda.amp, example with 20% memory increase compared to apex/amp",https://github.com/pytorch/pytorch/issues/49653,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2020-12-20T16:54:08Z +issue,25814,(PyTorch1.1 and 1.2) RuntimeError: Can't detach views in-place. Use detach() instead,https://github.com/pytorch/pytorch/issues/25814,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: optimizer",scan when time allows,2020-12-20T12:08:42Z +issue,49282,Bug when combining AMP with channels_last memory format,https://github.com/pytorch/pytorch/issues/49282,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: memory format",scan when time allows,2020-12-20T10:47:28Z +issue,49427,Inserting named tensor into other fails,https://github.com/pytorch/pytorch/issues/49427,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-12-19T10:13:01Z +issue,40659,[typing] overly restrictive List[int],https://github.com/pytorch/pytorch/issues/40659,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2020-12-18T20:24:53Z +issue,42460,Training with AMP gives overflow warning,https://github.com/pytorch/pytorch/issues/42460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda",scan when time allows,2020-12-18T10:07:35Z +issue,49023,Automatically rerun tests with CUDA_LAUNCH_BLOCKING=1 when they fail with CUDA errors in CI,https://github.com/pytorch/pytorch/issues/49023,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tests",scan when time allows,2020-12-18T07:28:42Z +issue,49475,Write link in README.md,https://github.com/pytorch/pytorch/issues/49475,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-12-18T06:17:09Z +issue,35180,hasSpecialCase INTERNAL ASSERT FAILED: We don't have an op for aten::to but it isn't a special case.,https://github.com/pytorch/pytorch/issues/35180,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-12-17T09:04:24Z +issue,49321,Backward hangs with DDP during training.,https://github.com/pytorch/pytorch/issues/49321,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, oncall: distributed",scan when time allows,2020-12-17T00:32:55Z +issue,49477,Callbacks may not be automatically synchronized in a single NCCL future chain,https://github.com/pytorch/pytorch/issues/49477,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-12-16T19:15:07Z +issue,49320,how could I print the log in source code,https://github.com/pytorch/pytorch/issues/49320,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: logging",scan when time allows,2020-12-16T16:04:01Z +issue,8561,cleanup BLAS detection,https://github.com/pytorch/pytorch/issues/8561,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: linear algebra",scan when time allows,2020-12-16T15:27:30Z +issue,8386,Properly release NCCL resources,https://github.com/pytorch/pytorch/issues/8386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nccl,scan when time allows,2020-12-16T15:22:16Z +issue,44182,Support keep stride for neg with requires_grad=False,https://github.com/pytorch/pytorch/issues/44182,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: numpy",scan when time allows,2020-12-15T20:29:55Z +issue,49332,global_pruning costs memory after model is trained,https://github.com/pytorch/pytorch/issues/49332,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: pruning",scan when time allows,2020-12-15T15:21:14Z +issue,49347,[pre-commit hook] fails on `BLK100 Black would make changes.`,https://github.com/pytorch/pytorch/issues/49347,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-12-15T05:40:08Z +issue,49300,is this a typo in optimizer.pyi ? it says `statue` instead of `state`,https://github.com/pytorch/pytorch/issues/49300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-12-15T04:35:59Z +issue,8097,manager.cpp:64: undefined reference to `shm_open (when building with GCC 5.x (sic)),https://github.com/pytorch/pytorch/issues/8097,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-12-11T16:38:34Z +issue,8107,Export CC is ignored when I build pytorch,https://github.com/pytorch/pytorch/issues/8107,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-12-11T16:37:51Z +issue,17920,Build compact libtorch from source with cmake,https://github.com/pytorch/pytorch/issues/17920,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2020-12-11T13:24:05Z +issue,49217,Bug in profiler on CI machines when profiling NCCL distributed calls,https://github.com/pytorch/pytorch/issues/49217,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2020-12-11T04:42:33Z +issue,49076,Add List Type Data through add_hparam method in torch.utils.tensorboard.SummaryWriter,https://github.com/pytorch/pytorch/issues/49076,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-12-11T02:02:48Z +issue,49178,[NNC] Some ops have type promotion logic which adds extra casts & does compute in different dtype than eager,https://github.com/pytorch/pytorch/issues/49178,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2020-12-10T22:53:44Z +issue,49177,Investigate `torch.linalg.norm` performance,https://github.com/pytorch/pytorch/issues/49177,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: performance",scan when time allows,2020-12-10T22:41:06Z +issue,49166,Implement fill_out on complex,https://github.com/pytorch/pytorch/issues/49166,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-12-10T20:23:07Z +issue,49082,swa_utils.bn_update is too opinionated in how it calls the model,https://github.com/pytorch/pytorch/issues/49082,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-12-10T20:15:25Z +issue,49139,"Hello everyone, why does libtorch half precision float 16 leak memory?大家好，为什么libtorch半精度float16会内存泄漏？",https://github.com/pytorch/pytorch/issues/49139,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-12-10T20:11:32Z +issue,45493,Build LibTorch for cpu using OpenBLAS: Had to manually remove a path in Caffe2Targets.cmake,https://github.com/pytorch/pytorch/issues/45493,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-12-10T15:23:58Z +issue,49133,Unknown failure for test_reference_numerics_sinc_cpu_complex64,https://github.com/pytorch/pytorch/issues/49133,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: trigonometric functions",scan when time allows,2020-12-10T06:24:53Z +issue,48526,CARU: A Content-Adaptive Recurrent Unit for the Transition of Hidden State in NLP,https://github.com/pytorch/pytorch/issues/48526,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2020-12-09T08:30:45Z +issue,48945,CUDA error: illegal memory access Conv3d,https://github.com/pytorch/pytorch/issues/48945,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2020-12-08T17:30:20Z +issue,48998,Torch1.2 can't be downloaded!,https://github.com/pytorch/pytorch/issues/48998,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-12-08T17:29:17Z +issue,48645,Can sublist a nn.Sequential subclass,https://github.com/pytorch/pytorch/issues/48645,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-12-08T17:06:49Z +issue,48641,"[bug] `torch.{sinh, cosh}`: Incorrect values for vectorized path",https://github.com/pytorch/pytorch/issues/48641,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: nans and infs, module: sleef, module: trigonometric functions",scan when time allows,2020-12-08T11:09:05Z +issue,9282,[gradcheck] warn about the case that mulitple inputs share storage,https://github.com/pytorch/pytorch/issues/9282,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-12-07T20:07:18Z +issue,48878,Libtorch: nvrtc: error: invalid value for --gpu-architecture (-arch),https://github.com/pytorch/pytorch/issues/48878,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-12-07T19:59:08Z +issue,48929,Domain Specific Batch Normalization,https://github.com/pytorch/pytorch/issues/48929,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-12-07T19:28:53Z +issue,28245,PyTorch RPC should expose critical metrics to the application.,https://github.com/pytorch/pytorch/issues/28245,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-12-07T02:32:19Z +issue,48892,is_non_overlapping_and_dense() does not error out for sparse tensors,https://github.com/pytorch/pytorch/issues/48892,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-12-06T03:42:03Z +issue,46564,Clean up request_callback_no_python.cpp,https://github.com/pytorch/pytorch/issues/46564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-12-04T23:45:36Z +issue,48291,pytorch is not linked with support for cuda devices,https://github.com/pytorch/pytorch/issues/48291,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-12-04T12:41:35Z +issue,48706,test_inverse setup is flaky using MKL>=2020.1 on certain CPUs and fails on CUDA,https://github.com/pytorch/pytorch/issues/48706,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: mkl",scan when time allows,2020-12-03T17:51:31Z +issue,7944,Better error message in DataChannelTCP::_receive,https://github.com/pytorch/pytorch/issues/7944,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2020-12-03T16:30:46Z +issue,8035,[PyTorch] EmbeddingBag comparison vs Embedding fails w/ small max_norm on CUDA,https://github.com/pytorch/pytorch/issues/8035,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: norms and normalization,scan when time allows,2020-12-03T16:21:34Z +issue,26218,Default adam epsilon to 1e-7 when on fp16,https://github.com/pytorch/pytorch/issues/26218,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: half, module: numerical-stability, module: optimizer",scan when time allows,2020-12-03T07:53:28Z +issue,48466,[master] new default signature: op=from `op=,https://github.com/pytorch/pytorch/issues/48466,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: distributed",scan when time allows,2020-12-03T01:03:58Z +issue,48684,Remove remaining native:: references from non-native ATen,https://github.com/pytorch/pytorch/issues/48684,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-12-02T14:54:06Z +issue,7834,torch.Tensor.new() disappeared in 0.4 doc,https://github.com/pytorch/pytorch/issues/7834,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: docs, module: tensor creation",scan when time allows,2020-12-02T07:14:08Z +issue,48649,[docs] A single additional summary page for diag* methods,https://github.com/pytorch/pytorch/issues/48649,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-12-01T22:44:16Z +issue,48621,[FR] loss reduction argument accepts None,https://github.com/pytorch/pytorch/issues/48621,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2020-12-01T22:37:47Z +issue,48648,Confused on project website pointer.,https://github.com/pytorch/pytorch/issues/48648,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-12-01T22:33:36Z +issue,48650,[feature request] Keyword-only device argument (and maybe dtype) for torch.meshgrid,https://github.com/pytorch/pytorch/issues/48650,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-12-01T22:31:47Z +issue,48667,Operator registration doesn't work with noexcept functions on some compilers (c10::guts::is_function_type rejects noexcept),https://github.com/pytorch/pytorch/issues/48667,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-12-01T22:28:17Z +issue,48153,More robust list comprehension,https://github.com/pytorch/pytorch/issues/48153,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-12-01T18:49:04Z +issue,48548,at::size documentation conflict,https://github.com/pytorch/pytorch/issues/48548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2020-12-01T16:35:55Z +issue,46464,[META] Assert that expected libraries appear in libtorch,https://github.com/pytorch/pytorch/issues/46464,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: static linking,scan when time allows,2020-12-01T15:54:54Z +issue,48083,CPU Tensor with Python MP Freezing in Docker Container,https://github.com/pytorch/pytorch/issues/48083,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docker, module: multiprocessing",scan when time allows,2020-12-01T15:41:48Z +issue,48587,[feature request] Give `torch.cholesky` an optional fallback to test whether the tensor is positive definite,https://github.com/pytorch/pytorch/issues/48587,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2020-12-01T07:56:05Z +issue,48353,[FR] bool tensor should support basic arithmetics,https://github.com/pytorch/pytorch/issues/48353,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: numpy, module: type promotion",scan when time allows,2020-11-30T18:09:25Z +issue,48235,Softplus differs depending on number of elements in tensor,https://github.com/pytorch/pytorch/issues/48235,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-11-30T18:04:45Z +issue,48281,Batched inplace mm changes stride when out size is correct,https://github.com/pytorch/pytorch/issues/48281,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: linear algebra, module: memory format",scan when time allows,2020-11-30T17:52:49Z +issue,44533,“doxygenfunction: Unable to resolve multiple matches for function ...” in C++ documentation,https://github.com/pytorch/pytorch/issues/44533,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2020-11-30T17:43:43Z +issue,48518,C++ optimizer check for duplicate parameters,https://github.com/pytorch/pytorch/issues/48518,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-11-30T15:40:11Z +issue,48377,Management of Gpu memory to avoid memory errors,https://github.com/pytorch/pytorch/issues/48377,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-11-27T16:05:02Z +issue,48366,Add Go to high level interface,https://github.com/pytorch/pytorch/issues/48366,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-11-27T07:27:03Z +issue,48450,"Qt 5.14.2, libtorch1.7.0 cuda10.2，Error after using header file",https://github.com/pytorch/pytorch/issues/48450,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: static linking,scan when time allows,2020-11-27T07:13:50Z +issue,48511,Ambiguous RuntimeError raised when device of input PackedSequence does not match with nn.LSTM's device,https://github.com/pytorch/pytorch/issues/48511,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2020-11-27T07:01:05Z +issue,7900,Deprecate torch.Tensor,https://github.com/pytorch/pytorch/issues/7900,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: tensor creation",scan when time allows,2020-11-27T06:58:46Z +issue,48306,[doc] example for pairwise distance matrix,https://github.com/pytorch/pytorch/issues/48306,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: docs, module: nn",scan when time allows,2020-11-26T23:17:22Z +issue,48305,[RFC] CUDA-aware future for distributed,https://github.com/pytorch/pytorch/issues/48305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: rpc, oncall: distributed",scan when time allows,2020-11-26T19:51:26Z +issue,48457,compile error immintrin.h,https://github.com/pytorch/pytorch/issues/48457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-26T15:44:25Z +issue,46202,pytorch not uploading data to tensorboard page,https://github.com/pytorch/pytorch/issues/46202,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-11-26T07:31:09Z +issue,28218,"index_sub, index_mul and index_div",https://github.com/pytorch/pytorch/issues/28218,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-11-26T06:30:15Z +issue,48338,torch.float128 datatype,https://github.com/pytorch/pytorch/issues/48338,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-11-26T00:01:29Z +issue,7773,[feature request] Add Local Contrast Normalization,https://github.com/pytorch/pytorch/issues/7773,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-11-25T23:51:21Z +issue,7865,[PyTorch] weight tensor dimension assumption,https://github.com/pytorch/pytorch/issues/7865,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-11-25T23:47:20Z +issue,44248,Additional clutter memory allocated on GPU 0 when training on GPU 1+,https://github.com/pytorch/pytorch/issues/44248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-11-25T22:11:27Z +issue,34799,Is it possible to run an object detection android app by using Pytorch Mobile?,https://github.com/pytorch/pytorch/issues/34799,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: android,scan when time allows,2020-11-25T14:48:48Z +issue,22454,Tracing an RNN does not support torch.nn.utils.rnn.PackedSequence as input,https://github.com/pytorch/pytorch/issues/22454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-11-25T11:40:13Z +issue,33463,torch._C.Node.scopeName() missing in pytorch 1.4,https://github.com/pytorch/pytorch/issues/33463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-11-25T08:26:02Z +issue,25004,[C++] Call find_package(Torch REQUIRED) more than one time in downstream project causes CMake configuration error,https://github.com/pytorch/pytorch/issues/25004,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-25T08:19:25Z +issue,22791,How to use mpi backend without CUDA_aware,https://github.com/pytorch/pytorch/issues/22791,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mpi,scan when time allows,2020-11-25T06:05:04Z +issue,48202,"""Found no NVIDIA driver"" produces too much log",https://github.com/pytorch/pytorch/issues/48202,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-11-23T17:05:28Z +issue,7740,Conv3D can be optimized for cases when kernel is spatial (probably),https://github.com/pytorch/pytorch/issues/7740,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: performance",scan when time allows,2020-11-20T16:53:07Z +issue,45459,Backward for sparse tensor item select does not work,https://github.com/pytorch/pytorch/issues/45459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2020-11-20T14:48:34Z +issue,47307,[Feature request] Stochastic Frank-Wolfe optimizer,https://github.com/pytorch/pytorch/issues/47307,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-11-20T08:21:57Z +issue,33785,Deepcopy fails with nn.parallel.replicate,https://github.com/pytorch/pytorch/issues/33785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-11-20T02:12:30Z +issue,48167,[libtorch] Build shared library with libtorch and compile a static library,https://github.com/pytorch/pytorch/issues/48167,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: static linking,scan when time allows,2020-11-20T01:59:49Z +issue,48227,Modifying values() tensor of COO tensor requiring grad throws an odd error message,https://github.com/pytorch/pytorch/issues/48227,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2020-11-19T22:33:18Z +issue,33628,Remove .data subset1 for fixathon,https://github.com/pytorch/pytorch/issues/33628,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-11-19T03:36:25Z +issue,28742,torch.as_tensor fails to create named tensors,https://github.com/pytorch/pytorch/issues/28742,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: named tensor, module: tensor creation",scan when time allows,2020-11-18T15:49:40Z +issue,38765,LibTorch 1.5.0 not supporting GLIBC < 2.23,https://github.com/pytorch/pytorch/issues/38765,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2020-11-18T11:58:29Z +issue,47906,amp for custom op,https://github.com/pytorch/pytorch/issues/47906,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: custom-operators",scan when time allows,2020-11-18T01:01:16Z +issue,7610,Inconsistent interactions of PyTorch tensors and NumPy ops,https://github.com/pytorch/pytorch/issues/7610,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-11-17T19:46:16Z +issue,48064,FAILED: test_api/CMakeFiles/test_api.dir/dataloader.cpp.o,https://github.com/pytorch/pytorch/issues/48064,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-17T16:24:01Z +issue,48088,[dataloader] Worker threads to print the signal they received before they die,https://github.com/pytorch/pytorch/issues/48088,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-11-17T15:45:28Z +issue,47588,Can we apply Weight normalization hook to a method other than `forward`?,https://github.com/pytorch/pytorch/issues/47588,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: norms and normalization",scan when time allows,2020-11-17T08:12:43Z +issue,47841,Factorial & Binomial Coefficient,https://github.com/pytorch/pytorch/issues/47841,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-11-17T06:48:17Z +issue,47954,[Feature] torch.bitwise_where,https://github.com/pytorch/pytorch/issues/47954,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-11-17T01:56:13Z +issue,48033,Missing -fopenmp when used in torchvision,https://github.com/pytorch/pytorch/issues/48033,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: build warnings, module: cpp-extensions, module: openmp",scan when time allows,2020-11-16T20:16:02Z +issue,23434,einsum equation with conditional mask works in numpy but not in PyTorch,https://github.com/pytorch/pytorch/issues/23434,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: type promotion",scan when time allows,2020-11-16T18:18:20Z +issue,45447,Retire usages of CUDA_tensor_apply helpers in ATen,https://github.com/pytorch/pytorch/issues/45447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: internals",scan when time allows,2020-11-16T18:00:07Z +issue,45578,Why is RTX3080 slower than RTX2020-Ti?,https://github.com/pytorch/pytorch/issues/45578,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-11-16T17:54:21Z +issue,47912,Code becomes more than x20 slower after upgrading torch version,https://github.com/pytorch/pytorch/issues/47912,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2020-11-16T17:10:51Z +issue,47901,"Get forward output, grad, hessian all at once",https://github.com/pytorch/pytorch/issues/47901,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-11-16T16:25:46Z +issue,7801,Checkpointing is slow on nn.DataParallel models,https://github.com/pytorch/pytorch/issues/7801,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: performance",scan when time allows,2020-11-15T04:54:57Z +issue,25714,Backtrace prints many ,https://github.com/pytorch/pytorch/issues/25714,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-14T15:13:47Z +issue,15378,Projective Transformation grid generator,https://github.com/pytorch/pytorch/issues/15378,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-11-14T05:39:45Z +issue,47824,"DDP error, when I use ""if"" branch in forward function.",https://github.com/pytorch/pytorch/issues/47824,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: ddp",scan when time allows,2020-11-14T04:55:55Z +issue,47904,Memory leak in nn.MaxPool2d layer when run on iOS,https://github.com/pytorch/pytorch/issues/47904,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ios, module: memory usage",scan when time allows,2020-11-14T00:13:33Z +issue,46918,Jacobians computed by autograd.functional.jacobian with compute_graph sometimes set requires_grad True,https://github.com/pytorch/pytorch/issues/46918,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-11-13T17:37:02Z +issue,47751,Make libtorch modular,https://github.com/pytorch/pytorch/issues/47751,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build, module: selective build",scan when time allows,2020-11-13T17:28:17Z +issue,7353,"Windows MAGMA binary requires explicit linking against MKL LAPACK, or it will silently give incorrect results",https://github.com/pytorch/pytorch/issues/7353,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-11-13T16:07:06Z +issue,7352,Check for F2C convention (for blas) at runtime,https://github.com/pytorch/pytorch/issues/7352,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-13T16:06:34Z +issue,9375,[Caffe2] Not handle attribute 'spatial' of ONNX operator 'BatchNormalization',https://github.com/pytorch/pytorch/issues/9375,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-11-13T07:16:03Z +issue,30867,save model definition file,https://github.com/pytorch/pytorch/issues/30867,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-11-13T02:42:38Z +issue,47753,"Attempted to read a PyTorch file with version 3, but the maximum supported version for reading is 2",https://github.com/pytorch/pytorch/issues/47753,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-11-12T15:21:21Z +issue,47649,error: reference to __host__ function 'parallel_for 1,https://github.com/pytorch/pytorch/issues/7423,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-11-09T22:33:00Z +issue,7358,DataParallel on list inputs,https://github.com/pytorch/pytorch/issues/7358,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-11-09T22:31:39Z +issue,7365,nn.DataParallel fills None grads with 0,https://github.com/pytorch/pytorch/issues/7365,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-11-09T22:31:26Z +issue,47563,Handling multiple large-scale datasets efficiently,https://github.com/pytorch/pytorch/issues/47563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-11-09T18:21:16Z +issue,47560,Minumul LR is never reached in,https://github.com/pytorch/pytorch/issues/47560,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-11-09T16:11:14Z +issue,42173,Exclusive flag in torch.cumsum,https://github.com/pytorch/pytorch/issues/42173,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-11-08T19:05:12Z +issue,7546,[Caffe2] Build broken on macOS High Sierra: can't find sys headers in /usr/local/include.,https://github.com/pytorch/pytorch/issues/7546,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-11-08T04:58:38Z +issue,3818,Memory leak when doing backward with grad as yourself,https://github.com/pytorch/pytorch/issues/3818,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",scan when time allows,2020-11-07T19:33:24Z +issue,38630,Torchvision error TypeError: _resolve_type_from_object(),https://github.com/pytorch/pytorch/issues/38630,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-11-07T08:46:06Z +issue,47447,"thread blocked when moving a tensor from GPU to CPU, by calling the function .cpu() in pytorch. This kind of block can be stop by any window event like mouse moving/clicking or keyboard pressing.",https://github.com/pytorch/pytorch/issues/47447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda",scan when time allows,2020-11-06T05:38:18Z +issue,47357,Conv1D formula in docs is wrong,https://github.com/pytorch/pytorch/issues/47357,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2020-11-06T05:36:02Z +issue,47341,ConvTranspose1d groups=channels is very slow!!!,https://github.com/pytorch/pytorch/issues/47341,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2020-11-05T16:26:57Z +issue,46807,[cudatoolkit 11.0] segfaults,https://github.com/pytorch/pytorch/issues/46807,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-11-05T16:15:34Z +issue,41142,Support for More Loss Functions,https://github.com/pytorch/pytorch/issues/41142,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2020-11-04T19:04:29Z +issue,26097,[RFC] TensorBoard extensions and improvements for PyTorch,https://github.com/pytorch/pytorch/issues/26097,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-11-04T16:26:39Z +issue,40968,Pytorch 1.4 compilation hangs on AMD Epyc,https://github.com/pytorch/pytorch/issues/40968,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vectorization",scan when time allows,2020-11-03T18:54:57Z +issue,35116,JIT does not support class instance attribute type annotation,https://github.com/pytorch/pytorch/issues/35116,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-11-03T16:01:09Z +issue,15918,Broadcasting and Additional Dimensions for pairwise_distance,https://github.com/pytorch/pytorch/issues/15918,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: nn",scan when time allows,2020-11-03T05:39:42Z +issue,7214,Do not put system paths in RPATH,https://github.com/pytorch/pytorch/issues/7214,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-03T04:22:59Z +issue,47208,[RFC] Add test execution time analysis CI workflow,https://github.com/pytorch/pytorch/issues/47208,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2020-11-02T18:43:06Z +issue,25150,Problems with install python from source,https://github.com/pytorch/pytorch/issues/25150,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-11-02T18:30:21Z +issue,46248,"""distributed"" NCCL tests fail when having more than 3 GPUs",https://github.com/pytorch/pytorch/issues/46248,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ddp, module: nccl, module: tests, oncall: distributed",scan when time allows,2020-11-02T15:14:35Z +issue,32531,Batched torch.eig() and gradient of torch.eig() for real eigenvalues,https://github.com/pytorch/pytorch/issues/32531,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: batching, module: linear algebra",scan when time allows,2020-11-02T14:34:18Z +issue,35535,Apple Review Rejected. ITMS-90338: Non-public API usage In Pytorch For iOS,https://github.com/pytorch/pytorch/issues/35535,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ios,scan when time allows,2020-11-02T05:30:57Z +issue,46947,Add Optimistic Mirror Descent (OMD),https://github.com/pytorch/pytorch/issues/46947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-11-02T03:47:24Z +issue,7032,[feature request] norm argument for RNNCells,https://github.com/pytorch/pytorch/issues/7032,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2020-10-30T20:20:34Z +issue,3823,Fuse bias to CuDNN convolution,https://github.com/pytorch/pytorch/issues/3823,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-30T19:30:17Z +issue,47111,THCudaShutdown should be called before THCState_free,https://github.com/pytorch/pytorch/issues/47111,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2020-10-30T13:30:56Z +issue,46428,send_object/recv_object APIs for c10d,https://github.com/pytorch/pytorch/issues/46428,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2020-10-29T17:16:55Z +issue,46999,"torch.sum(tensor(2.), dim=0) (and probably other reduction functions) doesn't make sense",https://github.com/pytorch/pytorch/issues/46999,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: reductions",scan when time allows,2020-10-29T16:06:16Z +issue,46522,ModuleNotFoundError when installing PyTorch via pip on aarch64 environment,https://github.com/pytorch/pytorch/issues/46522,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries",scan when time allows,2020-10-29T14:44:48Z +issue,46980,torch.acos not supported for sparse layout,https://github.com/pytorch/pytorch/issues/46980,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-10-28T17:28:09Z +issue,43352,"Remove warning, and update documentation.",https://github.com/pytorch/pytorch/issues/43352,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2020-10-28T14:56:35Z +issue,46929,Pooling code does not allow sliding window starting in right padded region,https://github.com/pytorch/pytorch/issues/46929,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2020-10-27T20:19:07Z +issue,46820,simple v *= v_scale error,https://github.com/pytorch/pytorch/issues/46820,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-10-27T17:45:12Z +issue,46899,CUDA out of memory when using torch.load,https://github.com/pytorch/pytorch/issues/46899,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-10-27T15:18:17Z +issue,46869,[META] DO NOT DELETE THIS LABEL,https://github.com/pytorch/pytorch/issues/46869,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-26T19:34:51Z +issue,46819,Specify input dimensions where they are not obvious,https://github.com/pytorch/pytorch/issues/46819,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-10-26T16:56:01Z +issue,46787,Clearer error messages for 'different devices' error in the RNN module,https://github.com/pytorch/pytorch/issues/46787,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-26T16:41:37Z +issue,39662,[RFC][distributed] RFC: c10d ProcessGroup extension and C++ API change,https://github.com/pytorch/pytorch/issues/39662,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-10-26T16:16:24Z +issue,46760,How to define a new data type in native_functions.yaml?,https://github.com/pytorch/pytorch/issues/46760,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-10-26T15:16:34Z +issue,4241,single-gpu works but multi-gpu hangs,https://github.com/pytorch/pytorch/issues/4241,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: data parallel",scan when time allows,2020-10-26T05:54:13Z +issue,32909,Some questions Concerning Intel's DNNL(MKLDNN) support in Pytorch (adding support for Intel Processors GPUs) by transitioning to DNNL,https://github.com/pytorch/pytorch/issues/32909,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mkldnn,scan when time allows,2020-10-25T13:34:05Z +issue,46766,Unable to compile from code v1.6.0,https://github.com/pytorch/pytorch/issues/46766,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-24T06:18:55Z +issue,46724,PROBLEM WITH INSTALLATION PYTORCH FROM SOURCE,https://github.com/pytorch/pytorch/issues/46724,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-24T05:41:35Z +issue,46753,Memory leak when creating new tensors inside nn.DataParallel on multiple GPUs,https://github.com/pytorch/pytorch/issues/46753,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-10-24T01:09:36Z +issue,44026,[Feature] Einsum like ShapeGuard,https://github.com/pytorch/pytorch/issues/44026,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-23T20:06:54Z +issue,46712,The documentation for c10::Dict is completely empty.,https://github.com/pytorch/pytorch/issues/46712,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs, module: internals",scan when time allows,2020-10-23T15:06:03Z +issue,46711,A random split function that return the datasets following specific target (label) distribution.,https://github.com/pytorch/pytorch/issues/46711,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data,scan when time allows,2020-10-22T17:47:51Z +issue,46659,Excluding image processing operators due to no opencv,https://github.com/pytorch/pytorch/issues/46659,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-22T17:36:36Z +issue,46382,"Code asserts when register new aten operations' implementation thru ""c10::RegisterOperators::op"" API.",https://github.com/pytorch/pytorch/issues/46382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-10-22T01:25:49Z +issue,46643,Autograd support for the tensor multiplication of sparse tensors,https://github.com/pytorch/pytorch/issues/46643,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2020-10-21T17:02:27Z +issue,46597,Suggestion: link individual APIs in docs with tutorials that use them,https://github.com/pytorch/pytorch/issues/46597,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-10-21T14:54:04Z +issue,34451,Mac build from source error,https://github.com/pytorch/pytorch/issues/34451,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: xnnpack, oncall: mobile",scan when time allows,2020-10-21T10:01:04Z +issue,45855,Do we have plan to offer C++ binding for prune related features.,https://github.com/pytorch/pytorch/issues/45855,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: pruning",scan when time allows,2020-10-21T04:02:14Z +issue,46530,[Request] Batched Dataset->DataLoader interface,https://github.com/pytorch/pytorch/issues/46530,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-10-20T15:24:11Z +issue,46559,Exception raised in rpc_async context is silently handled,https://github.com/pytorch/pytorch/issues/46559,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2020-10-19T21:07:24Z +issue,46311,"Compile error when use c10::RegisterOperators::Options::kernel with the return type ""std::tuple""",https://github.com/pytorch/pytorch/issues/46311,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-10-19T17:42:08Z +issue,46178,"torch.cuda.amp! when I use @autocast() on DCN(DeformConv), the error ""RuntimeError:expect scalar type Float but Half """,https://github.com/pytorch/pytorch/issues/46178,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda",scan when time allows,2020-10-19T02:30:20Z +issue,37409,Compatibility of subset dataset with disabled batch sampling,https://github.com/pytorch/pytorch/issues/37409,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-10-17T11:03:24Z +issue,46367,[chore] Test pybind11 2.6.0 (RC 2?),https://github.com/pytorch/pytorch/issues/46367,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2020-10-17T00:34:18Z +issue,46350,torch.distributions.half_normal.HalfNormal.cdf returns negative values,https://github.com/pytorch/pytorch/issues/46350,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-10-16T22:18:37Z +issue,42882,Support for defining/saving custom operators with dynamic schema,https://github.com/pytorch/pytorch/issues/42882,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-10-16T21:15:46Z +issue,46379,"Reduce ""reserved"" memory by PyTorch.",https://github.com/pytorch/pytorch/issues/46379,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2020-10-16T04:55:00Z +issue,43429,Inference performance regression caused by hacky_wrapper_for_legacy_signatures,https://github.com/pytorch/pytorch/issues/43429,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dispatch, module: internals, module: performance",scan when time allows,2020-10-16T01:45:42Z +issue,46381,[Feature] Imbalanced batch scattering in th.nn.DataParallel,https://github.com/pytorch/pytorch/issues/46381,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-10-15T22:01:11Z +issue,25045,[Distance functions] F.pdist backward CUDA invalid configuration,https://github.com/pytorch/pytorch/issues/25045,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: distance functions",scan when time allows,2020-10-15T20:36:50Z +issue,15253,[Feature Request] cdist: pairwise distances between two sets of tensors with batch mode,https://github.com/pytorch/pytorch/issues/15253,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: distance functions",scan when time allows,2020-10-15T20:36:14Z +issue,29528,Half precision cdist,https://github.com/pytorch/pytorch/issues/29528,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: distance functions, module: half",scan when time allows,2020-10-15T20:35:44Z +issue,37363,"cdist docs imply only one batch dimension is possible, but actually arbitrarily many are allowed",https://github.com/pytorch/pytorch/issues/37363,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: docs",scan when time allows,2020-10-15T20:35:31Z +issue,37408,torch.cdist() implementation without using contiguous() calls,https://github.com/pytorch/pytorch/issues/37408,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: performance",scan when time allows,2020-10-15T20:35:23Z +issue,37734,torch.cdist returns inconsistent result,https://github.com/pytorch/pytorch/issues/37734,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distance functions, module: docs, module: numerical-stability",scan when time allows,2020-10-15T20:35:19Z +issue,35363,NCCL version upgrade for PyTorch,https://github.com/pytorch/pytorch/issues/35363,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-15T20:26:12Z +issue,46400,Make Undefined reprensentable in DispatchKeySet.,https://github.com/pytorch/pytorch/issues/46400,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-15T18:51:47Z +issue,21411,Implementation of Group equivariant convolutions,https://github.com/pytorch/pytorch/issues/21411,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-15T00:50:48Z +issue,46276,test_distributed_* does not show error details from the subprocess,https://github.com/pytorch/pytorch/issues/46276,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",scan when time allows,2020-10-14T19:37:50Z +issue,46102,test_cat_cuda (__main__.TestTensorExprFuser) fails,https://github.com/pytorch/pytorch/issues/46102,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-10-14T17:55:46Z +issue,46065,TestDataLoader.test_proper_exit takes 2.5min to finish,https://github.com/pytorch/pytorch/issues/46065,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: dataloader, module: tests",scan when time allows,2020-10-14T16:01:44Z +issue,46319,Remove header declarations for CPUType/TypeDefault,https://github.com/pytorch/pytorch/issues/46319,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-10-14T15:09:58Z +issue,46084,"In PyTorch Tutorials, RuntimeError: CUDA error: out of memory happen",https://github.com/pytorch/pytorch/issues/46084,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-10-14T04:11:32Z +issue,37104,[JIT] Use of global value creates confusing error message,https://github.com/pytorch/pytorch/issues/37104,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-10-14T03:34:40Z +issue,46241,ubuntu16.04+vscode compile libtorch error,https://github.com/pytorch/pytorch/issues/46241,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-14T01:50:12Z +issue,32589,User-friendly handling of types and devices,https://github.com/pytorch/pytorch/issues/32589,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: ux",scan when time allows,2020-10-13T22:53:35Z +issue,46254,Enable named tensor inputs to `torch.linalg.norm`,https://github.com/pytorch/pytorch/issues/46254,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-10-13T21:39:26Z +issue,22024,Consolidate definition of operators/gradients where possible,https://github.com/pytorch/pytorch/issues/22024,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-10-13T19:20:02Z +issue,1750,Add support for colors (and maybe other attributes) to NVTX API,https://github.com/pytorch/pytorch/issues/1750,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-10-12T21:32:50Z +issue,46184,redesign of Dropout,https://github.com/pytorch/pytorch/issues/46184,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-12T19:52:57Z +issue,46176,BucketSampler for easy variable-length input batching,https://github.com/pytorch/pytorch/issues/46176,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-10-12T19:36:56Z +issue,46149,seeding does not work when I initialize a linear model even if I do not use it in the code,https://github.com/pytorch/pytorch/issues/46149,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-12T16:59:41Z +issue,46139,"Policy CMP0012 is not set, CMake, Building For C++ With PyTorch CUDA",https://github.com/pytorch/pytorch/issues/46139,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-12T16:53:16Z +issue,46093,"Spawn wrapper should catch BaseException, not Exception",https://github.com/pytorch/pytorch/issues/46093,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2020-10-12T14:37:47Z +issue,46066,TestXNNPACKConv1dTransformPass.test_conv1d_with_relu_fc takes 2+ min to finsh,https://github.com/pytorch/pytorch/issues/46066,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",scan when time allows,2020-10-12T14:26:02Z +issue,46061,Support undispatched ops in codegen,https://github.com/pytorch/pytorch/issues/46061,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2020-10-12T14:22:05Z +issue,43467,Data loader struct pack issue(overflow)?,https://github.com/pytorch/pytorch/issues/43467,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2020-10-10T12:00:24Z +issue,37682,LSTMCell consumes x1.5 more memory on CUDA on pytorch >=1.3 comparing to pytorch 1.2,https://github.com/pytorch/pytorch/issues/37682,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: rnn",scan when time allows,2020-10-10T11:28:21Z +issue,10172,"[Feature request] Batch eig/symeig functions (for small matrices, with CUDA)",https://github.com/pytorch/pytorch/issues/10172,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: linear algebra",scan when time allows,2020-10-10T10:19:35Z +issue,19685,Bad overload order for zeros_like,https://github.com/pytorch/pytorch/issues/19685,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pybind, module: tensor creation",scan when time allows,2020-10-10T10:19:03Z +issue,23382,Tensor from mmaped storage loads the entire file into memory,https://github.com/pytorch/pytorch/issues/23382,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-10-10T10:18:48Z +issue,26428,Remove TensorOptions logic from generated code,https://github.com/pytorch/pytorch/issues/26428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-10-10T10:18:36Z +issue,33040,[debatable] Better infer dtype in torch.as_tensor,https://github.com/pytorch/pytorch/issues/33040,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2020-10-10T10:17:48Z +issue,42597,Logsigmoid as chaining operation,https://github.com/pytorch/pytorch/issues/42597,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-10T10:16:20Z +issue,2180,Discrepancy in BCEWithLogitsLoss and ClassNLLLoss,https://github.com/pytorch/pytorch/issues/2180,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: loss",scan when time allows,2020-10-10T10:11:52Z +issue,1747,In-place bernoulli_ has more functionality than torch.bernoulli with output parameter,https://github.com/pytorch/pytorch/issues/1747,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-10-10T10:11:40Z +issue,1328,Unhelpful CrossEntropyLoss dimension error message,https://github.com/pytorch/pytorch/issues/1328,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking, module: loss",scan when time allows,2020-10-10T10:11:19Z +issue,20935,torch.full_like missing documentation for out input variable,https://github.com/pytorch/pytorch/issues/20935,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-10-10T10:08:04Z +issue,16599,Feature Request: Earth Mover's Distance Loss,https://github.com/pytorch/pytorch/issues/16599,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2020-10-10T10:04:39Z +issue,20323,Support size to `torch.normal`,https://github.com/pytorch/pytorch/issues/20323,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-10-10T10:03:41Z +issue,21457,downsampling with grid_sample doesn't match interpolate,https://github.com/pytorch/pytorch/issues/21457,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: vision",scan when time allows,2020-10-10T09:59:15Z +issue,21688,Batched Conv2d for sequence data,https://github.com/pytorch/pytorch/issues/21688,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: convolution",scan when time allows,2020-10-10T09:58:47Z +issue,21856,Batch Normalization axis,https://github.com/pytorch/pytorch/issues/21856,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: norms and normalization",scan when time allows,2020-10-10T09:56:19Z +issue,22550,Eigen Tensor library for convolutions on CPU,https://github.com/pytorch/pytorch/issues/22550,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: convolution, module: cpu, module: performance",scan when time allows,2020-10-10T09:51:44Z +issue,22707,torch.fill_() exists and modifies the input tensor: Expected or bug?,https://github.com/pytorch/pytorch/issues/22707,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: deprecation, module: ux",scan when time allows,2020-10-10T09:48:36Z +issue,22330,Training CNNs with deconvolution,https://github.com/pytorch/pytorch/issues/22330,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: vision",scan when time allows,2020-10-10T09:47:52Z +issue,22487,"""Floating point exception"" after trying the method from the issue #22382",https://github.com/pytorch/pytorch/issues/22487,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-10-10T09:47:11Z +issue,22513,No assertion when using scatter_ on a non-contiguous tensor,https://github.com/pytorch/pytorch/issues/22513,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: partial aliasing, module: scatter & gather ops",scan when time allows,2020-10-10T09:46:19Z +issue,22803,Port `fmod` operator from the TH code to Aten,https://github.com/pytorch/pytorch/issues/22803,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: porting",scan when time allows,2020-10-10T09:45:17Z +issue,22919,Unify tensor shape formatting in shape checks,https://github.com/pytorch/pytorch/issues/22919,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: error checking",scan when time allows,2020-10-10T09:44:48Z +issue,23159,"Can't `torch.sum(tensor, dim)` where `dim >= 64`",https://github.com/pytorch/pytorch/issues/23159,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: reductions, module: tensoriterator",scan when time allows,2020-10-10T09:44:03Z +issue,23305,make torch.utils._download_url_from_file public and add a docstring,https://github.com/pytorch/pytorch/issues/23305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: hub, module: ux",scan when time allows,2020-10-10T09:43:51Z +issue,24031,fractional_max_pool2d_with_indices silently ignores output_ratio if output_size is provided,https://github.com/pytorch/pytorch/issues/24031,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: pooling",scan when time allows,2020-10-10T09:38:34Z +issue,24041,tensor.var_mean variant for existing torch.var_mean (and same for std_mean),https://github.com/pytorch/pytorch/issues/24041,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2020-10-10T09:38:10Z +issue,24155,"""To compact weights again call flatten_parameters()"" is printed every step for every GPU",https://github.com/pytorch/pytorch/issues/24155,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nn, module: rnn",scan when time allows,2020-10-10T09:37:30Z +issue,24261,"""PyTorch core"" thread local flag",https://github.com/pytorch/pytorch/issues/24261,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-10-10T09:33:50Z +issue,24303,Port `masked_fill` operator from the TH code to Aten,https://github.com/pytorch/pytorch/issues/24303,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: cuda, module: porting",scan when time allows,2020-10-10T09:31:53Z +issue,24313,Value_select to perform region-wise selection,https://github.com/pytorch/pytorch/issues/24313,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-10T09:31:46Z +issue,24168,torch.unique is inconsistent with NumPy's unique,https://github.com/pytorch/pytorch/issues/24168,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: numpy",scan when time allows,2020-10-10T09:23:17Z +issue,25039,Add new interpolation modes to `grid_sample`,https://github.com/pytorch/pytorch/issues/25039,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn",scan when time allows,2020-10-10T09:11:10Z +issue,25156,euclidean distance between two tensors,https://github.com/pytorch/pytorch/issues/25156,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distance functions,scan when time allows,2020-10-10T09:10:08Z +issue,25978,Provide a way to select SVD algorithm in PyTorch?,https://github.com/pytorch/pytorch/issues/25978,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2020-10-10T09:08:21Z +issue,28214,reflective padding for 5D tensor,https://github.com/pytorch/pytorch/issues/28214,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: padding,scan when time allows,2020-10-10T09:01:57Z +issue,28223,Convert manually bound `cuda` `cpu` `byte` `float` operators to native_functions,https://github.com/pytorch/pytorch/issues/28223,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2020-10-10T09:01:39Z +issue,27147,Stop binding in-place methods to `torch.*`,https://github.com/pytorch/pytorch/issues/27147,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2020-10-10T09:00:40Z +issue,27936,Specifying `pos_weight` in F.binary_cross_entropy_with_logits leads to RuntimeError: class size not match,https://github.com/pytorch/pytorch/issues/27936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: nn",scan when time allows,2020-10-10T08:59:38Z +issue,29510,torch.stack: bad shape error message,https://github.com/pytorch/pytorch/issues/29510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: error checking",scan when time allows,2020-10-10T08:57:56Z +issue,30263,pairwise_dist eps argument is confusing,https://github.com/pytorch/pytorch/issues/30263,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: docs, module: numerical-stability",scan when time allows,2020-10-10T08:53:25Z +issue,30565,`F.interpolate` returns unexpected result when dealing with output size `1`,https://github.com/pytorch/pytorch/issues/30565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: interpolation, module: nn, module: vision",scan when time allows,2020-10-10T08:53:06Z +issue,31261,pinned memory requires DeviceGuard in multi-process envs,https://github.com/pytorch/pytorch/issues/31261,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: multiprocessing",scan when time allows,2020-10-10T08:47:51Z +issue,31458,Integrate `torch.xxx` and `Tensor.xxx`,https://github.com/pytorch/pytorch/issues/31458,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2020-10-10T08:46:34Z +issue,31474,Slow clip_grad_norm_ because of .item() calls when run on device,https://github.com/pytorch/pytorch/issues/31474,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: norms and normalization, module: performance",scan when time allows,2020-10-10T08:46:05Z +issue,31655,pytorch out of memory when calculation squared difference of unfold,https://github.com/pytorch/pytorch/issues/31655,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-10-10T08:44:29Z +issue,31679,broadcast randperm with dim specification,https://github.com/pytorch/pytorch/issues/31679,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-10T08:44:17Z +issue,31776,Documentation for `scatter` incorrectly states that index values must be unique,https://github.com/pytorch/pytorch/issues/31776,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: docs, module: scatter & gather ops",scan when time allows,2020-10-10T08:43:33Z +issue,32549,"""Tried to register multiple operators with the same name and the same overload name"" error is confusing",https://github.com/pytorch/pytorch/issues/32549,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-10-10T08:29:47Z +issue,32614,Logical AND and OR for Tensors in C++ API.,https://github.com/pytorch/pytorch/issues/32614,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-10-10T08:02:25Z +issue,32872,Confusing error messages of tensor.scatter_ on both CPU and CUDA,https://github.com/pytorch/pytorch/issues/32872,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: error checking, module: scatter & gather ops",scan when time allows,2020-10-10T07:58:10Z +issue,33035,Support batch linear transformation,https://github.com/pytorch/pytorch/issues/33035,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: batching, module: nn",scan when time allows,2020-10-10T07:56:44Z +issue,33041,bytearray(tensor) behaves very differently from bytearray(tensor.numpy()),https://github.com/pytorch/pytorch/issues/33041,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: printing",scan when time allows,2020-10-10T07:56:04Z +issue,33373,torch.clamp_ not inplace during backward,https://github.com/pytorch/pytorch/issues/33373,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",scan when time allows,2020-10-10T07:42:29Z +issue,33571,empty_sparse shouldn't be called with memory layout but is,https://github.com/pytorch/pytorch/issues/33571,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, module: sparse, module: tensor creation",scan when time allows,2020-10-10T07:39:12Z +issue,34544,Strange behaviour of F.interpolate with bicubic mode.,https://github.com/pytorch/pytorch/issues/34544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-10T07:34:24Z +issue,34576,Significant speed difference between P100 and V100,https://github.com/pytorch/pytorch/issues/34576,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda, module: performance",scan when time allows,2020-10-10T07:33:44Z +issue,34675,[feature request]Support dilation parameter for unfold2d_* function (slow cpu maxpool2d #28733),https://github.com/pytorch/pytorch/issues/34675,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, module: nn, module: performance, module: pooling",scan when time allows,2020-10-10T07:30:42Z +issue,34874,Extend nn.functional.softmax for arbitrary dimensions,https://github.com/pytorch/pytorch/issues/34874,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-10T07:28:46Z +issue,35553,caffe2 `DEPTHWISE3x3.Conv` test is broken,https://github.com/pytorch/pytorch/issues/35553,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-10-10T07:25:21Z +issue,37863,Deprecate type() and type_as() call,https://github.com/pytorch/pytorch/issues/37863,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: ux",scan when time allows,2020-10-10T07:15:53Z +issue,38394,4D tensor support for torch.nn.functionnal.fold() (col2im),https://github.com/pytorch/pytorch/issues/38394,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-10T07:15:36Z +issue,35775,Unexpected behaviour for affine_grid and grid_sample with 3D inputs,https://github.com/pytorch/pytorch/issues/35775,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: numerical-stability",scan when time allows,2020-10-10T07:10:45Z +issue,46035,"Error installing on source: ""src.cxx:1:10: fatal error: glog/stl_logging.h: No such file or directory""",https://github.com/pytorch/pytorch/issues/46035,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-10T04:41:13Z +issue,41916,ASAN build broken when using USE_ASAN=1,https://github.com/pytorch/pytorch/issues/41916,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2020-10-09T19:49:14Z +issue,41918,ASAN build breaks when using third_party/protobuf,https://github.com/pytorch/pytorch/issues/41918,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power, module: protobuf",scan when time allows,2020-10-09T19:49:10Z +issue,40095,Bump up NCCL to 2.7.3,https://github.com/pytorch/pytorch/issues/40095,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2020-10-09T00:20:04Z +issue,31913,`index_add_` with multidimensional index,https://github.com/pytorch/pytorch/issues/31913,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: advanced indexing,scan when time allows,2020-10-08T21:19:07Z +issue,45690,Complex and real results do not agree when computing reciprocal or pow(-1) of 0,https://github.com/pytorch/pytorch/issues/45690,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2020-10-08T20:18:01Z +issue,45851,The same function can have different signatures in the torch and torch.nn.functional namespaces,https://github.com/pytorch/pytorch/issues/45851,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-08T16:55:50Z +issue,26354,torch.tensor / torch.as_tensor not working with list of tensors,https://github.com/pytorch/pytorch/issues/26354,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: numpy, module: tensor creation",scan when time allows,2020-10-08T08:05:17Z +issue,26527,Multilinear map,https://github.com/pytorch/pytorch/issues/26527,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-08T08:03:08Z +issue,30658,"Dedicated inverse AdaptiveMaxPool1d operation (e.g., AdaptiveMaxUnpool1d)",https://github.com/pytorch/pytorch/issues/30658,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling, module: vision",scan when time allows,2020-10-08T07:54:17Z +issue,30934,Spurious negative output in convolution of positive tensors,https://github.com/pytorch/pytorch/issues/30934,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: docs",scan when time allows,2020-10-08T07:50:52Z +issue,39632,F.affine_grid dispatch async issue,https://github.com/pytorch/pytorch/issues/39632,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn, module: performance",scan when time allows,2020-10-08T07:48:31Z +issue,40419,Add a `like` argument to creation ops,https://github.com/pytorch/pytorch/issues/40419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2020-10-08T07:47:35Z +issue,41793,General reduction mode selection for in-place and out-variants for wider range (hopefully all) of ops,https://github.com/pytorch/pytorch/issues/41793,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: reductions,scan when time allows,2020-10-08T07:45:55Z +issue,42902,Avoid dynamic isCustomClassRegistered() checks in kernel call paths,https://github.com/pytorch/pytorch/issues/42902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-10-08T07:43:48Z +issue,27276,Add in-place view (view_),https://github.com/pytorch/pytorch/issues/27276,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2020-10-08T07:42:10Z +issue,13218,Support for integer interpolation (torch.nn.functional.interpolate),https://github.com/pytorch/pytorch/issues/13218,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-08T07:37:15Z +issue,18987,Performance issue of unique on CPU,https://github.com/pytorch/pytorch/issues/18987,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: sorting and selection",scan when time allows,2020-10-08T07:29:32Z +issue,19149,Make operators like logsumexp and cumsum operate over dimension 0 by default (or at least for 1D arrays),https://github.com/pytorch/pytorch/issues/19149,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy, module: reductions",scan when time allows,2020-10-08T07:28:28Z +issue,19150,Improve unit test coverage of torch.unique,https://github.com/pytorch/pytorch/issues/19150,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: sorting and selection, module: tests",scan when time allows,2020-10-08T07:27:42Z +issue,28090,[discussion] Smarter version of torch.reshape (can avoid realloc in some cases),https://github.com/pytorch/pytorch/issues/28090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: viewing and reshaping,scan when time allows,2020-10-08T07:16:16Z +issue,29116,torch.masked_fill missing out argument,https://github.com/pytorch/pytorch/issues/29116,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2020-10-08T07:07:03Z +issue,10735,Have all C++ modules expose a __file__ attribute,https://github.com/pytorch/pytorch/issues/10735,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-10-08T06:53:20Z +issue,36036,torch.multinomial is misnamed.,https://github.com/pytorch/pytorch/issues/36036,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: ux",scan when time allows,2020-10-08T06:42:17Z +issue,36309,"torch.ceil, torch.floor should accept a dtype argument",https://github.com/pytorch/pytorch/issues/36309,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: ux",scan when time allows,2020-10-08T06:40:42Z +issue,36314,[discussion] Refactor spectral_norm to use the newly merged lowrank solvers and proposal for Linear Algebra Cookbook page,https://github.com/pytorch/pytorch/issues/36314,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2020-10-08T06:36:53Z +issue,37160,Enhanced operator context when reporting errors,https://github.com/pytorch/pytorch/issues/37160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: shape checking,scan when time allows,2020-10-08T06:34:20Z +issue,17169,Global Second Order Pooling,https://github.com/pytorch/pytorch/issues/17169,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2020-10-08T03:51:49Z +issue,9570,[feature request] More options for Fractional Max Pooling,https://github.com/pytorch/pytorch/issues/9570,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2020-10-08T03:51:37Z +issue,3396,CUDA topk is slow for some input sizes,https://github.com/pytorch/pytorch/issues/3396,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance, module: sorting and selection",scan when time allows,2020-10-08T03:51:17Z +issue,27542,Make topk sort stable,https://github.com/pytorch/pytorch/issues/27542,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: sorting and selection",scan when time allows,2020-10-08T03:50:41Z +issue,15822,AvgPool2d doesn't test if kernel is smaller than input size,https://github.com/pytorch/pytorch/issues/15822,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: error checking, module: pooling",scan when time allows,2020-10-08T03:38:21Z +issue,18728,MaxPool with n-dimensional tensors,https://github.com/pytorch/pytorch/issues/18728,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: pooling",scan when time allows,2020-10-08T03:34:23Z +issue,6138,[feature request] SIGNUM an optimizer that takes the sign of gradient or momentum.,https://github.com/pytorch/pytorch/issues/6138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-10-07T20:51:41Z +issue,6165,Note about unusual stride situations in dev docs / make it easier to test for this in the library,https://github.com/pytorch/pytorch/issues/6165,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: memory format",scan when time allows,2020-10-07T20:50:27Z +issue,6328,[feature request] Include libomp support (macOS),https://github.com/pytorch/pytorch/issues/6328,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2020-10-07T20:46:53Z +issue,6350,worker assignments in torch.utils.dataloader.py,https://github.com/pytorch/pytorch/issues/6350,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-10-07T20:46:12Z +issue,41926,torch.exp() cannot be modified by an inplace operation,https://github.com/pytorch/pytorch/issues/41926,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2020-10-07T17:23:08Z +issue,31598,ATen not compiled with MKL support,https://github.com/pytorch/pytorch/issues/31598,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkl",scan when time allows,2020-10-07T17:10:14Z +issue,38550,Returning a tensor instead of a list in split and chunk,https://github.com/pytorch/pytorch/issues/38550,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-07T14:34:36Z +issue,45840,Error with DataParallel and dataclass,https://github.com/pytorch/pytorch/issues/45840,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-10-07T10:01:17Z +issue,42805,Unbuffered operation,https://github.com/pytorch/pytorch/issues/42805,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-10-07T09:47:36Z +issue,31710,AttributeError: module 'torch.distributed' has no attribute 'init_process_group' on torch 1.3 aarch64,https://github.com/pytorch/pytorch/issues/31710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-10-07T09:06:30Z +issue,18315,[FR] support default size (scalar) in torch.randint,https://github.com/pytorch/pytorch/issues/18315,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: tensor creation",scan when time allows,2020-10-07T05:24:16Z +issue,18634,Speed-up torch.cat on CPU,https://github.com/pytorch/pytorch/issues/18634,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2020-10-07T05:22:47Z +issue,42246,"`torch.tensor([1, 2], dtype=torch.int).fmod(torch.tensor(0, dtype=torch.float))` leads to ""RuntimeError: result type Float can't be cast to the desired output type Int""",https://github.com/pytorch/pytorch/issues/42246,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2020-10-07T05:16:47Z +issue,42529,renorm dim argument is extremely confusing,https://github.com/pytorch/pytorch/issues/42529,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: ux",scan when time allows,2020-10-07T05:15:16Z +issue,42625,F.grid_sample produces weird results on single-pixel images,https://github.com/pytorch/pytorch/issues/42625,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-07T05:11:38Z +issue,38470,[Feature] Option to have zeros/ones/full output tensor with zero strides,https://github.com/pytorch/pytorch/issues/38470,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2020-10-07T05:05:24Z +issue,39260,Naming inconsistency: padding_mode vs pad_mode + F.conv* docs ops miss padding_mode arg at all,https://github.com/pytorch/pytorch/issues/39260,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ux,scan when time allows,2020-10-07T04:59:05Z +issue,39224,Negative stride values in `as_strided`,https://github.com/pytorch/pytorch/issues/39224,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: internals, module: memory format",scan when time allows,2020-10-07T04:57:57Z +issue,39134,Crossentropy inconsistent results depending on tensor order,https://github.com/pytorch/pytorch/issues/39134,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: nn",scan when time allows,2020-10-07T04:57:34Z +issue,38955,fold variation,https://github.com/pytorch/pytorch/issues/38955,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-07T04:53:46Z +issue,39959,Different max_pool2d cpp signatures due to indices.,https://github.com/pytorch/pytorch/issues/39959,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pooling,scan when time allows,2020-10-07T04:45:45Z +issue,42858,"prelu_backward, hardshrink_backward shouldn't be a method",https://github.com/pytorch/pytorch/issues/42858,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-10-07T03:51:05Z +issue,41528,"torch.nn.functional.grid_sample()is doing bilinear interpolation when the input is 5D, i think the mode should add 'trilinear'",https://github.com/pytorch/pytorch/issues/41528,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-07T03:47:43Z +issue,45901,Add support for reading the whole file in from_file,https://github.com/pytorch/pytorch/issues/45901,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: serialization",scan when time allows,2020-10-06T21:25:51Z +issue,32210,"dist.send/recv ""IndexError: map::at"" error when bool tensors are used with mpi backend",https://github.com/pytorch/pytorch/issues/32210,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor,scan when time allows,2020-10-06T20:04:57Z +issue,45044,torch rpc cannot handle UnsupportedNodeError exception,https://github.com/pytorch/pytorch/issues/45044,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2020-10-06T19:41:59Z +issue,5272,Unsafe out= keyword argument with tensors sharing storage,https://github.com/pytorch/pytorch/issues/5272,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: correctness (silent), module: numpy, module: safe resize",scan when time allows,2020-10-06T19:05:35Z +issue,30459,[TensorBoard] Graph with objects other than torch.nn.Module can not be visualized.,https://github.com/pytorch/pytorch/issues/30459,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-10-06T14:04:49Z +issue,2129,[Feature request] truncated normal initializer(sampler),https://github.com/pytorch/pytorch/issues/2129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: initialization,scan when time allows,2020-10-06T10:05:28Z +issue,44917,Incorrect documentation of SGD momentum,https://github.com/pytorch/pytorch/issues/44917,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2020-10-06T09:49:18Z +issue,44968,[feature request] dtype argument for torch.sign,https://github.com/pytorch/pytorch/issues/44968,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-10-06T09:49:08Z +issue,44969,Tensordot does not support Bool,https://github.com/pytorch/pytorch/issues/44969,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor, module: numpy",scan when time allows,2020-10-06T09:48:11Z +issue,45444,Using operator[] in GenericPackedTensorAccessor impossible in cpu code in .cu files (tensor on cpu),https://github.com/pytorch/pytorch/issues/45444,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-06T09:46:41Z +issue,22447,Weak Symbols Resolution Causes Segmentation Fault in External Libraries,https://github.com/pytorch/pytorch/issues/22447,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-06T09:38:32Z +issue,38197,Why no `torch.randperm_like`?,https://github.com/pytorch/pytorch/issues/38197,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensor creation,scan when time allows,2020-10-06T09:37:13Z +issue,2400,Add CRF Layer,https://github.com/pytorch/pytorch/issues/2400,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-10-06T09:35:45Z +issue,2629,[feature request] add pairwise ranking loss,https://github.com/pytorch/pytorch/issues/2629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2020-10-06T09:35:32Z +issue,43544,AMP much worse performance with groupped Conv2d than fp32,https://github.com/pytorch/pytorch/issues/43544,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cuda, module: dependency bug, module: performance",scan when time allows,2020-10-06T07:33:02Z +issue,38034,Support Slicing of ModuleList during JIT model tracing/scripting,https://github.com/pytorch/pytorch/issues/38034,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",scan when time allows,2020-10-06T06:46:22Z +issue,44295,Request: Add PyTorch version to state dicts,https://github.com/pytorch/pytorch/issues/44295,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-10-06T04:01:15Z +issue,45767,`defineMethodsInModule` no longer exists,https://github.com/pytorch/pytorch/issues/45767,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-10-05T20:52:55Z +issue,14996,as_tensor does not use the device of the default tensor type,https://github.com/pytorch/pytorch/issues/14996,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2020-10-05T17:04:56Z +issue,41622,New Feature : A very fast algorithm for computing matrix rank,https://github.com/pytorch/pytorch/issues/41622,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: linear algebra, module: numpy",scan when time allows,2020-10-05T06:46:35Z +issue,18094,[JIT Script] Need support on distributions.Categorical.,https://github.com/pytorch/pytorch/issues/18094,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-10-04T12:13:52Z +issue,45717,Better documentation of vec256 API,https://github.com/pytorch/pytorch/issues/45717,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: vectorization",scan when time allows,2020-10-02T15:27:35Z +issue,13865,[c10d] Coordinated file truncation for FileStore,https://github.com/pytorch/pytorch/issues/13865,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-10-01T22:43:20Z +issue,16402,error with nccl when distributed training on caffe2,https://github.com/pytorch/pytorch/issues/16402,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-10-01T22:16:52Z +issue,16318,Pytorch with CUDA aware OpenMPI for Infiniband not working with HCOLL and MXM,https://github.com/pytorch/pytorch/issues/16318,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, oncall: distributed",scan when time allows,2020-10-01T22:16:10Z +issue,14372,[c10d] Configurable timeout per operation for MPI backend,https://github.com/pytorch/pytorch/issues/14372,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-10-01T22:07:37Z +issue,10800,[distributed] Synchronization on CUDA side with MPI backend,https://github.com/pytorch/pytorch/issues/10800,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: c10d, module: docs, oncall: distributed",scan when time allows,2020-10-01T22:01:25Z +issue,45573,Build Failed Ubuntu Cuda. CUDAHooks.cpp:97:15: error: ‘struct cudaPointerAttributes’ has no member named ‘type’ return attr.type == cudaMemoryTypeHost;,https://github.com/pytorch/pytorch/issues/45573,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-10-01T07:21:40Z +issue,41329,Reference the randomness issue in DataLoader & Dataset documentation.,https://github.com/pytorch/pytorch/issues/41329,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: docs",scan when time allows,2020-09-30T19:45:13Z +issue,38541,Doc update regarding predictability of experiments using Seeds and Workers,https://github.com/pytorch/pytorch/issues/38541,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: determinism, module: docs",scan when time allows,2020-09-30T19:44:52Z +issue,23061,The speed of `torch.einsum` and `torch.matmul` when using `fp16` is slow,https://github.com/pytorch/pytorch/issues/23061,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",scan when time allows,2020-09-30T19:15:14Z +issue,45569,Distutils Error in torch.hub Load(),https://github.com/pytorch/pytorch/issues/45569,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2020-09-30T15:58:30Z +issue,45540,Why does NLLLoss and CrossEntropyLoss require type long?,https://github.com/pytorch/pytorch/issues/45540,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-09-30T15:48:02Z +issue,45560,Android: allow preallocation of output buffers,https://github.com/pytorch/pytorch/issues/45560,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: mobile",scan when time allows,2020-09-30T15:45:55Z +issue,40916,Vectorized torch.eig(),https://github.com/pytorch/pytorch/issues/40916,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: vectorization",scan when time allows,2020-09-30T14:41:43Z +issue,45481,Combining add_scalar with add_hparams with different frequencey,https://github.com/pytorch/pytorch/issues/45481,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-09-29T20:28:27Z +issue,45483,Extra arguments included in the doc where they are not actually presented in the source code,https://github.com/pytorch/pytorch/issues/45483,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2020-09-29T20:27:47Z +issue,43347,backward for dense+sparse does not work,https://github.com/pytorch/pytorch/issues/43347,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2020-09-28T20:29:54Z +issue,45399,Load ExtraFilesMap with Java library,https://github.com/pytorch/pytorch/issues/45399,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: java,scan when time allows,2020-09-28T16:18:37Z +issue,6071,Repeated 'python setup.py install' with clang leads to -lcpuinfo not found,https://github.com/pytorch/pytorch/issues/6071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-09-28T15:29:23Z +issue,31570,Can't pin storage memory,https://github.com/pytorch/pytorch/issues/31570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2020-09-26T08:58:05Z +issue,45308,Blocked version of Cholesky backward,https://github.com/pytorch/pytorch/issues/45308,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: linear algebra",scan when time allows,2020-09-25T14:41:31Z +issue,44943,In pytorch 1.6。Run model with input no contiguous tensor will become very slow.,https://github.com/pytorch/pytorch/issues/44943,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-09-24T17:49:47Z +issue,45053,about benchmark issue,https://github.com/pytorch/pytorch/issues/45053,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2020-09-24T06:58:57Z +issue,44724,16-bit input + AMP breaks AdaptiveLogSoftmaxWithLoss,https://github.com/pytorch/pytorch/issues/44724,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2020-09-23T21:40:51Z +issue,5924,Install doesn't work with spaces in directory,https://github.com/pytorch/pytorch/issues/5924,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-09-23T19:49:54Z +issue,44504,FP16 inference latency after sleeping,https://github.com/pytorch/pytorch/issues/44504,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2020-09-23T01:26:54Z +issue,43646,c10::string_view pybind11 custom type caster.,https://github.com/pytorch/pytorch/issues/43646,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2020-09-22T22:17:18Z +issue,45129,Inconsistent wheel name in https://download.pytorch.org/whl/torch_stable.html,https://github.com/pytorch/pytorch/issues/45129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-09-22T16:25:52Z +issue,23730,Failed to build pytorch ...,https://github.com/pytorch/pytorch/issues/23730,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-09-22T10:15:16Z +issue,44948,Clarify use of GLog/GFlags,https://github.com/pytorch/pytorch/issues/44948,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: build warnings, module: dependency bug",scan when time allows,2020-09-22T07:08:27Z +issue,43502,Useless Exception traces when DataSet timing out,https://github.com/pytorch/pytorch/issues/43502,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: dataloader",scan when time allows,2020-09-21T18:56:52Z +issue,36333,Illegal memory access / CUDNN_STATUS_MAPPING_ERROR on Quadro 8000,https://github.com/pytorch/pytorch/issues/36333,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2020-09-21T06:59:38Z +issue,43815,Buffers are moved to another device by copy and not in-place,https://github.com/pytorch/pytorch/issues/43815,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-09-18T21:54:14Z +issue,44631,Multi-process Dataloader and multi-parameter exceptions,https://github.com/pytorch/pytorch/issues/44631,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-09-18T14:58:36Z +issue,44898,Don't query current device on stream construction,https://github.com/pytorch/pytorch/issues/44898,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2020-09-18T14:09:32Z +issue,43510,Make busy waiting time in DDP (kSynchronizeBusyWaitMillis) a configurable environment variable,https://github.com/pytorch/pytorch/issues/43510,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-09-17T22:58:43Z +issue,13447,Support gathering nested lists in DataParallel,https://github.com/pytorch/pytorch/issues/13447,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-09-17T07:16:34Z +issue,44817,ComplexHelper.h contains non-inline functions,https://github.com/pytorch/pytorch/issues/44817,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2020-09-17T03:54:48Z +issue,44771,Timed out RRef can still be used in subsequent RPCs,https://github.com/pytorch/pytorch/issues/44771,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-09-17T00:28:48Z +issue,42969,Enable torch.testing._internal typechecks during CI,https://github.com/pytorch/pytorch/issues/42969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2020-09-16T22:56:55Z +issue,44415,Batchnorm2d in PT1.6 couldn't turn off track_running_stats,https://github.com/pytorch/pytorch/issues/44415,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2020-09-16T22:12:11Z +issue,30402,There is no support for `weight_decay`/`momentum` in SGD for sparse tensors.,https://github.com/pytorch/pytorch/issues/30402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-09-16T14:39:16Z +issue,44741,cblas_gemv is not being used for gemv on complex on CPU,https://github.com/pytorch/pytorch/issues/44741,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2020-09-15T22:27:09Z +issue,44547,floating point exception (core dumped) in training 10^4 steps,https://github.com/pytorch/pytorch/issues/44547,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-09-15T20:03:25Z +issue,44704,Error message concerning dtype GRU not identical between devices,https://github.com/pytorch/pytorch/issues/44704,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2020-09-15T17:44:01Z +issue,44719,404 Page not found - inference_api.md & management_api.md,https://github.com/pytorch/pytorch/issues/44719,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2020-09-15T17:39:13Z +issue,44721,Get rid of copy_from,https://github.com/pytorch/pytorch/issues/44721,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-09-15T17:23:52Z +issue,31856,Fix softplus clampling issues using logsigmoid,https://github.com/pytorch/pytorch/issues/31856,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-09-15T12:52:14Z +issue,42428,KeyError: xxxxxxxxxx when calling optimizer.state_dict(),https://github.com/pytorch/pytorch/issues/42428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-09-15T05:30:41Z +issue,40400,New ProcessGroups created with dist.new_group may leak memory,https://github.com/pytorch/pytorch/issues/40400,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, oncall: distributed",scan when time allows,2020-09-14T23:44:14Z +issue,44657,Use macro to define `DispatchKey::toString` method,https://github.com/pytorch/pytorch/issues/44657,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-09-14T21:07:49Z +issue,27839,RuntimeError: invalid argument 0: Sizes of tensors must match except in dimension 1. Got 24 and 195 in dimension 0 at /opt/conda/conda-bld/pytorch_1565272271120/work/aten/src/TH/generic/THTensor.cpp:689,https://github.com/pytorch/pytorch/issues/27839,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, oncall: jit",scan when time allows,2020-09-14T20:04:10Z +issue,44611,MacOS CPU torch.tan and torch.tanh do not compute some values properly,https://github.com/pytorch/pytorch/issues/44611,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cpu, module: macos",scan when time allows,2020-09-14T19:37:34Z +issue,44624,CUDA RuntimeError unrecoverably bricks session,https://github.com/pytorch/pytorch/issues/44624,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-09-14T17:24:32Z +issue,43638,[FR] hub uses default github branch,https://github.com/pytorch/pytorch/issues/43638,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2020-09-14T17:24:02Z +issue,29010,torch.stack does not yet support named tensors,https://github.com/pytorch/pytorch/issues/29010,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-09-14T17:03:51Z +issue,44552,torch.norm does not broadcast for dim > 2,https://github.com/pytorch/pytorch/issues/44552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-09-14T16:45:37Z +issue,44528,DataLoader consumes extremely large shared memory (shm) in its initialization.,https://github.com/pytorch/pytorch/issues/44528,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-09-13T18:53:02Z +issue,19160,Suggest model.eval() in torch.no_grad (and vice versa),https://github.com/pytorch/pytorch/issues/19160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-09-13T16:14:00Z +issue,44545,too large data in Queue cause dead lock in Multiprocessing,https://github.com/pytorch/pytorch/issues/44545,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2020-09-11T21:23:04Z +issue,32988,[jit] Dict set item type mismatch error doesn't say the type was inferred,https://github.com/pytorch/pytorch/issues/32988,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-09-10T10:36:32Z +issue,44403,Investigate SyncBatchNorm cleanup with NCCL Async Error Handling,https://github.com/pytorch/pytorch/issues/44403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-09-10T01:49:40Z +issue,39358,SyncBatchNorm doesn't scale well across multiple nodes on large data sizes,https://github.com/pytorch/pytorch/issues/39358,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: distributed",scan when time allows,2020-09-09T22:10:50Z +issue,44421,[tools.codegen] Rename api.legacy_dispatcher to api.native,https://github.com/pytorch/pytorch/issues/44421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-09-09T21:09:24Z +issue,44084,[RFC] Manage CUDA Stream in TensorPipe RPC Agent,https://github.com/pytorch/pytorch/issues/44084,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe",scan when time allows,2020-09-09T17:41:34Z +issue,40959,Will the model run slower when deployed using libtorch ?,https://github.com/pytorch/pytorch/issues/40959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: performance",scan when time allows,2020-09-09T16:59:52Z +issue,44169,Deadlock with RPC and dist.barrier() for TensorPipeAgent and NCCL.,https://github.com/pytorch/pytorch/issues/44169,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, module: rpc, module: tensorpipe",scan when time allows,2020-09-09T16:47:50Z +issue,44179,REINFORCEMENT LEARNING (DQN) TUTORIAL is not working anymore due to matplotlib.pyplot.imshow(),https://github.com/pytorch/pytorch/issues/44179,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-09-09T16:46:40Z +issue,44185,caffe2: Python 3 deprecation warnings about inspect.getargspec,https://github.com/pytorch/pytorch/issues/44185,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-09-09T16:45:35Z +issue,44313,`Undefined symbol: torch::kNearest` when building App iOS,https://github.com/pytorch/pytorch/issues/44313,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ios, oncall: mobile",scan when time allows,2020-09-09T16:26:33Z +issue,44341,Consider cache effects in Timer,https://github.com/pytorch/pytorch/issues/44341,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2020-09-09T16:21:11Z +issue,44362,Libtorch CMakeList.txt config help,https://github.com/pytorch/pytorch/issues/44362,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-09-09T16:12:06Z +issue,43012,Implement map-style caching DataSet as PyTorch build-in DataSet.,https://github.com/pytorch/pytorch/issues/43012,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: dataloader",scan when time allows,2020-09-08T18:07:15Z +issue,17313,Implement Adaptive Input Representations for Neural Language Modeling,https://github.com/pytorch/pytorch/issues/17313,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-09-08T14:52:33Z +issue,44274,Slower speeds when using half().,https://github.com/pytorch/pytorch/issues/44274,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2020-09-08T01:14:04Z +issue,42405,slow data loading in VisionDataset - need to allow batch loading.,https://github.com/pytorch/pytorch/issues/42405,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: performance, module: vision",scan when time allows,2020-09-08T00:03:59Z +issue,44103,Item seems to affect the backward process of DataParallel,https://github.com/pytorch/pytorch/issues/44103,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: data parallel",scan when time allows,2020-09-05T02:58:50Z +issue,23849,Make MultiProcessTestCase pickable,https://github.com/pytorch/pytorch/issues/23849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",scan when time allows,2020-09-04T17:52:32Z +issue,28929,torch::tensor(scalar) behaves differently from at::tensor(scalar),https://github.com/pytorch/pytorch/issues/28929,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: pybind",scan when time allows,2020-09-03T16:42:28Z +issue,38622,Clarification for usage of negative loss with optim.lr_scheduler.ReduceLROnPlateau,https://github.com/pytorch/pytorch/issues/38622,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-09-03T16:00:55Z +issue,5435,Gaussian Sampling,https://github.com/pytorch/pytorch/issues/5435,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-09-03T14:35:31Z +issue,5528,TestNN.test_data_parallel takes 10G of memory,https://github.com/pytorch/pytorch/issues/5528,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory usage, module: tests",scan when time allows,2020-09-03T14:32:30Z +issue,37334,Why do we not use TorchScript to build graph for tensorboard,https://github.com/pytorch/pytorch/issues/37334,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",scan when time allows,2020-09-03T13:22:53Z +issue,43947,torch.cuda.synchronize Influence distributed training,https://github.com/pytorch/pytorch/issues/43947,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: data parallel, module: performance",scan when time allows,2020-09-03T11:38:24Z +issue,39522,Valgrind leak checking flags losses in libtorch,https://github.com/pytorch/pytorch/issues/39522,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-09-03T07:54:07Z +issue,43964,Make it so that leading underscore operators are truly private and can be changed without worry for BC,https://github.com/pytorch/pytorch/issues/43964,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-09-02T21:28:13Z +issue,5524,Redo torch.nn.functional docstring strategy,https://github.com/pytorch/pytorch/issues/5524,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-09-01T17:40:58Z +issue,43816,Move randperm() to DistributionTemplates,https://github.com/pytorch/pytorch/issues/43816,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2020-08-31T22:54:25Z +issue,14200,Failed to run 'bash ../tools/build_pytorch_libs.sh --use-nnpack --use-mkldnn --use-qnnpack caffe2',https://github.com/pytorch/pytorch/issues/14200,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-31T22:50:57Z +issue,43904,Documentation and `torch.sparse` alias for `torch.bmm` sparse-dense,https://github.com/pytorch/pytorch/issues/43904,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-08-31T22:44:17Z +issue,42912,"Add typing information into variable declarations for JIT script "".code"" output",https://github.com/pytorch/pytorch/issues/42912,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-08-31T22:09:50Z +issue,41006,Simplify Adam Optimizer,https://github.com/pytorch/pytorch/issues/41006,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-08-30T11:54:01Z +issue,42311,Inconsistent results when trying to enable Tensor Cores on NVIDIA T4,https://github.com/pytorch/pytorch/issues/42311,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision), module: cudnn, module: memory format",scan when time allows,2020-08-29T15:35:04Z +issue,40790,Encoding dimension argument for F.one_hot,https://github.com/pytorch/pytorch/issues/40790,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: memory format",scan when time allows,2020-08-29T10:06:52Z +issue,5096,[feature request]Add an env variable to cover different pathes when testing code with openmp,https://github.com/pytorch/pytorch/issues/5096,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-08-28T23:20:04Z +issue,43758,OSError: /lib64/libc.so.6: version `GLIBC_2.14' not found,https://github.com/pytorch/pytorch/issues/43758,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2020-08-28T18:21:59Z +issue,33855,Unclear output for Pytorch Profiler,https://github.com/pytorch/pytorch/issues/33855,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2020-08-28T13:04:07Z +issue,43604,ImportError: libtorch_cpu.so: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/43604,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-27T23:29:55Z +issue,41478,Implement backend fallback for Tracer,https://github.com/pytorch/pytorch/issues/41478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-08-27T21:13:46Z +issue,43706,checkpoint restore of optimizers changes dtype of Floating-point state,https://github.com/pytorch/pytorch/issues/43706,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-08-27T15:49:51Z +issue,42177,record_function no longer appears in the docs,https://github.com/pytorch/pytorch/issues/42177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-27T15:20:07Z +issue,43692,named tensor INTERNAL ASSERT FAILED when indexing with a list.,https://github.com/pytorch/pytorch/issues/43692,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-08-27T13:54:47Z +issue,11937,[feature request] Kumaraswamy distribution,https://github.com/pytorch/pytorch/issues/11937,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-08-27T13:28:23Z +issue,43486,scatter not place inputs into every device,https://github.com/pytorch/pytorch/issues/43486,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-27T06:55:02Z +issue,43685,Syncbatchnorm and DDP,https://github.com/pytorch/pytorch/issues/43685,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2020-08-27T05:34:32Z +issue,43677,"Some largeCUDATensorTest fails with OOM when running with the entire test suit, but not when running standalone",https://github.com/pytorch/pytorch/issues/43677,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-08-27T00:03:04Z +issue,43663,test_doc_template is not working correctly,https://github.com/pytorch/pytorch/issues/43663,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-08-26T23:23:14Z +issue,43195,one_hot tensors are channels_last but marked as contiguous,https://github.com/pytorch/pytorch/issues/43195,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2020-08-26T21:02:31Z +issue,43617,Type hints from _VariableFunctions and elsewhere clash,https://github.com/pytorch/pytorch/issues/43617,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2020-08-26T20:32:55Z +issue,43561,[RFC] RPC BENCHMARK,https://github.com/pytorch/pytorch/issues/43561,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-08-26T04:07:47Z +issue,5326,TestMultiprocessing.test_fd_sharing hangs with ASAN,https://github.com/pytorch/pytorch/issues/5326,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-08-26T03:01:01Z +issue,5385,Handle python_arg_parser dtype constants better,https://github.com/pytorch/pytorch/issues/5385,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-26T02:59:18Z +issue,5212,Weird error message in torch.split_size_or_sections,https://github.com/pytorch/pytorch/issues/5212,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-08-24T21:53:24Z +issue,42950,TorchBind C++ Enum Class,https://github.com/pytorch/pytorch/issues/42950,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: torchbind,scan when time allows,2020-08-24T20:34:03Z +issue,43490,Implement a set_printoptions method in libtorch,https://github.com/pytorch/pytorch/issues/43490,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-08-24T20:19:00Z +issue,43503,torch.quantization.quantize_dynamic document refers `module` as a parameter,https://github.com/pytorch/pytorch/issues/43503,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-24T19:56:09Z +issue,43453,CUDAGuard might not create CUcontext,https://github.com/pytorch/pytorch/issues/43453,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-08-24T19:21:21Z +issue,5231,[feature request] Stratified splits in random_split function,https://github.com/pytorch/pytorch/issues/5231,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-08-24T16:25:26Z +issue,5266,[feature request] warnings for functions with unspecified dim arguments,https://github.com/pytorch/pytorch/issues/5266,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-24T16:07:47Z +issue,43459,Trying to build PyTorch from source with LLD 8 fails,https://github.com/pytorch/pytorch/issues/43459,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: static linking",scan when time allows,2020-08-24T13:36:07Z +issue,43411,Add support of random state generator objects to nn.init module,https://github.com/pytorch/pytorch/issues/43411,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: initialization, module: nn",scan when time allows,2020-08-22T18:36:26Z +issue,40736,torch::autograd::Function should set AutoNonVariableTypeMode when running forward,https://github.com/pytorch/pytorch/issues/40736,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-08-21T16:20:19Z +issue,43385,OffsetCalculator.cuh(and THCIntegerDivider.cuh) should be available with PyTorch cpu-only binaries,https://github.com/pytorch/pytorch/issues/43385,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-08-21T13:21:54Z +issue,43402,ModuleDict does not preserve order of initializing dictionary,https://github.com/pytorch/pytorch/issues/43402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-08-21T13:17:36Z +issue,38853,Weight decay in AdamW,https://github.com/pytorch/pytorch/issues/38853,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-08-21T04:37:11Z +issue,43325,CUDA error: an illegal memory access was encountered (laugh_kernel at ...../cuda/CUDALoops.cuh:112),https://github.com/pytorch/pytorch/issues/43325,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cublas, module: cuda",scan when time allows,2020-08-21T04:03:34Z +issue,43113,Missing caffe2_pybind11_state in pip install after cmake.,https://github.com/pytorch/pytorch/issues/43113,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: pybind",scan when time allows,2020-08-21T00:53:41Z +issue,43245,mp.spawn 'args' are not clear,https://github.com/pytorch/pytorch/issues/43245,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: multiprocessing",scan when time allows,2020-08-20T19:04:29Z +issue,43281,Cannot find CUDA devices when the machine stays idle for a while,https://github.com/pytorch/pytorch/issues/43281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-08-20T17:25:57Z +issue,43253,Reduce number of stack frames used up by dispatcher,https://github.com/pytorch/pytorch/issues/43253,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: internals",scan when time allows,2020-08-20T17:11:49Z +issue,43109,[docs] LR schedulers' get_last_lr missing in docs online,https://github.com/pytorch/pytorch/issues/43109,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-20T16:47:47Z +issue,43123,Add CutMix transform to torchvision.transforms,https://github.com/pytorch/pytorch/issues/43123,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-08-20T15:36:59Z +issue,43238,Expose counters for torch.distributed to report communication overhead within a block.,https://github.com/pytorch/pytorch/issues/43238,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-08-20T07:59:13Z +issue,24984,DataLoader slow down when `pin_memory=False`,https://github.com/pytorch/pytorch/issues/24984,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing, module: performance",scan when time allows,2020-08-19T10:06:59Z +issue,43116,Accessing tensor by element is super slow,https://github.com/pytorch/pytorch/issues/43116,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2020-08-18T15:52:18Z +issue,43140,"Pytorch ""Import torch"" lead to ""core dump""",https://github.com/pytorch/pytorch/issues/43140,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: protobuf",scan when time allows,2020-08-18T15:41:34Z +issue,43115,torch.multinomial with replacement=True produces inaccurate results for large number of categories,https://github.com/pytorch/pytorch/issues/43115,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: numerical-stability",scan when time allows,2020-08-18T14:10:08Z +issue,43091,Enable torch.optim typechecks during CI,https://github.com/pytorch/pytorch/issues/43091,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2020-08-18T13:46:05Z +issue,37048,copy_ slowness,https://github.com/pytorch/pytorch/issues/37048,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-08-18T02:11:32Z +issue,43072,LSTM::permute_hidden breaks Liskov substitution principle,https://github.com/pytorch/pytorch/issues/43072,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rnn, module: typing",scan when time allows,2020-08-17T17:17:26Z +issue,37585,Unable to use torch.det() inside nn.DataParallel with multiple gpus,https://github.com/pytorch/pytorch/issues/37585,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: data parallel, module: linear algebra",scan when time allows,2020-08-15T18:46:35Z +issue,43064,torch.dot throws an error for input tensors of different dtypes,https://github.com/pytorch/pytorch/issues/43064,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: type promotion",scan when time allows,2020-08-14T18:07:22Z +issue,43040,undefined symbol: FLAGS_caffe2_keep_on_shrink,https://github.com/pytorch/pytorch/issues/43040,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-14T17:47:09Z +issue,43030,BUILD_PYTHON switched to OFF at second run of CMake,https://github.com/pytorch/pytorch/issues/43030,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-14T16:51:32Z +issue,43033,No vectorization for int8 and uint8,https://github.com/pytorch/pytorch/issues/43033,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vectorization,scan when time allows,2020-08-14T16:50:16Z +issue,42879,Tensorpipe requires setting both GLOO_SOCKET_IFNAME and TP_SOCKET_IFNAME,https://github.com/pytorch/pytorch/issues/42879,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-08-14T09:01:00Z +issue,42916,Build problems caffe2 -- pytorch from source & CUDA 11.0,https://github.com/pytorch/pytorch/issues/42916,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-08-14T01:29:41Z +issue,11324,Why does DistributedDataSampler not use default RNG?,https://github.com/pytorch/pytorch/issues/11324,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-08-13T18:20:10Z +issue,42949,Verify TorchBind works with nested class type,https://github.com/pytorch/pytorch/issues/42949,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: torchbind,scan when time allows,2020-08-13T18:19:33Z +issue,42863,TensorOptions extensibility has rusted shut,https://github.com/pytorch/pytorch/issues/42863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2020-08-13T18:05:47Z +issue,42971,Enable torch.quantization.fuse_modules typechecks during CI,https://github.com/pytorch/pytorch/issues/42971,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2020-08-13T04:56:34Z +issue,42807,build error redefinition of ‘struct c10::complex’ with release v1.6.0,https://github.com/pytorch/pytorch/issues/42807,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-12T19:50:53Z +issue,41400,test_jit.py fails on Power,https://github.com/pytorch/pytorch/issues/41400,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-08-12T15:09:22Z +issue,42729,Mismatch in docs and behavior of align_corners for nn.functional.interpolate,https://github.com/pytorch/pytorch/issues/42729,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2020-08-12T10:15:38Z +issue,32047,Sobol point implementation,https://github.com/pytorch/pytorch/issues/32047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2020-08-11T19:25:35Z +issue,5070,NVIDIA_DRIVER_CAPABILITIES env variable is missing in pytorch docker images,https://github.com/pytorch/pytorch/issues/5070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2020-08-11T14:35:38Z +issue,5106,Saving model with runtime code changes,https://github.com/pytorch/pytorch/issues/5106,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-08-11T14:33:23Z +issue,5159,TakeBackward taking a significant portion of backward time,https://github.com/pytorch/pytorch/issues/5159,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: performance",scan when time allows,2020-08-11T14:16:49Z +issue,5161,ASAN detected leaks on python -c 'import torch',https://github.com/pytorch/pytorch/issues/5161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-08-11T14:15:51Z +issue,42848,Add support for user defined types in serialization in libtorch,https://github.com/pytorch/pytorch/issues/42848,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: serialization",scan when time allows,2020-08-11T14:08:25Z +issue,42663,Why Conv3D is slower than Conv2D when its flops is smaller than Conv2D,https://github.com/pytorch/pytorch/issues/42663,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: performance",scan when time allows,2020-08-11T01:51:50Z +issue,42800,RuntimeError: cublas runtime error : the GPU program failed to execute at /pytorch/aten/src/THC/THCBlas.cu:27 for spectral norm,https://github.com/pytorch/pytorch/issues/42800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2020-08-11T01:36:34Z +issue,40679,[JIT] Tracing BCE loss throws error when using weight,https://github.com/pytorch/pytorch/issues/40679,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-08-10T22:26:59Z +issue,42774,Missing API reference for SWALR and AveragedModel,https://github.com/pytorch/pytorch/issues/42774,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-10T22:12:45Z +issue,13234,torch.utils.checkpoint is not compatible with nn.DataParallel,https://github.com/pytorch/pytorch/issues/13234,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: data parallel",scan when time allows,2020-08-10T22:00:25Z +issue,42829,torch/library.h doc rendering issue,https://github.com/pytorch/pytorch/issues/42829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-10T21:50:01Z +issue,42830,Undocumented parameters in the Variables section of the recurrent bidirectional layers regarding the backward pass,https://github.com/pytorch/pytorch/issues/42830,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-10T21:49:00Z +issue,42818,Can't registered boxed kernel for operator that doesn't support boxing,https://github.com/pytorch/pytorch/issues/42818,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-08-10T21:47:59Z +issue,16542,Error with setting tensors to use cpu in packed_padded_sequence when CUDA tensor is set as default,https://github.com/pytorch/pytorch/issues/16542,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: rnn",scan when time allows,2020-08-08T00:12:55Z +issue,40882,JIT fuser result of dropout doesn't fully match eager mode,https://github.com/pytorch/pytorch/issues/40882,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-08-07T23:11:30Z +issue,39022,Cap DDP total number of buckets,https://github.com/pytorch/pytorch/issues/39022,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-08-07T22:21:46Z +issue,42705,torch.distributed.rpc package not work well with generator and lambda,https://github.com/pytorch/pytorch/issues/42705,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-08-07T21:07:57Z +issue,41639,Error with minimal hogwild test (multiprocessing shared memory),https://github.com/pytorch/pytorch/issues/41639,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2020-08-07T17:07:01Z +issue,42671,[discussion] Support __round__ magic,https://github.com/pytorch/pytorch/issues/42671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-08-07T08:42:01Z +issue,42350,An error occurred deploying CRNN using libtorch,https://github.com/pytorch/pytorch/issues/42350,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: rnn, oncall: jit",scan when time allows,2020-08-07T02:50:54Z +issue,42498,pytorch tests failed,https://github.com/pytorch/pytorch/issues/42498,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-08-07T01:45:04Z +issue,42526,Python3 segfaults or doesn't load torch after installation,https://github.com/pytorch/pytorch/issues/42526,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-06T11:13:54Z +issue,40027,[Android Pytorch] TorchScript traced model returns inconsistent output tensors on each run,https://github.com/pytorch/pytorch/issues/40027,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",scan when time allows,2020-08-06T10:10:08Z +issue,42655,"When one distributed test fails in CI, the next one can fail spuriously",https://github.com/pytorch/pytorch/issues/42655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests, oncall: distributed",scan when time allows,2020-08-06T04:06:32Z +issue,42183,Skip all reduce globally unused parameters in DDP,https://github.com/pytorch/pytorch/issues/42183,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2020-08-05T23:34:43Z +issue,42544,torch rpc lose device information between rpc calls,https://github.com/pytorch/pytorch/issues/42544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-08-05T22:01:11Z +issue,16322,We're binding a bunch of crap to 'torch' namespace which shouldn't be there,https://github.com/pytorch/pytorch/issues/16322,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-05T21:06:11Z +issue,42439,How to use libtorch on Jetson TX2,https://github.com/pytorch/pytorch/issues/42439,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: arm, module: binaries",scan when time allows,2020-08-05T21:03:37Z +issue,42545,CuDNN RNN bindings are basically all deprecated in cudnn 8,https://github.com/pytorch/pytorch/issues/42545,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: deprecation, module: rnn",scan when time allows,2020-08-05T17:03:51Z +issue,42366,Stop implementing JIT stack as a std::vector,https://github.com/pytorch/pytorch/issues/42366,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-08-05T14:35:10Z +issue,13746,[caffe2] test depthwise3x3_conv_op_test fails to run,https://github.com/pytorch/pytorch/issues/13746,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-05T14:25:31Z +issue,15294,manylinux2014 compatible wheels,https://github.com/pytorch/pytorch/issues/15294,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-04T18:11:36Z +issue,29554,[feature request] Print some measure of fragmentation at CUDA out-of-memory,https://github.com/pytorch/pytorch/issues/29554,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-08-04T18:07:15Z +issue,42539,Docs for Tensor.copy_ non_blocking should mention pinned memory,https://github.com/pytorch/pytorch/issues/42539,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-04T17:55:33Z +issue,42478,Using hooks with `DataParallel` gets `autograd` error,https://github.com/pytorch/pytorch/issues/42478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-08-04T14:40:27Z +issue,42188,Eliminate warning when cloning a tensor using `torch.tensor(x)`,https://github.com/pytorch/pytorch/issues/42188,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-08-03T17:25:20Z +issue,42447,OneCycleLR argument `pct_start` used as a proportion not percentage,https://github.com/pytorch/pytorch/issues/42447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2020-08-03T17:18:15Z +issue,42440,[feature request].Time Series Algorithms,https://github.com/pytorch/pytorch/issues/42440,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-03T17:12:50Z +issue,31565,Force libtorch to use CUDA context,https://github.com/pytorch/pytorch/issues/31565,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-08-03T12:44:56Z +issue,18101,Android NNAPI support of Caffe2,https://github.com/pytorch/pytorch/issues/18101,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-08-03T07:34:14Z +issue,41816,install path can set prefix install dir,https://github.com/pytorch/pytorch/issues/41816,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-02T20:01:55Z +issue,4952,[Feature request] Optimize autograd/ATen when a gradient is clearly zero,https://github.com/pytorch/pytorch/issues/4952,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-08-01T15:29:51Z +issue,42362,libpytorch_jni is not provided in conda builds,https://github.com/pytorch/pytorch/issues/42362,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-08-01T08:16:16Z +issue,4927,[Feature Request]would PackedSequence support unsorted sequences?,https://github.com/pytorch/pytorch/issues/4927,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-08-01T08:00:21Z +issue,4829,[Feature Request] clip_grad_norm for sparse gradients,https://github.com/pytorch/pytorch/issues/4829,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-07-31T21:40:59Z +issue,42345,Generate unique `worker_id` for each node in the RPC framework for the store.,https://github.com/pytorch/pytorch/issues/42345,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-07-31T01:02:36Z +issue,42301,Is it possible that one-version pytorch supports two gpu (GTX TITAN X and TITAN) using at the same time?,https://github.com/pytorch/pytorch/issues/42301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-07-30T21:04:56Z +issue,42295,KeyError: 'track_running_stats' in batchnorm.extra_repr,https://github.com/pytorch/pytorch/issues/42295,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: printing, module: serialization",scan when time allows,2020-07-30T20:56:36Z +issue,42316,Proper testing of `nn.Module` loading backward compatibility,https://github.com/pytorch/pytorch/issues/42316,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: serialization, module: tests",scan when time allows,2020-07-30T20:42:28Z +issue,42109,Slow `index_add_` on `torch.long` tensors,https://github.com/pytorch/pytorch/issues/42109,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-07-30T16:06:01Z +issue,36989,pytorch latest update(1.4) broke CosineAnnealingWarmRestarts: T_cur is not define,https://github.com/pytorch/pytorch/issues/36989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-07-30T11:43:26Z +issue,42228,[docs] Bitwise ops miss arguments,https://github.com/pytorch/pytorch/issues/42228,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-30T01:51:19Z +issue,42094,Setting threads number to the number of default by torch.set_num_threads is faster than not setting it,https://github.com/pytorch/pytorch/issues/42094,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2020-07-29T02:08:13Z +issue,29516,RuntimeError: CUDA error: invalid device ordinal,https://github.com/pytorch/pytorch/issues/29516,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-07-28T20:23:19Z +issue,42159,[docs] Add example of interop with DLPack created in plain C,https://github.com/pytorch/pytorch/issues/42159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-28T17:22:00Z +issue,41322,SGD documentatiuon detail on g_{t+1},https://github.com/pytorch/pytorch/issues/41322,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-07-28T13:53:36Z +issue,42156,"RuntimeError: test_autograd failed! Build success, test failed on IBM POWER9",https://github.com/pytorch/pytorch/issues/42156,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: power",scan when time allows,2020-07-28T13:18:16Z +issue,12773,"Build the docker image from source, but torch.cuda.is_available()==false",https://github.com/pytorch/pytorch/issues/12773,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2020-07-28T11:46:10Z +issue,42074,Memory usage of torch.nn.functional.interpolate increased with v1.5.0 when run on numpy input,https://github.com/pytorch/pytorch/issues/42074,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: nn, module: numpy",scan when time allows,2020-07-27T18:44:47Z +issue,41797,[C++] Libtorch error in Release build only,https://github.com/pytorch/pytorch/issues/41797,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: abi,scan when time allows,2020-07-27T17:26:11Z +issue,42112,Function names in title should not be capitalized in documentation,https://github.com/pytorch/pytorch/issues/42112,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-27T16:56:43Z +issue,42012,EMNIST looks different to MNIST,https://github.com/pytorch/pytorch/issues/42012,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-07-26T19:14:31Z +issue,42049,"Code hangs when using `set_start_method('spawn', force=True)` in `torch.multiprocessing.pool`",https://github.com/pytorch/pytorch/issues/42049,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2020-07-26T18:58:55Z +issue,7807,LMDB read error for Mnist,https://github.com/pytorch/pytorch/issues/7807,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-07-26T12:49:21Z +issue,41904,Inconsistencies on ModelLayer,https://github.com/pytorch/pytorch/issues/41904,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-07-25T14:15:17Z +issue,4758,Clang color diagnostics don't work with ninja,https://github.com/pytorch/pytorch/issues/4758,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-07-24T14:56:31Z +issue,41999,PyBind11 submodule required even when pybind11_PREFER_third_party=OFF,https://github.com/pytorch/pytorch/issues/41999,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-07-24T14:48:53Z +issue,32351,"Problem with multiprocessing, custom __getstate__ with Tensors and forkserver",https://github.com/pytorch/pytorch/issues/32351,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing, module: serialization",scan when time allows,2020-07-24T08:25:42Z +issue,41823,bitwise_or / bitwise_and /... reductions across dim (+ multidim / keepdim for consistency),https://github.com/pytorch/pytorch/issues/41823,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-07-24T08:18:16Z +issue,38457,Named Tensor and Indexing,https://github.com/pytorch/pytorch/issues/38457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-07-24T07:24:20Z +issue,41955,can not use nn::Functional(torch::softmax(-1)) in Sequential,https://github.com/pytorch/pytorch/issues/41955,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2020-07-24T03:57:42Z +issue,41938,Installing pytorch from source on Power9 (PPC64LE) + CUDA 10.2 + RHLE7,https://github.com/pytorch/pytorch/issues/41938,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2020-07-23T23:04:33Z +issue,41950,[docs] Clarify behavior of torch.cuda.device_count() when torch compiled without CUDA (cpu-only) or when CUDA is not available,https://github.com/pytorch/pytorch/issues/41950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs",scan when time allows,2020-07-23T22:41:55Z +issue,41844,test_bottleneck_cuda fails on Power,https://github.com/pytorch/pytorch/issues/41844,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: power, module: tests",scan when time allows,2020-07-23T22:36:50Z +issue,41903,[C++] adding type checking or type casting to torch::PackedTensorAccessor indexing,https://github.com/pytorch/pytorch/issues/41903,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-07-23T18:21:16Z +issue,41915,Automating the various Installation processes for Linux,https://github.com/pytorch/pytorch/issues/41915,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-07-23T18:17:55Z +issue,41531,Test failure in test_shared_allgather_nccl: NCCL error in: ../torch/lib/c10d/ProcessGroupNCCL.cpp:537,https://github.com/pytorch/pytorch/issues/41531,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: power, oncall: distributed",scan when time allows,2020-07-23T18:14:29Z +issue,41920,Compilation on Power fails with clang due to vec_xl,https://github.com/pytorch/pytorch/issues/41920,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2020-07-23T18:05:18Z +issue,41921,Failure to compile Eigen on Power with clang,https://github.com/pytorch/pytorch/issues/41921,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2020-07-23T18:04:17Z +issue,41673,"../caffe2/perfkernels/common_avx2.cc:17:2: error: #error ( ""You found a build system error: __AVX2__ is defined (via e.g. -mavx2) "" ""but CAFFE2_PERF_WITH_AVX2 is not defined."");",https://github.com/pytorch/pytorch/issues/41673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-07-23T03:12:33Z +issue,40971,The values calculated according to the document isn't equal to the values calculated by framework,https://github.com/pytorch/pytorch/issues/40971,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: optimizer",scan when time allows,2020-07-22T22:14:28Z +issue,41623,Can we have a way to reset a scheduler back to epoch -1,https://github.com/pytorch/pytorch/issues/41623,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-07-22T22:06:02Z +issue,41369,[RFC] Device Placement API for RPC,https://github.com/pytorch/pytorch/issues/41369,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe",scan when time allows,2020-07-22T14:33:56Z +issue,29710,Reliable way to identify RuntimeErrors (CUDA),https://github.com/pytorch/pytorch/issues/29710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-07-22T00:19:07Z +issue,41801,Caffe2 GPU Inference not working on pytorch nightly version,https://github.com/pytorch/pytorch/issues/41801,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-07-21T18:41:46Z +issue,37354,Efficient handling special gradient values in the autograd,https://github.com/pytorch/pytorch/issues/37354,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: performance",scan when time allows,2020-07-21T09:43:03Z +issue,41694,Torch.multiprocessing.spawn can deadlock,https://github.com/pytorch/pytorch/issues/41694,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2020-07-20T23:26:43Z +issue,41660,Potential bug in net_printer,https://github.com/pytorch/pytorch/issues/41660,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-07-20T23:24:13Z +issue,4716,Compilation issue: problem with GPU capability check,https://github.com/pytorch/pytorch/issues/4716,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-07-20T23:06:20Z +issue,4703,Very slow on CPU,https://github.com/pytorch/pytorch/issues/4703,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance, module: rnn",scan when time allows,2020-07-20T23:05:37Z +issue,41392,"About the description of the mathematical formula of nn.RNN, I think it is wrong",https://github.com/pytorch/pytorch/issues/41392,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-20T00:56:03Z +issue,41525,c++ indexing vs python,https://github.com/pytorch/pytorch/issues/41525,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-07-19T05:29:08Z +issue,41561,[RPC] Should we support users _not_ calling rpc.shutdown()?,https://github.com/pytorch/pytorch/issues/41561,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-07-17T20:18:43Z +issue,41530,Parallel computation of the diagonal of a Jacobian,https://github.com/pytorch/pytorch/issues/41530,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-07-17T16:31:35Z +issue,41243,[discussion] Unite nn modules nn.Something*d to just nn.Something with appropriate options,https://github.com/pytorch/pytorch/issues/41243,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-07-17T16:26:48Z +issue,41428,Add a done() API to torch.futures.Future and ProcessGroup::Work,https://github.com/pytorch/pytorch/issues/41428,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-07-17T03:19:13Z +issue,39426,Format issue in `torch.quantization.add_quant_dequant` documentation parameter section,https://github.com/pytorch/pytorch/issues/39426,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-16T20:23:13Z +issue,40936,[RFC] [RPC] Automatic retries of all requests in TensorPipe agent,https://github.com/pytorch/pytorch/issues/40936,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe",scan when time allows,2020-07-16T09:29:56Z +issue,41399,Bottleneck when publishing the model using flask about 3 times slower.,https://github.com/pytorch/pytorch/issues/41399,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2020-07-16T03:01:10Z +issue,4622,Protect user from No module named _C import error,https://github.com/pytorch/pytorch/issues/4622,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: pybind",scan when time allows,2020-07-16T02:25:59Z +issue,4564,Mixed Tensor/TensorList arguments in ATen functions with explicit derivatives,https://github.com/pytorch/pytorch/issues/4564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-07-15T15:10:53Z +issue,41383,Large overhead (7 microseconds) for PyTorch operation,https://github.com/pytorch/pytorch/issues/41383,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2020-07-15T14:24:39Z +issue,4636,Better header hygiene in ATen,https://github.com/pytorch/pytorch/issues/4636,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-07-15T14:19:52Z +issue,40215,Move all torch.Tensor methods to codegen,https://github.com/pytorch/pytorch/issues/40215,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: pybind",scan when time allows,2020-07-15T12:19:20Z +issue,4574,`from` keyword in `random_` gives error,https://github.com/pytorch/pytorch/issues/4574,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-07-14T23:14:07Z +issue,4551,Met 'cudnnDestroyDropoutDescriptor' while run multiply gpu-based models in multiply processes,https://github.com/pytorch/pytorch/issues/4551,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: multi-gpu",scan when time allows,2020-07-14T23:13:36Z +issue,4530,descriptor 'add' of 'torch._C._VariableBase' object needs an argument,https://github.com/pytorch/pytorch/issues/4530,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-14T23:13:22Z +issue,4542,Carefully audit contiguity requirements of code,https://github.com/pytorch/pytorch/issues/4542,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2020-07-14T23:11:12Z +issue,41284,add a reparameterized version of inverse Gaussian distribution,https://github.com/pytorch/pytorch/issues/41284,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-07-14T07:21:18Z +issue,41187,test_upsampling_not_recompute_scale_factor fails with Eigen/OpenBLAS,https://github.com/pytorch/pytorch/issues/41187,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-07-14T06:33:47Z +issue,38788,Convergence issues when using pytorch's native AMP,https://github.com/pytorch/pytorch/issues/38788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: amp (automated mixed precision),scan when time allows,2020-07-13T18:32:01Z +issue,41311,ConvTranspose1d layer behaviour under different channel numbers,https://github.com/pytorch/pytorch/issues/41311,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: docs",scan when time allows,2020-07-13T16:24:40Z +issue,41282,[docs] Strange arg names of torch.bmm/mm and tensor.bmm/mm,https://github.com/pytorch/pytorch/issues/41282,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-13T12:27:16Z +issue,34126,torch::jit::script::Object::attr should throw AttributeError instead of RuntimeError,https://github.com/pytorch/pytorch/issues/34126,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-07-13T08:46:06Z +issue,1641,[Feature request] In-place 'max' method for Tensor,https://github.com/pytorch/pytorch/issues/1641,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-07-12T12:17:42Z +issue,40969,Regarding graphs page on site,https://github.com/pytorch/pytorch/issues/40969,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-11T17:32:05Z +issue,24236,Tensorboard: Add disable flag for debugging,https://github.com/pytorch/pytorch/issues/24236,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-07-11T03:19:27Z +issue,13023,Python dataloader Improvements,https://github.com/pytorch/pytorch/issues/13023,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-07-10T22:36:54Z +issue,41101,DDP should provide an option for not touching grad of locally unused params,https://github.com/pytorch/pytorch/issues/41101,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-07-10T16:35:50Z +issue,41236,"""out of scope"" variable `cmd` used",https://github.com/pytorch/pytorch/issues/41236,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-07-10T15:29:14Z +issue,41246,torch.abs(complex) is divergent from NumPy on vectorized NaN values,https://github.com/pytorch/pytorch/issues/41246,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: nans and infs",scan when time allows,2020-07-10T15:27:40Z +issue,41244,torch.tan(complex) on CUDA doesn't handle nonfinite values properly,https://github.com/pytorch/pytorch/issues/41244,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: cuda",scan when time allows,2020-07-10T15:27:15Z +issue,40729,Custom c++ extension build process doesn't preserve color from compiler,https://github.com/pytorch/pytorch/issues/40729,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: cpp-extensions",scan when time allows,2020-07-10T15:25:05Z +issue,41178,Failure to build on Power9 due to FXDIV_SOURCE_DIR not being set,https://github.com/pytorch/pytorch/issues/41178,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: power",scan when time allows,2020-07-10T08:11:13Z +issue,40470,jit's default dtype is different in sandcastle and test_jit.py,https://github.com/pytorch/pytorch/issues/40470,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: jit",scan when time allows,2020-07-10T01:51:47Z +issue,32130,Idempotency Keys for RPC Retry,https://github.com/pytorch/pytorch/issues/32130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-07-09T22:41:51Z +issue,41148,Source code for some /_modules/.../.../ not displaying on https://pytorch.org/docs/master/_modules/,https://github.com/pytorch/pytorch/issues/41148,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-09T15:12:35Z +issue,41135,libtorch cmake should PUBLIC ally advertise it as requiring C++14,https://github.com/pytorch/pytorch/issues/41135,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-07-09T15:11:05Z +issue,37142,Propagation of channels-last layout leads to massive slowdowns in 1.5 compared to 1.4,https://github.com/pytorch/pytorch/issues/37142,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: memory format, module: performance",scan when time allows,2020-07-08T22:34:21Z +issue,4501,Bind in Python _backward ATen functions,https://github.com/pytorch/pytorch/issues/4501,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-07-08T20:47:02Z +issue,4400,Installation Optimise For Chinese Users Who Behind the Wall,https://github.com/pytorch/pytorch/issues/4400,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-08T05:25:08Z +issue,4406,DistributedDataParallel doesn't converge well when using MPI,https://github.com/pytorch/pytorch/issues/4406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-07-08T05:24:24Z +issue,17901,should disable AVX on 32bit x86 / refine AVX availability tests,https://github.com/pytorch/pytorch/issues/17901,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-07-08T03:49:20Z +issue,40563,"Compiling PyTorch with 11.0, V11.0.167",https://github.com/pytorch/pytorch/issues/40563,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-07-08T02:42:25Z +issue,41106,Missing Code for Audio Classification Tutorial,https://github.com/pytorch/pytorch/issues/41106,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-07-08T00:12:32Z +issue,14959,get/set device in c++,https://github.com/pytorch/pytorch/issues/14959,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-07-06T19:58:18Z +issue,40997,Is this a bug? The values calculated according to the document isn't equal to the values calculated by framework,https://github.com/pytorch/pytorch/issues/40997,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-07-06T16:31:27Z +issue,31883,"The dependency target ""nccl_external"" of target ""gloo_cuda"" does not exist.",https://github.com/pytorch/pytorch/issues/31883,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: distributed",scan when time allows,2020-07-06T12:15:59Z +issue,40886,Improve error messaging when using dictionaries as inputs to a trace,https://github.com/pytorch/pytorch/issues/40886,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-07-06T04:22:29Z +issue,40978,[FR] NCCL and bool type,https://github.com/pytorch/pytorch/issues/40978,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nccl",scan when time allows,2020-07-04T20:27:13Z +issue,32937,Operation Registration Error,https://github.com/pytorch/pytorch/issues/32937,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-07-04T04:39:46Z +issue,40974,Add a launching script for RPC,https://github.com/pytorch/pytorch/issues/40974,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-07-04T01:13:55Z +issue,40891,[FR] torch.floor/ceil should support output int dtype,https://github.com/pytorch/pytorch/issues/40891,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2020-07-03T21:24:45Z +issue,40854,Build pytorch with cuda11,https://github.com/pytorch/pytorch/issues/40854,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-07-02T22:36:28Z +issue,40880,Allow TLS to keep distributed autograd context alive,https://github.com/pytorch/pytorch/issues/40880,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-07-01T19:21:17Z +issue,31759,Allow using digits in names of named tensors,https://github.com/pytorch/pytorch/issues/31759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-07-01T04:12:01Z +issue,40456,Print values (but not strings) when STRIP_ERROR_MESSAGES is defined,https://github.com/pytorch/pytorch/issues/40456,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: internals",scan when time allows,2020-06-30T22:24:55Z +issue,40590,Inconsistent behavior between numpy.exp and torch.exp on CPU for complex numbers,https://github.com/pytorch/pytorch/issues/40590,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy",scan when time allows,2020-06-30T21:29:55Z +issue,35446,[JIT] dropout fails on legacy mode,https://github.com/pytorch/pytorch/issues/35446,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-30T17:19:24Z +issue,40441,"[feature request] batch_apply, a general-purpose device-agnostic batch iterator",https://github.com/pytorch/pytorch/issues/40441,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: batching,scan when time allows,2020-06-30T15:06:59Z +issue,40763,Enhance supported types of functional.pad,https://github.com/pytorch/pytorch/issues/40763,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: numpy",scan when time allows,2020-06-30T14:19:04Z +issue,40761,Sparse Convolutional support,https://github.com/pytorch/pytorch/issues/40761,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-06-30T14:16:40Z +issue,40756,RuntimeError: rois.device().is_cpu() ASSERT FAILED at /vision/torchvision/csrc/cpu/ROIAlign_cpu.cpp:386,https://github.com/pytorch/pytorch/issues/40756,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-06-30T14:15:09Z +issue,38057,DNNL's backward pass much slower when using nn.grad.conv2d_input and nn.grad.conv2d_weight,https://github.com/pytorch/pytorch/issues/38057,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: mkldnn, module: nn",scan when time allows,2020-06-30T07:14:53Z +issue,40633,RuntimeError: broken pipe from NCCL,https://github.com/pytorch/pytorch/issues/40633,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2020-06-29T23:44:42Z +issue,14461,The speed of scatter is influenced by the data size while using nn.DataParallel,https://github.com/pytorch/pytorch/issues/14461,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: dataloader, module: performance",scan when time allows,2020-06-29T20:37:58Z +issue,40687,libtorch: macros in logging_is_not_google_glog.h have very common names like CHECK or LOG,https://github.com/pytorch/pytorch/issues/40687,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: logging,scan when time allows,2020-06-29T20:32:32Z +issue,40667,Segmentation fault in forward pass using DataParallel and multiple GPUs,https://github.com/pytorch/pytorch/issues/40667,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: data parallel",scan when time allows,2020-06-29T15:25:27Z +issue,36386,Docs of distributed module do not include the full documentation for torch.distributed.launch,https://github.com/pytorch/pytorch/issues/36386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-06-28T06:32:39Z +issue,40568,Converting NumPy dtype to Torch dtype when using `as_tensor`,https://github.com/pytorch/pytorch/issues/40568,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-06-26T15:26:29Z +issue,40269,caffe2 error during compilation of PyTorch with ROCm on Archlinux,https://github.com/pytorch/pytorch/issues/40269,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-26T09:00:07Z +issue,40492,Learning rate change is not applied at designated iteration with a scheduler,https://github.com/pytorch/pytorch/issues/40492,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-06-26T03:08:25Z +issue,40564,Stop registering kernels that use DispatchStub as catch all,https://github.com/pytorch/pytorch/issues/40564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-06-25T21:49:51Z +issue,40566,"If all parameters are unused by forward pass in a process, backward will not work with DDP.",https://github.com/pytorch/pytorch/issues/40566,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-06-25T21:49:21Z +issue,40548,"Embedding with DataParallel can return ""incomplete"" results",https://github.com/pytorch/pytorch/issues/40548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: data parallel",scan when time allows,2020-06-25T15:42:31Z +issue,40355,Tracking output dimensions of the convolutional layers,https://github.com/pytorch/pytorch/issues/40355,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-06-25T14:35:47Z +issue,38911,[FR] cuda.comm.broadcast/reduce_add support `out=`,https://github.com/pytorch/pytorch/issues/38911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2020-06-24T20:13:33Z +issue,39444,Torch hub: object has no attribute nms,https://github.com/pytorch/pytorch/issues/39444,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docker,scan when time allows,2020-06-24T19:08:40Z +issue,40114,ppc64le: test cpp_extensions/rng_extension.cpp failure (without altivec override),https://github.com/pytorch/pytorch/issues/40114,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: power, module: random",scan when time allows,2020-06-24T19:08:33Z +issue,31232,Default shuffle behavior of DistributedSampler,https://github.com/pytorch/pytorch/issues/31232,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, oncall: distributed",scan when time allows,2020-06-24T17:16:08Z +issue,13786,PyTorch streams are not cuda-memcheck clean,https://github.com/pytorch/pytorch/issues/13786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-06-23T19:55:41Z +issue,40417,Missing explanation in torch.utils.tensorboard.add_histogram(),https://github.com/pytorch/pytorch/issues/40417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: tensorboard",scan when time allows,2020-06-23T19:38:08Z +issue,37556,"torch.cartesian_prod(*tensors) error when you have tensors with [x,y]",https://github.com/pytorch/pytorch/issues/37556,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: linear algebra,scan when time allows,2020-06-22T17:04:59Z +issue,39990,libtorch 1.5 macos crash when loading on some mac,https://github.com/pytorch/pytorch/issues/39990,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: mkldnn",scan when time allows,2020-06-22T14:58:21Z +issue,4123,Use the int64 version of MKL calls,https://github.com/pytorch/pytorch/issues/4123,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: mkl",scan when time allows,2020-06-19T21:38:04Z +issue,33085,not able to import * from fastai.vision in Google collab,https://github.com/pytorch/pytorch/issues/33085,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-19T20:20:26Z +issue,4159,Variable outputs of stochastic functions should never require grad,https://github.com/pytorch/pytorch/issues/4159,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-06-19T18:53:02Z +issue,4186,Feature Request: CPU performance optimization with MKL-DNN,https://github.com/pytorch/pytorch/issues/4186,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: mkldnn,scan when time allows,2020-06-19T18:25:37Z +issue,4181,Fused RNN refactor plan,https://github.com/pytorch/pytorch/issues/4181,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2020-06-19T18:25:12Z +issue,4148,ATen explicitly differentiated native function resolution hazard (call is ambiguous),https://github.com/pytorch/pytorch/issues/4148,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-06-19T18:23:40Z +issue,40266,CUDA error: out of memory when running tensorpipe test_cuda,https://github.com/pytorch/pytorch/issues/40266,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe",scan when time allows,2020-06-19T18:18:42Z +issue,40250,Include expanded TensorOptions version of op in at:: namespace,https://github.com/pytorch/pytorch/issues/40250,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2020-06-19T18:15:20Z +issue,36125,build Pytorch-1.5.0-rc1 from source fail,https://github.com/pytorch/pytorch/issues/36125,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-06-19T13:08:09Z +issue,37079,Result parameters from Single-Process Multi-GPU DDP training on RNN do not match local training,https://github.com/pytorch/pytorch/issues/37079,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-06-19T06:20:11Z +issue,40237,Move torch cpp Errors to c10::Error,https://github.com/pytorch/pytorch/issues/40237,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: internals",scan when time allows,2020-06-18T20:23:58Z +issue,40098,Updating learning rate with Libtorch 1.5 and optimiser options,https://github.com/pytorch/pytorch/issues/40098,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: optimizer",scan when time allows,2020-06-18T16:11:45Z +issue,40086,TestListwiseL2rOps::test_lambda_rank_loss fails,https://github.com/pytorch/pytorch/issues/40086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-06-18T14:16:16Z +issue,40030,matchTensor tests. Add tests to make sure that `isSubtypeOf` and `matchTensor` return equivalent results,https://github.com/pytorch/pytorch/issues/40030,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-16T22:49:59Z +issue,39356,Torch script to support dictionary with keys of type tuple,https://github.com/pytorch/pytorch/issues/39356,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-16T17:39:21Z +issue,39852,Shared file-system initialization in pytorch distributed is slow,https://github.com/pytorch/pytorch/issues/39852,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: distributed",scan when time allows,2020-06-16T07:32:49Z +issue,40020,Caffe2 operation switches current CUDA stream,https://github.com/pytorch/pytorch/issues/40020,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-06-16T03:27:08Z +issue,40071,[JIT][to-backend] Finalize the PyTorchBackendInterface API/naming for external sharing,https://github.com/pytorch/pytorch/issues/40071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-16T01:23:44Z +issue,40070,[JIT][to-backend] Support type refinement for container types in the generated code,https://github.com/pytorch/pytorch/issues/40070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-16T01:19:48Z +issue,39745,[JIT] exported dunder methods are ignored,https://github.com/pytorch/pytorch/issues/39745,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-15T20:54:40Z +issue,39806,Simplify layers of optionals in `VaryingShape` and `Stride`,https://github.com/pytorch/pytorch/issues/39806,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-15T20:53:37Z +issue,39719,User Objects aren't recursively scripted as nn.Module attributes,https://github.com/pytorch/pytorch/issues/39719,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-15T17:19:25Z +issue,39993,How do I derive weights for CrossEntropy Loss on my custom dataset?,https://github.com/pytorch/pytorch/issues/39993,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-06-15T14:37:49Z +issue,39972,Port old registration API to new one,https://github.com/pytorch/pytorch/issues/39972,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-15T13:49:00Z +issue,39947,[Caffe2] compilation in c++ undefined reference to `caffe2,https://github.com/pytorch/pytorch/issues/39947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-06-12T17:36:22Z +issue,4085,GridSampler behaviours,https://github.com/pytorch/pytorch/issues/4085,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2020-06-12T04:34:00Z +issue,4102,Make the generator tools data model more explicit,https://github.com/pytorch/pytorch/issues/4102,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: codegen,scan when time allows,2020-06-12T04:32:02Z +issue,32078,Reduce degrees of freedom in strides so that they unambiguously specify layout permutation,https://github.com/pytorch/pytorch/issues/32078,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-12T02:12:23Z +issue,35609,Randomly error reports,https://github.com/pytorch/pytorch/issues/35609,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-11T21:49:35Z +issue,37010,Make it an error to def() an operator multiple times,https://github.com/pytorch/pytorch/issues/37010,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bootcamp,scan when time allows,2020-06-11T19:46:47Z +issue,39836,Feedback to split certain doc pages into sub-topics,https://github.com/pytorch/pytorch/issues/39836,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-06-11T17:24:23Z +issue,39864,Parallelize arguments serde for RPC with TorchScript functions.,https://github.com/pytorch/pytorch/issues/39864,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: rpc, oncall: jit",scan when time allows,2020-06-11T16:13:55Z +issue,36572,[1.4.1] cmake3 not found,https://github.com/pytorch/pytorch/issues/36572,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-06-11T12:33:32Z +issue,32497,Don't unnecesarily send cleanup dist autograd context RPCs to other nodes,https://github.com/pytorch/pytorch/issues/32497,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: rpc",scan when time allows,2020-06-11T05:54:10Z +issue,39064,Add cpack support to CMakeLists.txt,https://github.com/pytorch/pytorch/issues/39064,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2020-06-11T01:43:07Z +issue,39799,Tensor.size() API not correctly documented?,https://github.com/pytorch/pytorch/issues/39799,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-06-10T21:25:01Z +issue,36570,Libtorch build error when setting both `USE_GLOO` and `USE_SYSTEM_NCCL` to `ON`,https://github.com/pytorch/pytorch/issues/36570,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs, module: nccl, oncall: distributed",scan when time allows,2020-06-10T12:48:03Z +issue,39463,JIT test suite has dependencies across tests,https://github.com/pytorch/pytorch/issues/39463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-09T18:52:59Z +issue,30373,Refactor (a bit) `torch.hub(.load)`,https://github.com/pytorch/pytorch/issues/30373,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: hub,scan when time allows,2020-06-09T18:31:31Z +issue,39706,[distributed] calling nccl reduce with inconsistent dst hangs,https://github.com/pytorch/pytorch/issues/39706,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: nccl, oncall: distributed",scan when time allows,2020-06-09T16:12:17Z +issue,13716,depthwise convolution are slow on cpu,https://github.com/pytorch/pytorch/issues/13716,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cpu, module: performance",scan when time allows,2020-06-09T16:06:05Z +issue,39705,[doc] [distributed] example of specifying url etc in url,https://github.com/pytorch/pytorch/issues/39705,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: distributed",scan when time allows,2020-06-09T15:46:55Z +issue,39682,Run `clang-tidy` on the `aten` folder?,https://github.com/pytorch/pytorch/issues/39682,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build warnings, module: lint",scan when time allows,2020-06-09T01:22:51Z +issue,39665,Better err msg for tensor ctor from sequence,https://github.com/pytorch/pytorch/issues/39665,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-09T00:28:55Z +issue,39603,friend constexpr in templated struct loses constexpr-ness in nvcc,https://github.com/pytorch/pytorch/issues/39603,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-08T17:07:54Z +issue,39226,Try Address Sanitizer in MSVC builds,https://github.com/pytorch/pytorch/issues/39226,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: windows,scan when time allows,2020-06-08T15:58:20Z +issue,39435,BCEWithLogitsLoss() not equal to BCELoss() with sigmoid(),https://github.com/pytorch/pytorch/issues/39435,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2020-06-06T22:27:24Z +issue,39625,All keys matched successfully missing when loading state dict on optimizers,https://github.com/pytorch/pytorch/issues/39625,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-06-06T21:45:56Z +issue,39619,Switching from CPU build to CUDA build in conda environments is a bit tricky,https://github.com/pytorch/pytorch/issues/39619,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-06-06T20:37:37Z +issue,29026,RuntimeError: !t.is_cuda() INTERNAL ASSERT FAILED at /pytorch/aten/src/ATen/native/sparse/SparseTensorMath.cpp:591,https://github.com/pytorch/pytorch/issues/29026,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: sparse",scan when time allows,2020-06-06T07:57:59Z +issue,39362,"The conversion of Inplace aten::append to outplace is incomplete, creating incorrect subgraphs",https://github.com/pytorch/pytorch/issues/39362,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-06-06T07:05:30Z +issue,39537,Segfault = docker + tensorboard + pytorch,https://github.com/pytorch/pytorch/issues/39537,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-06-06T00:08:12Z +issue,39123,The MacOS compiler is generating illegal instruction for the division of c10::complex,https://github.com/pytorch/pytorch/issues/39123,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2020-06-05T14:40:04Z +issue,39524,test_nn_module_tests should run less tests,https://github.com/pytorch/pytorch/issues/39524,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: tests",scan when time allows,2020-06-04T19:42:40Z +issue,27840,caffe2 install VS2019 CUDA 10.1 lib\torch.lib : fatal error LNK1248: image size (10028FA9F) exceeds maximum allowable size (FFFFFFFF),https://github.com/pytorch/pytorch/issues/27840,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-04T09:34:39Z +issue,38974,ATen registrable operator list,https://github.com/pytorch/pytorch/issues/38974,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-06-03T18:15:09Z +issue,39317,send a Tensor to Cuda very slow,https://github.com/pytorch/pytorch/issues/39317,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-06-03T15:19:00Z +issue,38643,torch while loading weighs found runtime error on storage has wrong size: expected 4254413747647032608 got 1024,https://github.com/pytorch/pytorch/issues/38643,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-06-03T00:39:12Z +issue,39399,Update RPC doc to recommend async user functions for non-blocking server execution,https://github.com/pytorch/pytorch/issues/39399,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rpc",scan when time allows,2020-06-02T15:05:07Z +issue,38973,ATen operator API versioning,https://github.com/pytorch/pytorch/issues/38973,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-06-02T00:55:33Z +issue,39355,Clean up GIL that used to guard deleted RRef destructions,https://github.com/pytorch/pytorch/issues/39355,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-06-01T21:37:10Z +issue,32162,Using Tensor.to(device) after distributed all_reduce intermittently causes deadlock with NCCL,https://github.com/pytorch/pytorch/issues/32162,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2020-06-01T19:46:44Z +issue,38233,When TorchScripted module has bad type annotation you get bad error message,https://github.com/pytorch/pytorch/issues/38233,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",scan when time allows,2020-06-01T19:04:35Z +issue,38677,TestCase.assertEqual does not distinguish Python builtin types and single-element Tensor,https://github.com/pytorch/pytorch/issues/38677,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-06-01T17:25:03Z +issue,39298,Caffe 2 android app not working with proguard and minifyEnabled,https://github.com/pytorch/pytorch/issues/39298,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-06-01T06:55:33Z +issue,39124,[TensorPipe] Errors in pipeWrite should clear out the future in pendingResponseMessage,https://github.com/pytorch/pytorch/issues/39124,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, module: tensorpipe",scan when time allows,2020-06-01T01:43:55Z +issue,39280,Let future.wait() take in an optional timeout,https://github.com/pytorch/pytorch/issues/39280,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-06-01T01:40:47Z +issue,39305,Compilation errors,https://github.com/pytorch/pytorch/issues/39305,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn",scan when time allows,2020-06-01T01:40:37Z +issue,39270,Current implementation of c10::complex does not support being used in shared memory,https://github.com/pytorch/pytorch/issues/39270,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2020-06-01T01:32:20Z +issue,39130,`torch.backends` undocumented,https://github.com/pytorch/pytorch/issues/39130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-06-01T01:24:25Z +issue,39082,[TensorPipe] Avoid wrapping the future message in order to do atomic test-and-set,https://github.com/pytorch/pytorch/issues/39082,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorpipe,scan when time allows,2020-06-01T01:23:23Z +issue,39101,Build issue when installing pytorch with USE_FFMPEG=1 USE_OPENCV=1,https://github.com/pytorch/pytorch/issues/39101,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-06-01T01:22:58Z +issue,38051,Div by zero error not triggered and inf not returned when dividing by 0 for some dtypes,https://github.com/pytorch/pytorch/issues/38051,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numerical-stability, module: type promotion",scan when time allows,2020-05-31T17:59:52Z +issue,30458,add method to make tensor constant for debug purposes,https://github.com/pytorch/pytorch/issues/30458,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-05-29T09:26:41Z +issue,39128,"Problems implementing complex support for acosh, asinh, tanh",https://github.com/pytorch/pytorch/issues/39128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: complex,scan when time allows,2020-05-28T14:56:17Z +issue,38239,Failed to link torch_library using cmake,https://github.com/pytorch/pytorch/issues/38239,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2020-05-28T09:05:27Z +issue,38975,Use general ATen dispatch mechanism,https://github.com/pytorch/pytorch/issues/38975,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-05-28T00:19:13Z +issue,38996,[FR] API consistency for cuda.comm and distributed,https://github.com/pytorch/pytorch/issues/38996,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, oncall: distributed",scan when time allows,2020-05-26T19:34:05Z +issue,38997,[doc] document cuda.nccl,https://github.com/pytorch/pytorch/issues/38997,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: docs, module: nccl",scan when time allows,2020-05-26T19:33:57Z +issue,38980,runtime error while using default arguments in add_graph() function,https://github.com/pytorch/pytorch/issues/38980,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensorboard, oncall: visualization",scan when time allows,2020-05-26T18:55:28Z +issue,38863,[JIT] named_(parameters | buffers) do not support recursive iteration.,https://github.com/pytorch/pytorch/issues/38863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-26T17:47:13Z +issue,38782,SPMG in DDP does not have gradient computation and communication overlap?,https://github.com/pytorch/pytorch/issues/38782,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-05-26T17:32:58Z +issue,38940,[FR] Support SyncBatchNorm in DataParallel,https://github.com/pytorch/pytorch/issues/38940,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nccl",scan when time allows,2020-05-26T17:32:21Z +issue,33668,"/usr/bin/x86_64-linux-gnu-ld: warning: libcusparse.so.10.0, needed by /pytorch_master/build/lib/libtorch_cuda.so, not found (try using -rpath or -rpath-link)",https://github.com/pytorch/pytorch/issues/33668,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2020-05-26T16:26:04Z +issue,3791,Make pytest stop printing docstrings in its default diagnostic output,https://github.com/pytorch/pytorch/issues/3791,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-05-26T16:14:48Z +issue,3697,Deprecate inplace argument in torch.nn.functional,https://github.com/pytorch/pytorch/issues/3697,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bc-breaking, module: deprecation, module: nn",scan when time allows,2020-05-26T16:08:17Z +issue,38876,Override `__call__` instead of `forward`,https://github.com/pytorch/pytorch/issues/38876,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: typing,scan when time allows,2020-05-26T13:52:37Z +issue,34879,Addition of Siamese loss or contrastive loss function,https://github.com/pytorch/pytorch/issues/34879,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2020-05-25T16:47:47Z +issue,10119,"[Caffe2] MNIST Tutorial LMDB Error ""Cannot open db""",https://github.com/pytorch/pytorch/issues/10119,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-24T09:37:02Z +issue,34651,Support passing memoryview to torch.as_tensor,https://github.com/pytorch/pytorch/issues/34651,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-23T22:17:44Z +issue,38737,equivalent of tensorflow's embedding_lookup_sparse,https://github.com/pytorch/pytorch/issues/38737,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-23T20:21:48Z +issue,37837,[DRAFT] Channels Last + AMP support plan for 1.6 release,https://github.com/pytorch/pytorch/issues/37837,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2020-05-22T23:38:11Z +issue,38936,[FR] DataParallel arg rename device_ids->devices,https://github.com/pytorch/pytorch/issues/38936,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-05-22T19:34:01Z +issue,14699,[Feature Request] linux distribution friendly build system,https://github.com/pytorch/pytorch/issues/14699,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-22T16:11:13Z +issue,36575,One confusion about the CompilationUnit destructuring process in torch/jit/__init__.py,https://github.com/pytorch/pytorch/issues/36575,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-22T03:48:34Z +issue,36378,CMake targets wrongly forward unknown options to NVCC (v1.5+),https://github.com/pytorch/pytorch/issues/36378,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-21T22:53:32Z +issue,38855,Can not compile GridSamplerKernel.cpp with gcc-9.3,https://github.com/pytorch/pytorch/issues/38855,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-21T03:56:54Z +issue,38854,Can't compile QuantizedOpKernels.cpp using gcc-9.3,https://github.com/pytorch/pytorch/issues/38854,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-21T03:56:27Z +issue,38719,Building NVCC (Device) failed when building from source,https://github.com/pytorch/pytorch/issues/38719,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-20T02:52:23Z +issue,38437,[JIT] torch.tensor needs a Tensor overload,https://github.com/pytorch/pytorch/issues/38437,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: jit",scan when time allows,2020-05-19T18:24:30Z +issue,38640,Maxunpool seems to give a weird error message,https://github.com/pytorch/pytorch/issues/38640,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-19T16:13:13Z +issue,38662,Unify CPU/CUDA exponential transformation formula,https://github.com/pytorch/pytorch/issues/38662,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2020-05-19T15:40:35Z +issue,38720,DOC: add documentation for undocumented classes and methods,https://github.com/pytorch/pytorch/issues/38720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-05-19T15:40:06Z +issue,38611,Investigate using -cospi(u) / sinpi(u) instead of tan(pi * (u - 0.5)) in transformation::cauchy,https://github.com/pytorch/pytorch/issues/38611,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: random",scan when time allows,2020-05-19T15:21:27Z +issue,38613,Investigate using log1p instead of log in transformation functions(TransformationHelper.h),https://github.com/pytorch/pytorch/issues/38613,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: random",scan when time allows,2020-05-19T15:21:11Z +issue,38619,"Can you add NMS,RoIAlign,RoIPool for libtorch?",https://github.com/pytorch/pytorch/issues/38619,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-05-19T14:49:52Z +issue,38709,NaN Loss for FasterRCNN on Multiclass Object Detection on Custom Dataset COCO,https://github.com/pytorch/pytorch/issues/38709,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-05-19T14:43:40Z +issue,38612,Investigate exponential distribution improvements,https://github.com/pytorch/pytorch/issues/38612,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: random",scan when time allows,2020-05-19T14:16:54Z +issue,38595,torch.norm p/ord parameter documentation is wrong,https://github.com/pytorch/pytorch/issues/38595,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-05-19T14:05:45Z +issue,37136,[Design][RFC] RemoteModule API Design,https://github.com/pytorch/pytorch/issues/37136,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: jit",scan when time allows,2020-05-19T06:31:14Z +issue,38703,torch.as_tensor(np_array) is sometimes much faster than torch.tensor(np_array),https://github.com/pytorch/pytorch/issues/38703,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: performance",scan when time allows,2020-05-19T04:20:56Z +issue,37373,Move bernoulli_() to DistributionTemplates,https://github.com/pytorch/pytorch/issues/37373,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2020-05-19T01:30:04Z +issue,21794,Linker errors when building project with OpenCV,https://github.com/pytorch/pytorch/issues/21794,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-17T21:39:26Z +issue,38553,"When I use cuda(), wg = th.matmul(extra_obs, extra_obs.transpose(-2, -1)) take a mistake",https://github.com/pytorch/pytorch/issues/38553,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2020-05-17T06:01:51Z +issue,38472,[torch.jit.trace] torch.jit.trace fixed batch size CNN,https://github.com/pytorch/pytorch/issues/38472,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-17T04:59:03Z +issue,38474,Scale parameter downcasted and rounded down in pytorch.distributions.Normal,https://github.com/pytorch/pytorch/issues/38474,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-05-15T23:08:31Z +issue,38564,tensorboard projector mode with custom metadata_header with only one label name,https://github.com/pytorch/pytorch/issues/38564,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-05-15T22:20:17Z +issue,38543,Make rebuildBucket() to be async in c10d reducer,https://github.com/pytorch/pytorch/issues/38543,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2020-05-15T21:49:51Z +issue,35937,MKLDNN_conv2d 2X slower than the native TH implementation,https://github.com/pytorch/pytorch/issues/35937,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: mkldnn, module: performance",scan when time allows,2020-05-15T15:56:24Z +issue,38548,Add Distributed LR Scheduler to RPC,https://github.com/pytorch/pytorch/issues/38548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: rpc",scan when time allows,2020-05-15T14:34:37Z +issue,29816,AdamSparse fails to run,https://github.com/pytorch/pytorch/issues/29816,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-05-15T05:42:43Z +issue,38510,[DISCUSSION] RPC server-side ThreadLocalState,https://github.com/pytorch/pytorch/issues/38510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: rpc, oncall: jit",scan when time allows,2020-05-14T22:07:06Z +issue,38473,Python builtin function next() is currently not supported in Torchscript,https://github.com/pytorch/pytorch/issues/38473,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-14T17:33:22Z +issue,33672,[dev] `RecursiveScriptModule` does not expose `jit.ignore`d methods,https://github.com/pytorch/pytorch/issues/33672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-14T11:53:28Z +issue,38204,missing dependency `protobuf` for caffe2,https://github.com/pytorch/pytorch/issues/38204,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-14T04:07:45Z +issue,38354,"Can I add more ""Project Documentation"" on the PYPI webpage?",https://github.com/pytorch/pytorch/issues/38354,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs",scan when time allows,2020-05-14T01:48:23Z +issue,38319,c10/macros/cmake_macros.h not exists,https://github.com/pytorch/pytorch/issues/38319,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: mobile,scan when time allows,2020-05-13T07:38:24Z +issue,3600,making .cuda() falls back to an identity function when gpu is not available,https://github.com/pytorch/pytorch/issues/3600,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-05-12T21:05:26Z +issue,9046,[feature request] freeze() for nn.Module,https://github.com/pytorch/pytorch/issues/9046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-05-12T19:22:36Z +issue,38241,SIGXCPU at test_cholesky_solve with AMD EPYC 7742 64-Core Processor,https://github.com/pytorch/pytorch/issues/38241,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-05-12T16:38:15Z +issue,38208,nn.Module Abstract Class attribute overrides Child.,https://github.com/pytorch/pytorch/issues/38208,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-12T16:36:38Z +issue,38212,"Please help, building failing with MAGMA support",https://github.com/pytorch/pytorch/issues/38212,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-12T16:34:06Z +issue,38170,"OneCycleLR raises ""Tried to step { step_num + 1 } times"" after the value is more than expected.",https://github.com/pytorch/pytorch/issues/38170,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-12T16:24:11Z +issue,38323,Support alternate casting rules,https://github.com/pytorch/pytorch/issues/38323,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: type promotion",scan when time allows,2020-05-12T15:56:32Z +issue,38328,Provide an issubdtype API,https://github.com/pytorch/pytorch/issues/38328,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: numpy, module: type promotion",scan when time allows,2020-05-12T15:56:06Z +issue,38272,"PyTorch's rot90 returns a new tensor, inconsistent with NumPy's returning a view",https://github.com/pytorch/pytorch/issues/38272,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-05-12T04:26:53Z +issue,33089,test_baddbmm_cpu_float32 fails locally for me when built with DEBUG=1,https://github.com/pytorch/pytorch/issues/33089,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-05-12T02:59:36Z +issue,32868,__cuda_array_interface__ conversion does not support readonly arrays,https://github.com/pytorch/pytorch/issues/32868,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: internals, module: numba",scan when time allows,2020-05-11T23:40:10Z +issue,38245,Make torch.rpc accept store as optional parameter,https://github.com/pytorch/pytorch/issues/38245,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-05-11T20:10:07Z +issue,18283,Doubly freed pointer in torch::cat error handling when called via pybind11.,https://github.com/pytorch/pytorch/issues/18283,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: error checking",scan when time allows,2020-05-11T17:42:42Z +issue,18776,Value of torch.backends.cudnn.benchmark Baked into JIT-Traced Modules ( 150x slowdown on ConvTranspose2d() ) [jit] [libtorch] [cudnn],https://github.com/pytorch/pytorch/issues/18776,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-11T09:54:55Z +issue,38009,Resnet Model always predicting same label,https://github.com/pytorch/pytorch/issues/38009,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-05-11T09:16:11Z +issue,38202,IndexError reports the wrong dimension when fancy indexing,https://github.com/pytorch/pytorch/issues/38202,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2020-05-10T15:02:52Z +issue,38177,valgrind says libtorch has memory leak,https://github.com/pytorch/pytorch/issues/38177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-05-10T14:22:47Z +issue,21779,"Use of word ""elements"" in `torch.utils.data` samplers",https://github.com/pytorch/pytorch/issues/21779,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-05-10T09:27:23Z +issue,35414,"MacOS Error: subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '12']' returned non-zero exit status 1.",https://github.com/pytorch/pytorch/issues/35414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: macos",scan when time allows,2020-05-09T15:25:03Z +issue,38190,test_ReplicationPad3d (test_nn.TestNN) takes too long to run,https://github.com/pytorch/pytorch/issues/38190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-05-09T07:45:21Z +issue,38189,test_LocalResponseNorm_3d_custom_params (test_nn.TestNN) takes too long to run,https://github.com/pytorch/pytorch/issues/38189,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-05-09T07:44:46Z +issue,38188,test_interpolate_nearest_scale_3d in test_nn takes too long to run,https://github.com/pytorch/pytorch/issues/38188,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-05-09T07:43:57Z +issue,18351,Port SpatialConvolutionMM and VolumetricConvolutionMM to ATen,https://github.com/pytorch/pytorch/issues/18351,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: porting,scan when time allows,2020-05-09T07:16:18Z +issue,7890,[feature request] batch_first of RNN hidden weight for Multi GPU training,https://github.com/pytorch/pytorch/issues/7890,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nn",scan when time allows,2020-05-09T06:35:39Z +issue,38184,Feature Request: Log uniform candidate sampler,https://github.com/pytorch/pytorch/issues/38184,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-05-09T06:18:29Z +issue,2691,Feature Request: Support grad of grad in fused RNNs,https://github.com/pytorch/pytorch/issues/2691,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: double backwards, module: nn",scan when time allows,2020-05-09T06:09:52Z +issue,1932,Feature Request: ReLU on LSTMs and GRUs,https://github.com/pytorch/pytorch/issues/1932,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2020-05-09T05:58:48Z +issue,1686,Feature Request: load_state_dict should take filenames,https://github.com/pytorch/pytorch/issues/1686,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-05-09T05:53:37Z +issue,4987,MultiGPU hangs Titan Xp in multiprocessing/queue.py,https://github.com/pytorch/pytorch/issues/4987,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: macos, module: multiprocessing",scan when time allows,2020-05-08T22:27:00Z +issue,38142,[JIT] Aliased Python references to script::Object escape IValue reference counting,https://github.com/pytorch/pytorch/issues/38142,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-08T22:02:08Z +issue,37985,torch.save incompatible with lzma file,https://github.com/pytorch/pytorch/issues/37985,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-05-08T16:06:53Z +issue,38118,Add helpers to save/load RPC-based models,https://github.com/pytorch/pytorch/issues/38118,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-05-08T14:35:39Z +issue,38010,Split more pages into sub-topics,https://github.com/pytorch/pytorch/issues/38010,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2020-05-08T14:12:43Z +issue,27958,how to use libtorch library in cuda file with nvcc compiler(c++)?,https://github.com/pytorch/pytorch/issues/27958,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-05-08T08:30:40Z +issue,24463,Crashes on torch.cuda.memory_allocated(device),https://github.com/pytorch/pytorch/issues/24463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2020-05-08T05:42:47Z +issue,38012,Formatting of topic and sub-topic pages,https://github.com/pytorch/pytorch/issues/38012,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2020-05-08T03:35:44Z +issue,33867,cuDNN batchnorm with non-contiguous running mean silently discards updates,https://github.com/pytorch/pytorch/issues/33867,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2020-05-07T18:12:31Z +issue,38000,"New padding size format for F.pad, allowing named tensors and more clear syntax overall",https://github.com/pytorch/pytorch/issues/38000,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-05-07T17:36:07Z +issue,7786,[feature request] Simple and Efficient way to get gradients of each element of a sum,https://github.com/pytorch/pytorch/issues/7786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-05-07T14:41:06Z +issue,37988,JIT compilation of NamedTuple Containing a NamedTuple fails,https://github.com/pytorch/pytorch/issues/37988,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-07T03:43:38Z +issue,37987,Special methods on torchscript custom class,https://github.com/pytorch/pytorch/issues/37987,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-05-07T03:40:12Z +issue,37860,autograd engine callbacks don't respect non-default cuda streams,https://github.com/pytorch/pytorch/issues/37860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cuda",scan when time allows,2020-05-06T21:11:40Z +issue,37880,"[RFC] Don't install CI dependencies in build scripts, install them in underlying docker images",https://github.com/pytorch/pytorch/issues/37880,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2020-05-06T20:54:53Z +issue,37895,torch.addmv can't take as input tensors with different dtypes,https://github.com/pytorch/pytorch/issues/37895,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2020-05-06T18:14:35Z +issue,37814,DistributedDataParallel does not support Modules that take no inputs.,https://github.com/pytorch/pytorch/issues/37814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-05-06T02:10:05Z +issue,3281,"DataLoader ""casting"" non statndard objects to lists",https://github.com/pytorch/pytorch/issues/3281,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-05-05T23:27:37Z +issue,3364,type of torch.bernoulli and torch.multinomial inconsistent,https://github.com/pytorch/pytorch/issues/3364,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-05-05T23:15:28Z +issue,3390,High CPU use by clock_gettime syscall,https://github.com/pytorch/pytorch/issues/3390,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-05-05T23:14:36Z +issue,37847,llvmlite version issue when upgrading CI docker image to python 3.8,https://github.com/pytorch/pytorch/issues/37847,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2020-05-05T22:58:29Z +issue,37588,"test_qnnpack_sigmoid failing on old CPU, built pytorch 1.5.0",https://github.com/pytorch/pytorch/issues/37588,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2020-05-05T21:24:54Z +issue,27752,Version number is still duplicated in a bunch of places,https://github.com/pytorch/pytorch/issues/27752,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2020-05-05T20:37:33Z +issue,34606,Multiple CMake target errors ever since commit 0e52627358,https://github.com/pytorch/pytorch/issues/34606,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-05T19:52:26Z +issue,37512,Improve visibility in test suite timings,https://github.com/pytorch/pytorch/issues/37512,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests",scan when time allows,2020-05-05T05:37:35Z +issue,36804,c++: error: unrecognized command line option '-Wthread-safety',https://github.com/pytorch/pytorch/issues/36804,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-05T05:05:29Z +issue,37629,logging_is_not_google_glog.h:24:11: error: 'const int ERROR' redeclared as different kind of symbol (v1.3.0),https://github.com/pytorch/pytorch/issues/37629,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-05T03:04:02Z +issue,7250,feature request: support for new/future hardware accelerators,https://github.com/pytorch/pytorch/issues/7250,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-05T00:53:38Z +issue,37684,"Add Compound key, make custom ops default to it (but keep internal users using CatchAll)",https://github.com/pytorch/pytorch/issues/37684,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-05T00:31:39Z +issue,37731,Variational Dropout In RNN,https://github.com/pytorch/pytorch/issues/37731,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2020-05-04T22:04:18Z +issue,37782,Guard Gloo and TensorPipe related code in RPC with #ifdef,https://github.com/pytorch/pytorch/issues/37782,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: rpc, module: tensorpipe",scan when time allows,2020-05-04T19:19:51Z +issue,37457,[RuntimeError] Tensor creation using storage fails,https://github.com/pytorch/pytorch/issues/37457,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2020-05-02T23:24:31Z +issue,37529,Implement generic function scheduler in c10/util,https://github.com/pytorch/pytorch/issues/37529,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-05-02T01:38:47Z +issue,37444,Multi-Process Single-GPU is bad,https://github.com/pytorch/pytorch/issues/37444,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2020-05-02T00:29:40Z +issue,37434,rsub incorrectly exposed in torch,https://github.com/pytorch/pytorch/issues/37434,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-05-01T23:03:42Z +issue,37549,Are there any differences in kernel memory between RX2080's and Quadro RTX4000?,https://github.com/pytorch/pytorch/issues/37549,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2020-05-01T21:50:34Z +issue,37510,"Cannot build pytorch with linker arguments in C{,XX}FLAGS",https://github.com/pytorch/pytorch/issues/37510,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-05-01T21:02:19Z +issue,30929,How to set not to build libtorch_cpu.so and libmkl_*.so dependencies?,https://github.com/pytorch/pytorch/issues/30929,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkl",scan when time allows,2020-05-01T18:47:25Z +issue,37598,NCCL fails to find cuda include dir,https://github.com/pytorch/pytorch/issues/37598,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nccl, oncall: distributed",scan when time allows,2020-05-01T13:10:43Z +issue,37441,Tensor.as_strided_ is not documented,https://github.com/pytorch/pytorch/issues/37441,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-05-01T04:05:53Z +issue,37440,Tensor.is_same_size not documented,https://github.com/pytorch/pytorch/issues/37440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-05-01T04:04:15Z +issue,3076,BN slows down double-backprop enormously,https://github.com/pytorch/pytorch/issues/3076,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2020-04-30T22:07:49Z +issue,32444,DataParallel does not work with sparse parameters,https://github.com/pytorch/pytorch/issues/32444,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-04-30T20:56:19Z +issue,32919,"Torchscript used to work, but now it fails with VariableTensorId error",https://github.com/pytorch/pytorch/issues/32919,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: android, oncall: jit, oncall: mobile",scan when time allows,2020-04-30T17:07:25Z +issue,31579,Option to apply weights to gradients when using DistributedDataParallel,https://github.com/pytorch/pytorch/issues/31579,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: data parallel, oncall: distributed",scan when time allows,2020-04-30T02:21:52Z +issue,37484,No speedup from channels_last with DataParallel,https://github.com/pytorch/pytorch/issues/37484,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: memory format, module: performance",scan when time allows,2020-04-29T23:33:07Z +issue,37282,python setup.py install error,https://github.com/pytorch/pytorch/issues/37282,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-04-29T23:21:40Z +issue,36058,Half type promotion with Numpy arrays is incorrect,https://github.com/pytorch/pytorch/issues/36058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2020-04-29T19:19:15Z +issue,33327,Creating Torch tensors slows OpenCV video reading a lot,https://github.com/pytorch/pytorch/issues/33327,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: performance",scan when time allows,2020-04-29T18:48:41Z +issue,36307,[docs] Matrix transpose missing in formula in nn.functional.bilinear docs,https://github.com/pytorch/pytorch/issues/36307,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-04-28T17:05:49Z +issue,37416,[docs] Unclear return type of torch.randint and extra comma in arg spec,https://github.com/pytorch/pytorch/issues/37416,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-04-28T15:55:04Z +issue,33055,[docs] Strange order of items in docs contents in left pane,https://github.com/pytorch/pytorch/issues/33055,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-04-28T15:32:15Z +issue,37153,CPU out of bound memory access in CUDA reduction kernel config,https://github.com/pytorch/pytorch/issues/37153,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tensoriterator",scan when time allows,2020-04-28T14:38:13Z +issue,31615,'torch.load' report 'bad pickle data',https://github.com/pytorch/pytorch/issues/31615,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-04-28T09:09:27Z +issue,34998,Better examples in functional autograd functions,https://github.com/pytorch/pytorch/issues/34998,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2020-04-27T18:23:27Z +issue,37204,Make it harder to make SIOF bugs for torchbind classes referenced by schemas,https://github.com/pytorch/pytorch/issues/37204,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-26T20:30:05Z +issue,37300,Data caching module a la `Sampler`,https://github.com/pytorch/pytorch/issues/37300,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-26T20:18:14Z +issue,37314,equal_nan keyword not implemented for complex torch.isclose,https://github.com/pytorch/pytorch/issues/37314,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy",scan when time allows,2020-04-26T09:22:37Z +issue,20613,groupby function,https://github.com/pytorch/pytorch/issues/20613,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-25T14:31:57Z +issue,36004,C++ tensor print doesn't show requires_grad and grad_fn like Python tensor print,https://github.com/pytorch/pytorch/issues/36004,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: printing",scan when time allows,2020-04-25T01:06:23Z +issue,37103,Legacy fuser doesn't do remainder consistently with `aten::remainder`,https://github.com/pytorch/pytorch/issues/37103,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-24T01:26:11Z +issue,37149,Cuda profiler + DataParallel + manual profiling start = strange profiling overhead pattern,https://github.com/pytorch/pytorch/issues/37149,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-04-23T22:54:00Z +issue,36713,variable name N_ conflicts with an internationalization macro in glib,https://github.com/pytorch/pytorch/issues/36713,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: jit",scan when time allows,2020-04-23T20:38:19Z +issue,8550,caffe2: Does caffe support the model running on different device (CPU and GPU) same time ?,https://github.com/pytorch/pytorch/issues/8550,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-23T07:58:47Z +issue,36917,"jit trace failed due to ""failed to differentiate `prim::ListConstruct`""",https://github.com/pytorch/pytorch/issues/36917,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-22T19:04:11Z +issue,36931,Pull hacked twins out of prim ops,https://github.com/pytorch/pytorch/issues/36931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-22T19:00:16Z +issue,37065,Deprecated mask fill mask type can causes pages and pages of repeated messages,https://github.com/pytorch/pytorch/issues/37065,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-22T18:10:02Z +issue,3152,Sparse tensor .new(size) can be confusing,https://github.com/pytorch/pytorch/issues/3152,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2020-04-22T17:50:48Z +issue,3176,Add safety checks in `index_add`/`scatter_add`,https://github.com/pytorch/pytorch/issues/3176,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-22T17:49:29Z +issue,36754,Clearer guidance on when to define an operator as a method on torchbind'ed class versus standalone function,https://github.com/pytorch/pytorch/issues/36754,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-22T14:49:47Z +issue,28863,Command for downloading torch 1.12.0 CUDA 10 linked to cu92 version,https://github.com/pytorch/pytorch/issues/28863,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2020-04-22T12:18:49Z +issue,37004,"RuntimeError: NCCL error in ProcessGroupNCCL.cpp:290, unhandled system error",https://github.com/pytorch/pytorch/issues/37004,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dependency bug, module: nccl, oncall: distributed",scan when time allows,2020-04-22T01:52:04Z +issue,36739,Custom class type name is very wordy,https://github.com/pytorch/pytorch/issues/36739,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-21T20:35:37Z +issue,36874,Memory leak upon exception for interactive consoles,https://github.com/pytorch/pytorch/issues/36874,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-04-21T16:55:39Z +issue,36999,Add a CI configuration to test USE_DISTRIBUTED=0,https://github.com/pytorch/pytorch/issues/36999,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: ci, module: tests, oncall: distributed",scan when time allows,2020-04-21T14:33:23Z +issue,32018,DistributedDataParallel non-floating point dtype parameter with requires_grad=False,https://github.com/pytorch/pytorch/issues/32018,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-04-20T19:12:24Z +issue,36891,pytorch and c++ inference disagree,https://github.com/pytorch/pytorch/issues/36891,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-04-20T16:42:27Z +issue,36792,[1.4.1] Intel MPI libs not found.,https://github.com/pytorch/pytorch/issues/36792,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn",scan when time allows,2020-04-20T07:02:53Z +issue,17774,RuntimeError: cuDNN error: CUDNN_STATUS_EXECUTION_FAILED when batch size is too large,https://github.com/pytorch/pytorch/issues/17774,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2020-04-18T17:46:16Z +issue,36638,Poor elmenetwise_kernel performance becomes critical on small mini-batch sizes,https://github.com/pytorch/pytorch/issues/36638,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance, module: tensoriterator",scan when time allows,2020-04-18T05:27:18Z +issue,36753,When you try to register a kernel that make boxed from unboxed functor doesn't support you get a horrible error message,https://github.com/pytorch/pytorch/issues/36753,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-18T04:48:12Z +issue,36733,Custom Generators don't work in JIT,https://github.com/pytorch/pytorch/issues/36733,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-17T21:15:43Z +issue,36517,XNNPACK operators are not actually registered under xnnpack namespace,https://github.com/pytorch/pytorch/issues/36517,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: xnnpack,scan when time allows,2020-04-16T22:33:52Z +issue,36407,Unable to build wheel from RC3,https://github.com/pytorch/pytorch/issues/36407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-04-16T04:27:06Z +issue,34754,General Purpose Faulty RPC Agent,https://github.com/pytorch/pytorch/issues/34754,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2020-04-14T19:08:43Z +issue,36444,Comparison ops for Complex Tensors,https://github.com/pytorch/pytorch/issues/36444,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy",scan when time allows,2020-04-14T16:15:06Z +issue,36553,Typecasting issue in MSELoss,https://github.com/pytorch/pytorch/issues/36553,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: type promotion",scan when time allows,2020-04-14T14:20:53Z +issue,36519,Restructure test_c10d.py and test_distributed.py,https://github.com/pytorch/pytorch/issues/36519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-04-13T21:33:11Z +issue,36516,DDP should divide bucket contents by the number of global replicas instead of world size,https://github.com/pytorch/pytorch/issues/36516,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-04-13T21:30:44Z +issue,19911,"BatchNorm1d does not support batchsize>65535 in eval mode with 3 dimension (NxCxL), raise CUDNN_STATUS_NOT_SUPPORTED",https://github.com/pytorch/pytorch/issues/19911,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: dependency bug",scan when time allows,2020-04-13T15:06:19Z +issue,31264,Failed to config caffe2_rocksdb in cmake,https://github.com/pytorch/pytorch/issues/31264,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-04-13T05:18:58Z +issue,36436,"After `create_graph=True`, calculating `backward()` on sparse Tensor fails",https://github.com/pytorch/pytorch/issues/36436,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-13T04:51:04Z +issue,36469,Wrong results for multiplication of non-finite complex numbers with real numbers,https://github.com/pytorch/pytorch/issues/36469,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: complex, module: numpy",scan when time allows,2020-04-13T04:28:25Z +issue,36356,Memory leak issue still exists in CI,https://github.com/pytorch/pytorch/issues/36356,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2020-04-13T04:18:25Z +issue,36370,Jit doesn't match schema like eager mode does,https://github.com/pytorch/pytorch/issues/36370,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-10T20:59:45Z +issue,36317,Updating a ModelDict instance with another ModelDict instance generates error.,https://github.com/pytorch/pytorch/issues/36317,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-04-10T13:50:11Z +issue,32916,from torch._C import * (ImportError: DLL load failed),https://github.com/pytorch/pytorch/issues/32916,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-10T11:40:08Z +issue,32370,Backward `Functional.conv3d` is slow when cuDNN is enabled,https://github.com/pytorch/pytorch/issues/32370,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2020-04-09T16:05:28Z +issue,32287,Python 3.8 Windows JIT test failure,https://github.com/pytorch/pytorch/issues/32287,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-09T15:40:10Z +issue,34715,Maybe gpu_kernel shouldn't ASSERT_HOST_DEVICE_LAMBDA,https://github.com/pytorch/pytorch/issues/34715,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-09T10:25:02Z +issue,23110,[RFC] RPC Based Distributed Model Parallel,https://github.com/pytorch/pytorch/issues/23110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-04-09T04:38:01Z +issue,35942,Build PyTorch-1.4.0 from source failed,https://github.com/pytorch/pytorch/issues/35942,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-04-09T01:00:04Z +issue,36160,Decouple DDP from CUDA,https://github.com/pytorch/pytorch/issues/36160,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2020-04-08T08:20:37Z +issue,14219,C++ model load error,https://github.com/pytorch/pytorch/issues/14219,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-07T18:35:59Z +issue,36140,Inconsistent ProcessGroupMPI work data structure for send/recv and collectives,https://github.com/pytorch/pytorch/issues/36140,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: distributed",scan when time allows,2020-04-07T14:28:18Z +issue,36099,"In AutogradContext, get_saved_variables() should be renamed to get_saved_tensors()",https://github.com/pytorch/pytorch/issues/36099,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-04-06T20:59:03Z +issue,36062,RuntimeError CUDA error despite CUDA available and GPU supported,https://github.com/pytorch/pytorch/issues/36062,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: error checking",scan when time allows,2020-04-06T20:23:02Z +issue,35944,"Copy activations from one parts to another part in tensor, but report error",https://github.com/pytorch/pytorch/issues/35944,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-06T17:25:31Z +issue,35832,Linking error after marking an op `manual_kernel_registration: True`,https://github.com/pytorch/pytorch/issues/35832,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-04-06T17:17:02Z +issue,35964,TensorOptions shouldn't provide default values,https://github.com/pytorch/pytorch/issues/35964,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-04-06T16:41:17Z +issue,36028,About tensorboard in pytorch record graph in different state,https://github.com/pytorch/pytorch/issues/36028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: visualization,scan when time allows,2020-04-06T14:59:10Z +issue,36059,Some tutorials cannot be found from both side panel and tutorial welcome page,https://github.com/pytorch/pytorch/issues/36059,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-04-06T14:50:46Z +issue,30528,Warn about driver CUDA mismatch in torch.cuda.is_available(),https://github.com/pytorch/pytorch/issues/30528,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-04-05T12:52:11Z +issue,29093,error: 'SO_REUSEPORT' was not declared in this scope,https://github.com/pytorch/pytorch/issues/29093,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2020-04-04T22:41:03Z +issue,35788,[jit] Structural typing shouldn't work,https://github.com/pytorch/pytorch/issues/35788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-03T21:07:56Z +issue,31916,I'm not able to build pytorch with tensorrt (current master),https://github.com/pytorch/pytorch/issues/31916,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-04-01T19:24:29Z +issue,35820,[JIT] Tracer bakes List[Tensor] attribute as constant,https://github.com/pytorch/pytorch/issues/35820,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-04-01T19:08:42Z +issue,26139,c10 List API hard to use,https://github.com/pytorch/pytorch/issues/26139,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-04-01T17:15:12Z +issue,35766,Clarify tensor storage communication behavior in RPC,https://github.com/pytorch/pytorch/issues/35766,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rpc",scan when time allows,2020-04-01T16:49:01Z +issue,35655,Wrong conv2d output on GPU when kernel has many zeros,https://github.com/pytorch/pytorch/issues/35655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: dependency bug, module: numerical-stability",scan when time allows,2020-04-01T13:32:19Z +issue,35770,BatchNormFuncOptions object cant be printed in C++,https://github.com/pytorch/pytorch/issues/35770,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2020-04-01T13:25:23Z +issue,35749,Cannot JIT functions with custom backwards (e.g. swish),https://github.com/pytorch/pytorch/issues/35749,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: jit",scan when time allows,2020-04-01T13:24:28Z +issue,35759,how to do 3d data augmentation in parallel on the gpu?,https://github.com/pytorch/pytorch/issues/35759,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader",scan when time allows,2020-04-01T13:23:34Z +issue,35755,Expected object of scalar type Double but got scalar type Float for argument #2 'mat2' in call to _th_mm,https://github.com/pytorch/pytorch/issues/35755,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-31T20:38:48Z +issue,35758,UserRRef should store error if it sees any and prevent subsequent usage,https://github.com/pytorch/pytorch/issues/35758,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-03-31T14:58:52Z +issue,35681,TensorBoard add_scalars throws error when dict has keys of type int,https://github.com/pytorch/pytorch/issues/35681,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-31T08:27:00Z +issue,34939,Enable OpaqueTensor to possess Storage then allow it to view from CPUTensor,https://github.com/pytorch/pytorch/issues/34939,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: mkldnn",scan when time allows,2020-03-31T06:45:20Z +issue,33227,[v1.5] Python/C++ API parity master tracking task,https://github.com/pytorch/pytorch/issues/33227,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-03-31T01:58:00Z +issue,12764,[feature request] Operator Overloading,https://github.com/pytorch/pytorch/issues/12764,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-30T23:00:30Z +issue,35648,Caffe2 ReshapeOpGPUTest crashes on Windows,https://github.com/pytorch/pytorch/issues/35648,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-03-30T22:47:48Z +issue,35651,Caffe2 utility_ops_gpu_test fails on Windows,https://github.com/pytorch/pytorch/issues/35651,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-03-30T22:47:38Z +issue,35640,How do you change Adam learning rate since the latest commits ?,https://github.com/pytorch/pytorch/issues/35640,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-03-30T22:38:52Z +issue,2854,Add the new lr_scheduler which called poly,https://github.com/pytorch/pytorch/issues/2854,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-03-30T22:25:27Z +issue,2849,Learning rate scheduler have different APIs,https://github.com/pytorch/pytorch/issues/2849,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-03-30T22:25:17Z +issue,35678,USE_AVX/USE_AVX2 does not affect __AVX2__ macro defition,https://github.com/pytorch/pytorch/issues/35678,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2020-03-30T22:21:17Z +issue,35671,allgather_coalesced for tensors of different types seems to be broken,https://github.com/pytorch/pytorch/issues/35671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-03-30T22:14:06Z +issue,35408,torch.jit.script works for x*sigmoid(x) but not for x*sin(x),https://github.com/pytorch/pytorch/issues/35408,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-30T03:39:58Z +issue,35118,Customize batch size based on gpu id,https://github.com/pytorch/pytorch/issues/35118,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-03-30T00:38:30Z +issue,31822,Libtorch's files conflict with glog's file?,https://github.com/pytorch/pytorch/issues/31822,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-03-28T17:43:10Z +issue,28746,super().__init__() not called in torch.nn.Module.__init__,https://github.com/pytorch/pytorch/issues/28746,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-03-27T12:39:08Z +issue,9604,WARNING:root:This caffe2 python run does not have GPU support. Will run in CPU only mode. CRITICAL:root:Cannot load caffe2.python. Error: libcaffe2.so: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/9604,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-27T08:00:11Z +issue,34367,Better testing of the autograd engine,https://github.com/pytorch/pytorch/issues/34367,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-03-27T02:02:57Z +issue,2739,Proposal: simplify overloaded Tensor function signatures,https://github.com/pytorch/pytorch/issues/2739,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-27T01:31:15Z +issue,35160,Documentation doesn't cover MWE using launch.py script,https://github.com/pytorch/pytorch/issues/35160,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-03-25T21:43:23Z +issue,34300,Inconsistent semantics of converting inf/-inf to long,https://github.com/pytorch/pytorch/issues/34300,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-03-25T16:28:07Z +issue,35234,Raspberry Pi Zero build fails,https://github.com/pytorch/pytorch/issues/35234,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-03-25T14:16:17Z +issue,35035,How install old version pytorch 1.2.0 from source?,https://github.com/pytorch/pytorch/issues/35035,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-03-24T19:58:57Z +issue,35208,The BatchNorm error in `DataParallel`,https://github.com/pytorch/pytorch/issues/35208,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nn",scan when time allows,2020-03-24T19:45:29Z +issue,35078,Generator C++ API should match Python API,https://github.com/pytorch/pytorch/issues/35078,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: random",scan when time allows,2020-03-24T17:38:10Z +issue,35097,undefined symbols when using libtorch and ITK,https://github.com/pytorch/pytorch/issues/35097,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-03-23T21:12:43Z +issue,22961,performance much worse on 2080ti than 1080ti,https://github.com/pytorch/pytorch/issues/22961,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-03-23T19:12:34Z +issue,35040,Deprecate and remove RegisterOperators,https://github.com/pytorch/pytorch/issues/35040,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-03-23T13:51:03Z +issue,35038,Rename Dispatcher::findSchema to Dispatcher::findOperator,https://github.com/pytorch/pytorch/issues/35038,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2020-03-23T13:50:20Z +issue,35128,Autograd Engine leaks reentrant threadpool threads on deletion,https://github.com/pytorch/pytorch/issues/35128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-03-20T23:26:03Z +issue,34878,Set the number of CUDAStream,https://github.com/pytorch/pytorch/issues/34878,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-03-20T16:32:34Z +issue,2703,Have ppc64le docker images?,https://github.com/pytorch/pytorch/issues/2703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: power,scan when time allows,2020-03-20T05:21:34Z +issue,2724,DataLoader gets stuck after model initialization,https://github.com/pytorch/pytorch/issues/2724,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-03-20T05:20:33Z +issue,34483,JIT: Tracing faster than scripting,https://github.com/pytorch/pytorch/issues/34483,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-19T14:00:50Z +issue,32641,changed format of trace graph in torch 1.4.0,https://github.com/pytorch/pytorch/issues/32641,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-19T10:13:34Z +issue,34941,Issues with DataParallel on Multiple GPUs,https://github.com/pytorch/pytorch/issues/34941,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-03-19T08:16:38Z +issue,29190,How to run two different jit models in two GPUs respectively in one scrip?,https://github.com/pytorch/pytorch/issues/29190,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-19T06:14:43Z +issue,27975,"Sending CUDA tensors via queue between processes, memory of Consumer process grows infinitely",https://github.com/pytorch/pytorch/issues/27975,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: multiprocessing",scan when time allows,2020-03-19T02:47:41Z +issue,31478,DataParallel has different tensor copy behavior if batch size = 1,https://github.com/pytorch/pytorch/issues/31478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-03-19T00:23:29Z +issue,34949,Equality operator for torch.distribution.*,https://github.com/pytorch/pytorch/issues/34949,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-03-18T23:32:24Z +issue,34812,"Semantic differences between forward and backward of upsampling2d/3d in channels last (NHWC, NDHWC) format.",https://github.com/pytorch/pytorch/issues/34812,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory format,scan when time allows,2020-03-18T18:57:03Z +issue,34412,error in pytorch Docs: https://pytorch.org/docs/stable/distributions.html,https://github.com/pytorch/pytorch/issues/34412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-03-18T18:42:21Z +issue,34210,[docs] Unclear arg spec for torch.full,https://github.com/pytorch/pytorch/issues/34210,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: numpy",scan when time allows,2020-03-18T16:37:09Z +issue,34833,NO_CUDA and NO_DISTRIBUTED referenced in docs,https://github.com/pytorch/pytorch/issues/34833,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-03-18T15:05:29Z +issue,34765,Add ability to return a copy to Module's state_dict member function.,https://github.com/pytorch/pytorch/issues/34765,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-03-18T14:42:02Z +issue,34881,libtorch memory leak,https://github.com/pytorch/pytorch/issues/34881,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: memory usage",scan when time allows,2020-03-18T03:17:01Z +issue,31467,Distributed hangs on process termination with world_size=1,https://github.com/pytorch/pytorch/issues/31467,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",scan when time allows,2020-03-18T02:39:06Z +issue,25416,Add GPU support to c10d allreduce_coalesced,https://github.com/pytorch/pytorch/issues/25416,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-03-17T22:55:52Z +issue,29108,a retrained and saved jit module could not be reload.,https://github.com/pytorch/pytorch/issues/29108,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-16T23:17:22Z +issue,31545,IndexExpressions (or slice) for jit.script functions,https://github.com/pytorch/pytorch/issues/31545,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-16T21:04:53Z +issue,34728,CPU softmax performance poor when dim is not the last dimension,https://github.com/pytorch/pytorch/issues/34728,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2020-03-16T20:21:06Z +issue,25267,JIT leaks memory when I change the max sequence length,https://github.com/pytorch/pytorch/issues/25267,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-16T19:13:57Z +issue,34814,[JIT] Tensor.add_ incorrect schema matching on overloads,https://github.com/pytorch/pytorch/issues/34814,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-16T17:02:09Z +issue,32214,Pytorch compilation error,https://github.com/pytorch/pytorch/issues/32214,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-03-16T12:38:38Z +issue,32851,Silent failing of batch_sampler when the data points are lists of tensors.,https://github.com/pytorch/pytorch/issues/32851,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: nestedtensor",scan when time allows,2020-03-13T21:59:44Z +issue,34649,Confusing error message of tensor constructor when passing a storage,https://github.com/pytorch/pytorch/issues/34649,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-12T22:42:00Z +issue,2628,BatchNorm{1-2-3}d are redundant,https://github.com/pytorch/pytorch/issues/2628,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-03-12T22:34:15Z +issue,34323,PyTorch GPU memory allocation,https://github.com/pytorch/pytorch/issues/34323,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2020-03-12T22:28:40Z +issue,6760,[Feature request] LayerNormLSTMCell and LayerNormLSTM,https://github.com/pytorch/pytorch/issues/6760,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-12T15:24:17Z +issue,33129,Python package using CMake,https://github.com/pytorch/pytorch/issues/33129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-03-12T00:46:20Z +issue,34621,A better way to show users all build options,https://github.com/pytorch/pytorch/issues/34621,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-03-11T23:43:21Z +issue,33166,TensorIterator does not work with different input/output types,https://github.com/pytorch/pytorch/issues/33166,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tensoriterator, module: vectorization",scan when time allows,2020-03-11T16:29:39Z +issue,34409,PyTorch build with Ideep support?,https://github.com/pytorch/pytorch/issues/34409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn",scan when time allows,2020-03-11T08:22:47Z +issue,34361,[JIT legacy executor] device propagation regression,https://github.com/pytorch/pytorch/issues/34361,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-10T20:48:53Z +issue,34516,Expose chunk_sizes for DataParallel,https://github.com/pytorch/pytorch/issues/34516,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: nn",scan when time allows,2020-03-10T20:29:39Z +issue,25641,[distributed] all_gather on a List of Tensors directly,https://github.com/pytorch/pytorch/issues/25641,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",scan when time allows,2020-03-09T23:07:23Z +issue,33876,Torchscript incompatible with torch.cat for tensor lists,https://github.com/pytorch/pytorch/issues/33876,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-03-09T17:10:18Z +issue,34207,Support vars/__dict__ on torch.return_types tuples e.g. topk,https://github.com/pytorch/pytorch/issues/34207,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-07T16:28:11Z +issue,33787,MSE Loss Implementation,https://github.com/pytorch/pytorch/issues/33787,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-06T19:43:56Z +issue,33799,Add `start_process_in_context` to `torch.multiprocessing`,https://github.com/pytorch/pytorch/issues/33799,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2020-03-06T02:13:33Z +issue,34063,[docs] Unclear description of indices arg in torch.index_put_,https://github.com/pytorch/pytorch/issues/34063,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-03-06T00:19:29Z +issue,34052,Model loaded in C++ runtime is not thread safe,https://github.com/pytorch/pytorch/issues/34052,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: sanitizers",scan when time allows,2020-03-05T22:18:37Z +issue,34082,Autograd deep copy avoidance optimization unsound in the presence of views,https://github.com/pytorch/pytorch/issues/34082,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-03-04T22:56:38Z +issue,33898,Docs for uniform_ don't make any sense,https://github.com/pytorch/pytorch/issues/33898,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-03-04T21:33:46Z +issue,34172,END_HANDLE_TH_ERRORS_PYBIND prevents pybind11 Exception translation,https://github.com/pytorch/pytorch/issues/34172,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2020-03-04T04:17:49Z +issue,20963,official libtorch static build zip file error,https://github.com/pytorch/pytorch/issues/20963,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2020-03-03T23:24:49Z +issue,18906,Provide option to use alias method in Categorical.sample(),https://github.com/pytorch/pytorch/issues/18906,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-03-03T21:38:41Z +issue,34086,Enable profiler tracing tests on windows,https://github.com/pytorch/pytorch/issues/34086,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-03-03T16:11:43Z +issue,34068,[C++ API Parity] Incorrect documentation for optim initialization in serialization docs,https://github.com/pytorch/pytorch/issues/34068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs, module: optimizer",scan when time allows,2020-03-02T20:44:19Z +issue,34080,RPC API Changes for TensorPipes,https://github.com/pytorch/pytorch/issues/34080,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-03-02T20:21:39Z +issue,34001,_broadcast_coalesced_reshape doesn't respect zero-size tensor,https://github.com/pytorch/pytorch/issues/34001,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-03-02T19:39:21Z +issue,33991,TensorPipes RPC Agent Default Args/Result Device Mapping,https://github.com/pytorch/pytorch/issues/33991,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-03-02T19:36:59Z +issue,33990,TensorPipes RPC Agent Multiple Placement Retries,https://github.com/pytorch/pytorch/issues/33990,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-03-02T19:04:41Z +issue,33989,TensorPipes RPC Agent Message Acknowledgements,https://github.com/pytorch/pytorch/issues/33989,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-03-02T19:04:25Z +issue,33983,TensorPipes RPC Agent Listener shortcut,https://github.com/pytorch/pytorch/issues/33983,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-03-02T19:04:05Z +issue,33982,Initialize TensorPipe RPC Agent Transport,https://github.com/pytorch/pytorch/issues/33982,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-03-02T19:02:20Z +issue,33979,TensorPipes RPC Agent CUDA Support,https://github.com/pytorch/pytorch/issues/33979,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: rpc",scan when time allows,2020-03-02T19:00:44Z +issue,17658,C++ API: Crash in cudnnDestroy() when deconstructing,https://github.com/pytorch/pytorch/issues/17658,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: abi, module: cpp, module: cudnn",scan when time allows,2020-03-02T17:15:45Z +issue,17703,Training hangs when using DistributedDataParallel in two pod on two nodes,https://github.com/pytorch/pytorch/issues/17703,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-03-01T07:16:52Z +issue,2545,Counter-intuitive Patience & Cooldown of ReduceLROnPlateau,https://github.com/pytorch/pytorch/issues/2545,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-02-28T23:15:08Z +issue,24915,Shared Dataset Functionality,https://github.com/pytorch/pytorch/issues/24915,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-02-28T21:28:17Z +issue,30635,Cannot use torch.jit.script with nn.DataParallel,https://github.com/pytorch/pytorch/issues/30635,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: jit",scan when time allows,2020-02-28T20:25:04Z +issue,33887,[jit] Python arg parser / TorchScript incompatibilities,https://github.com/pytorch/pytorch/issues/33887,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:57:00Z +issue,33844,[jit] Returning different types with `Any` segfaults,https://github.com/pytorch/pytorch/issues/33844,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:57:00Z +issue,33782,[JIT] BroadcastingList annotations don't work with ignore'd functions,https://github.com/pytorch/pytorch/issues/33782,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:56:59Z +issue,33501,"When a Node fails to resolve to an Operator, print out the types of arguments, and all ""close matches"" in known operators",https://github.com/pytorch/pytorch/issues/33501,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:56:59Z +issue,32544,[jit] Use `typing.get_type_hints` instead of parsing types manually,https://github.com/pytorch/pytorch/issues/32544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:56:58Z +issue,30788,Master Task: JIT/C++ Parity With Pytorch Python API,https://github.com/pytorch/pytorch/issues/30788,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:56:58Z +issue,33353,[jit] `Node` callstack is incorrect,https://github.com/pytorch/pytorch/issues/33353,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:54:57Z +issue,32751,PyTorch freezes on second call to scripted densenet model from torchvision,https://github.com/pytorch/pytorch/issues/32751,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: vision, oncall: jit",scan when time allows,2020-02-28T18:54:55Z +issue,32690,How to customize build torchscript model to be used in end devices codebase,https://github.com/pytorch/pytorch/issues/32690,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",scan when time allows,2020-02-28T18:54:55Z +issue,32565,Reduce RPC branches for Python/Built-inOp/TorchScript,https://github.com/pytorch/pytorch/issues/32565,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: jit",scan when time allows,2020-02-28T18:54:54Z +issue,32463,[JIT] Make `torch.jit.script` work on all objects which we can represent as IValues,https://github.com/pytorch/pytorch/issues/32463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:54:54Z +issue,32337,Save LocalResponceNorm as a single aten node when convert a pytorch model to jit model,https://github.com/pytorch/pytorch/issues/32337,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:54:54Z +issue,32253,[TorchScript] Device comparison implemented in eq but not ne,https://github.com/pytorch/pytorch/issues/32253,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:54:53Z +issue,30138,Unable to register custom JIT Operator with AliasAnalysisKind::CONSERVATIVE,https://github.com/pytorch/pytorch/issues/30138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:54:52Z +issue,28308,jit script fails with `AttributeError: 'str' object has no attribute 'lineno'`,https://github.com/pytorch/pytorch/issues/28308,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:54:51Z +issue,27479,[JIT] Figure out how to easily investigate memory usage issues issues,https://github.com/pytorch/pytorch/issues/27479,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:54:51Z +issue,31953,JIT tests are linked directly into libtorch and register operators even when unused,https://github.com/pytorch/pytorch/issues/31953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:51Z +issue,31882,Example cmakelists for custom cuda operator?,https://github.com/pytorch/pytorch/issues/31882,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:51Z +issue,31574,attribute and register_buffer are not the same on gpu,https://github.com/pytorch/pytorch/issues/31574,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:50Z +issue,31551,TorchBind broken on rocm,https://github.com/pytorch/pytorch/issues/31551,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:50Z +issue,31430,[jit] `del` with slices doesn't work,https://github.com/pytorch/pytorch/issues/31430,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:49Z +issue,31386,[jit] Python type hints in TorchScript classes don't work,https://github.com/pytorch/pytorch/issues/31386,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:49Z +issue,31202,Request for Lint Pass to Detect Modification on Parameters/Attributes during TorchScript Inference,https://github.com/pytorch/pytorch/issues/31202,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:49Z +issue,30226,Wrong substitution of aten::to,https://github.com/pytorch/pytorch/issues/30226,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: oncall: jit, oncall: mobile",scan when time allows,2020-02-28T18:52:48Z +issue,30049,CPU Memory Leak for JIT ScriptModule on DataParallel,https://github.com/pytorch/pytorch/issues/30049,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:47Z +issue,29987,Move the attributes of a module to the given device,https://github.com/pytorch/pytorch/issues/29987,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:47Z +issue,29822,[jit] Traced `cat` on GPU doesn't support negative indexing,https://github.com/pytorch/pytorch/issues/29822,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:46Z +issue,29750,[jit] Printing the graph doesn't include function calls,https://github.com/pytorch/pytorch/issues/29750,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:46Z +issue,29642,error: ‘struct torch::jit::RegisterOperators’ has no member named ‘op’,https://github.com/pytorch/pytorch/issues/29642,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:45Z +issue,29597,[jit] Script class attributes aren't automatically added,https://github.com/pytorch/pytorch/issues/29597,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:18Z +issue,29063,Trace of torch.tensor is impressively convoluted,https://github.com/pytorch/pytorch/issues/29063,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:17Z +issue,28549,Expose DifferentiableGraphBackward to python,https://github.com/pytorch/pytorch/issues/28549,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, oncall: jit",scan when time allows,2020-02-28T18:52:17Z +issue,28519,[jit] C++ Documentation,https://github.com/pytorch/pytorch/issues/28519,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:16Z +issue,28517,[JIT] __repr__ support for ScriptModules,https://github.com/pytorch/pytorch/issues/28517,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:16Z +issue,28307,[jit] scripted module and user defined methods,https://github.com/pytorch/pytorch/issues/28307,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:16Z +issue,27505,[JIT] ndimension not supported,https://github.com/pytorch/pytorch/issues/27505,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:14Z +issue,27497,TSAN failure related to mkldnn,https://github.com/pytorch/pytorch/issues/27497,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, oncall: jit",scan when time allows,2020-02-28T18:52:13Z +issue,27475,[JIT] script::Module API parity with nnmodule,https://github.com/pytorch/pytorch/issues/27475,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:13Z +issue,27406,[jit] String frontend doesn't support default arg values,https://github.com/pytorch/pytorch/issues/27406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:13Z +issue,27003,[JIT] Script doesn't preserve builtin torch named tuples upon python return,https://github.com/pytorch/pytorch/issues/27003,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:52:12Z +issue,26086,[Feature Request] Trace / Script C++ models,https://github.com/pytorch/pytorch/issues/26086,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: jit",scan when time allows,2020-02-28T18:51:43Z +issue,25803,[jit] `random` module support,https://github.com/pytorch/pytorch/issues/25803,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:51:43Z +issue,25047,ScriptModule and nn.Module parameter ordering difference,https://github.com/pytorch/pytorch/issues/25047,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:51:41Z +issue,24248,Loading custom Torchscript C++ operators in python segfaults due to ABI compatibility issue between pytorch and libtorch,https://github.com/pytorch/pytorch/issues/24248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:51:41Z +issue,24229,Confusing error message for Custom Class type mismatch,https://github.com/pytorch/pytorch/issues/24229,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:51:40Z +issue,24225,Allow forward method to be defined with .define() in new TorchScript API,https://github.com/pytorch/pytorch/issues/24225,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:51:40Z +issue,23958,[jit] Python @property's not supported in TorchScript,https://github.com/pytorch/pytorch/issues/23958,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:51:40Z +issue,23890,[JIT] Can't use ndim in script,https://github.com/pytorch/pytorch/issues/23890,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-28T18:51:39Z +issue,24176,torch.fft crash when used with nn.DataParallel,https://github.com/pytorch/pytorch/issues/24176,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: data parallel",scan when time allows,2020-02-28T16:24:11Z +issue,33367,Expose `internal::GRAIN_SIZE` through Python API.,https://github.com/pytorch/pytorch/issues/33367,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2020-02-27T02:30:13Z +issue,20343,torch.distributions.Binomial.sample() uses a massive amount of memory,https://github.com/pytorch/pytorch/issues/20343,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: memory usage",scan when time allows,2020-02-26T23:16:24Z +issue,29548,Dispatch key reorganization,https://github.com/pytorch/pytorch/issues/29548,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-02-26T21:28:27Z +issue,2512,detach_() variant that affects all past uses too,https://github.com/pytorch/pytorch/issues/2512,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-02-26T18:18:19Z +issue,26447,parallel_for may hang when called in main process and then on daemon process,https://github.com/pytorch/pytorch/issues/26447,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deadlock, module: multiprocessing",scan when time allows,2020-02-25T21:09:07Z +issue,26691,Support FPGA Xilinx,https://github.com/pytorch/pytorch/issues/26691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: backend,scan when time allows,2020-02-25T20:32:16Z +issue,33482,Distributed Data Parallel for computation graphs that make RPCs in forward(),https://github.com/pytorch/pytorch/issues/33482,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2020-02-25T18:19:11Z +issue,33738,Support pipelining the backward pass and optimizer.step() for distributed autograd.,https://github.com/pytorch/pytorch/issues/33738,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2020-02-25T04:48:16Z +issue,33725,Tensor.random_ is not implemented for bfloat16 on CPU(but implemented on CUDA),https://github.com/pytorch/pytorch/issues/33725,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: random",scan when time allows,2020-02-25T00:55:48Z +issue,32510,Tensor.random_ is not implemented for bool on CUDA(but implemented on CPU),https://github.com/pytorch/pytorch/issues/32510,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: random",scan when time allows,2020-02-24T22:13:19Z +issue,24130,[RPC] Add type annotations for RPC-related Python files,https://github.com/pytorch/pytorch/issues/24130,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-02-24T19:02:18Z +issue,28055,Traced resnet101 leaks memory during `forward`,https://github.com/pytorch/pytorch/issues/28055,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2020-02-24T18:24:20Z +issue,33691,Some module has incorrect scope in complex naming situations in tensorboard graph export,https://github.com/pytorch/pytorch/issues/33691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-02-24T15:52:06Z +issue,33684,Decouple Lifetime of Local RRef and RPC,https://github.com/pytorch/pytorch/issues/33684,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-02-24T04:07:58Z +issue,33670,Some module info is missing in nested graph for tensorboard,https://github.com/pytorch/pytorch/issues/33670,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2020-02-23T19:35:03Z +issue,33664,torch.rand() not having same values on using torch.manual_seed(0),https://github.com/pytorch/pytorch/issues/33664,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2020-02-23T19:29:12Z +issue,33653,master build error,https://github.com/pytorch/pytorch/issues/33653,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-23T19:21:39Z +issue,33436,It seems nn.Sequential.add_module() could take duplicate names,https://github.com/pytorch/pytorch/issues/33436,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-02-23T10:58:32Z +issue,33360,Dropout of attention weights in function F.multi_head_attention_forward() breaks sum-to-1 constraint,https://github.com/pytorch/pytorch/issues/33360,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-02-21T21:17:43Z +issue,33611,Resnet 32x32 in caffe2 generates ResNet with wrong number of layers,https://github.com/pytorch/pytorch/issues/33611,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-21T15:35:48Z +issue,32576,Document different gradient wrt to jax when nesting,https://github.com/pytorch/pytorch/issues/32576,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: docs",scan when time allows,2020-02-20T20:49:29Z +issue,6613,[caffe2] benchmark performance for different operators,https://github.com/pytorch/pytorch/issues/6613,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-20T14:42:49Z +issue,33476,Make ArrayRef::size() return int64_t rather than size_t,https://github.com/pytorch/pytorch/issues/33476,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-02-20T03:23:36Z +issue,33477,"Add 32-bit CI (e.g., Raspberry PI CI)",https://github.com/pytorch/pytorch/issues/33477,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2020-02-20T03:22:53Z +issue,33514,"Make setter non-optional, e.g., TensorOptions::device(optional) -> device(Device), and add a device_opt setter",https://github.com/pytorch/pytorch/issues/33514,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-02-20T03:18:39Z +issue,4825,[feature request]Support AVX512F intrinstics to vectorize operations,https://github.com/pytorch/pytorch/issues/4825,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2020-02-20T00:07:25Z +issue,2407,Hard-negative mining using __getitem__ directive in Dataset class,https://github.com/pytorch/pytorch/issues/2407,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-02-19T21:24:04Z +issue,2466,Autograd test failure on ppc64le,https://github.com/pytorch/pytorch/issues/2466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: power,scan when time allows,2020-02-19T21:22:39Z +issue,28093,Move options logic in _like functions from parsing layer to implementations,https://github.com/pytorch/pytorch/issues/28093,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-19T18:40:14Z +issue,33396,PyTorch 1.4.0 does not support using `Module` or `ModuleList` in attribute annotations in ScriptModule,https://github.com/pytorch/pytorch/issues/33396,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-18T18:30:43Z +issue,33413,IterableDataset with num_workers > 0 and drop_last=True drops more instances than expected,https://github.com/pytorch/pytorch/issues/33413,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-02-18T18:13:31Z +issue,33363,Make it easier to add new messages in RPC layer,https://github.com/pytorch/pytorch/issues/33363,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-02-18T16:55:55Z +issue,33390,It is not good to separate the steps of modules making and forward computation,https://github.com/pytorch/pytorch/issues/33390,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-18T15:58:12Z +issue,21016,[utils.bottleneck] throws initialization error for cuda profiling,https://github.com/pytorch/pytorch/issues/21016,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-02-18T00:13:21Z +issue,32924,Connect timeout feature do not work in DDP with TCPStore,https://github.com/pytorch/pytorch/issues/32924,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-02-17T18:35:49Z +issue,33114,Move the custom pass execution back to the beginning of runNondiffOptimization,https://github.com/pytorch/pytorch/issues/33114,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-16T19:26:23Z +issue,32490,"RuntimeError: has_marked_unused_parameters_ INTERNAL ASSERT FAILED at /pytorch/torch/csrc/distributed/c10d/reducer.cpp:290, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/32490,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-02-16T00:09:59Z +issue,32390,[JIT] Compile group of functions/modules,https://github.com/pytorch/pytorch/issues/32390,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-15T16:05:57Z +issue,33342,malloc(): memory corruption (fast),https://github.com/pytorch/pytorch/issues/33342,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-02-14T18:11:46Z +issue,33341,how-to-adjust-learning-rate using libtorch,https://github.com/pytorch/pytorch/issues/33341,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-14T17:57:33Z +issue,31038,Pytorch 1.3.0 on RTX cards: CUDA error: an illegal memory access was encountered,https://github.com/pytorch/pytorch/issues/31038,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2020-02-13T21:06:50Z +issue,33303,torch::var_out and dimnames,https://github.com/pytorch/pytorch/issues/33303,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: named tensor",scan when time allows,2020-02-13T19:50:28Z +issue,33301,Don't take TensorOptions by reference,https://github.com/pytorch/pytorch/issues/33301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2020-02-13T18:59:39Z +issue,33249,Core dumps being created when running test_c10d.py and test_multiprocessing_spawn.py,https://github.com/pytorch/pytorch/issues/33249,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-02-13T16:43:26Z +issue,33241,"[feature request] Add ""groups"" argument to nn.Fold and nn.Unfold",https://github.com/pytorch/pytorch/issues/33241,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-02-13T08:28:41Z +issue,33180,Loading pretrained model,https://github.com/pytorch/pytorch/issues/33180,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-12T02:17:00Z +issue,31708,Computing dot product of columns sliced from large matrix causes illegal memory access in CUDA,https://github.com/pytorch/pytorch/issues/31708,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cublas, module: cuda, module: dependency bug",scan when time allows,2020-02-12T01:17:20Z +issue,2312,DataParallel is not compatible with pack_padded_sequence,https://github.com/pytorch/pytorch/issues/2312,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2020-02-11T18:58:23Z +issue,30291,[feature request] A way to restore/assign tensor _version,https://github.com/pytorch/pytorch/issues/30291,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-02-11T17:11:03Z +issue,33051,Recover from CUDA runtime error,https://github.com/pytorch/pytorch/issues/33051,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: third_party",scan when time allows,2020-02-11T02:14:42Z +issue,5900,"Build Fails on Gentoo with CUDA 9.1, GCC 6.4, Python 3.5",https://github.com/pytorch/pytorch/issues/5900,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-02-11T01:53:14Z +issue,26109,Support the AVX512 runtime dispatch,https://github.com/pytorch/pytorch/issues/26109,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-10T23:57:56Z +issue,31866,Negative indices in chunk could cause Out of Range access on loss.backward in JIT,https://github.com/pytorch/pytorch/issues/31866,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-10T18:39:00Z +issue,33115,[RFC] Add ability to get all remote parameters when constructing DistributedOptimizer.,https://github.com/pytorch/pytorch/issues/33115,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-02-10T12:01:25Z +issue,33081,DataParallel gives different gradients when using LSTMs,https://github.com/pytorch/pytorch/issues/33081,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: determinism",scan when time allows,2020-02-10T11:43:43Z +issue,33049,Warning when link libtorch and opencv4.2.0 together,https://github.com/pytorch/pytorch/issues/33049,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-02-10T11:32:08Z +issue,32406,KL divergence for diagonal Gaussian distributions,https://github.com/pytorch/pytorch/issues/32406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-02-10T10:14:15Z +issue,27610,[FR][RFC] Build a serving framework to host and serve trained PyTorch models,https://github.com/pytorch/pytorch/issues/27610,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-02-08T12:34:18Z +issue,22778,In-place updating the original value tensor should also update version counter of sparse tensor's values_ tensor,https://github.com/pytorch/pytorch/issues/22778,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: sparse",scan when time allows,2020-02-07T23:03:09Z +issue,32884,[RFC] Nested scopes in autograd profiler should support RPC calls properly.,https://github.com/pytorch/pytorch/issues/32884,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: rpc",scan when time allows,2020-02-07T20:55:14Z +issue,25941,TestAutograd.test_deep_reentrant fails with SIGBUS on macOS,https://github.com/pytorch/pytorch/issues/25941,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: macos, module: tests",scan when time allows,2020-02-07T20:17:54Z +issue,31185,Retain Subgraph or Save Intermediate Grad support?,https://github.com/pytorch/pytorch/issues/31185,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2020-02-07T19:27:15Z +issue,33007,Upper/Lower attributes for named dimensions for proper Ricci notation and to generalize matrix operations,https://github.com/pytorch/pytorch/issues/33007,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-02-07T19:06:29Z +issue,33006,matmul: no warning when contracting differently named dimensions,https://github.com/pytorch/pytorch/issues/33006,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2020-02-07T19:06:03Z +issue,29804,Parallel data loader performance degradation for IterableDataset with num_workers > 1 (but not for Dataset).,https://github.com/pytorch/pytorch/issues/29804,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-02-07T09:46:41Z +issue,33034,Scripting fails to preserve attribute aliasing,https://github.com/pytorch/pytorch/issues/33034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-06T23:22:53Z +issue,3790,"Add SGDR, SGDW, AdamW and AdamWR",https://github.com/pytorch/pytorch/issues/3790,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2020-02-06T21:52:44Z +issue,30380,Add Google Colab as an option on the 'Get Started' page.,https://github.com/pytorch/pytorch/issues/30380,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-02-05T11:44:26Z +issue,32972,"which pytorch do i install for running this project in my windows.link for this is :: https://github.com/xiaojunxu/SQLNet .IN this they have used python 2.7 ,but i am unable to install pytorch on python 2.7 environment .help me with this",https://github.com/pytorch/pytorch/issues/32972,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2020-02-05T07:55:31Z +issue,32754,JIT performance discrepancies,https://github.com/pytorch/pytorch/issues/32754,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: jit",scan when time allows,2020-02-04T19:37:45Z +issue,32840,Can we add support for Enum in scripted models?,https://github.com/pytorch/pytorch/issues/32840,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-02-03T22:03:55Z +issue,32698,Add Julia Bindings to Torch backend,https://github.com/pytorch/pytorch/issues/32698,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: language binding,scan when time allows,2020-02-03T18:25:19Z +issue,32687,Building PyTorch ignores my current Python version,https://github.com/pytorch/pytorch/issues/32687,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-02-03T18:17:32Z +issue,31528,cuCtxGetDevice error and seg fault with DDP and OpenMPI,https://github.com/pytorch/pytorch/issues/31528,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, oncall: distributed",scan when time allows,2020-01-31T01:50:55Z +issue,24835,NetworkX's Version,https://github.com/pytorch/pytorch/issues/24835,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-30T15:23:37Z +issue,32784,Implement Backend-Agnostic RPC functionality in RpcAgent,https://github.com/pytorch/pytorch/issues/32784,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-01-29T20:23:30Z +issue,32783,Comments Separating Class Methods from Different Classes in C++ Files,https://github.com/pytorch/pytorch/issues/32783,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-29T19:54:10Z +issue,32767,Documentation is not loaded by IDEs,https://github.com/pytorch/pytorch/issues/32767,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-01-29T19:38:04Z +issue,32752,RPC mock mode for unit tests.,https://github.com/pytorch/pytorch/issues/32752,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-01-29T18:56:58Z +issue,31752,Need a launch utility for Distributed RPC framework.,https://github.com/pytorch/pytorch/issues/31752,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-01-29T06:53:00Z +issue,30405,Codegen refactoring master task,https://github.com/pytorch/pytorch/issues/30405,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-28T21:12:09Z +issue,27517,Allow explicit gradients in torch.distributed.autograd.backward() API,https://github.com/pytorch/pytorch/issues/27517,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-01-28T02:05:34Z +issue,32452,Numba Enhancement Proposal (NBEP) 7: External Memory Management Plugins,https://github.com/pytorch/pytorch/issues/32452,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-27T22:46:21Z +issue,32098,torch.gt ge lt le,https://github.com/pytorch/pytorch/issues/32098,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: type promotion,scan when time allows,2020-01-27T22:46:09Z +issue,22577,Double backward 3 times slower for conv2d with padding = 1,https://github.com/pytorch/pytorch/issues/22577,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: dependency bug, module: double backwards, module: nn",scan when time allows,2020-01-27T17:48:59Z +issue,18643,Memory not being deallocated in backward(),https://github.com/pytorch/pytorch/issues/18643,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage",scan when time allows,2020-01-27T17:47:42Z +issue,32580,SGD optimizer with deprecation warning,https://github.com/pytorch/pytorch/issues/32580,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: deprecation, module: optimizer",scan when time allows,2020-01-24T17:17:33Z +issue,32504,LibTorch operates very slowly on data blobs from GPU,https://github.com/pytorch/pytorch/issues/32504,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2020-01-23T18:34:55Z +issue,32488,TracerWarning When Using Tensor Size in Torchscript Trace,https://github.com/pytorch/pytorch/issues/32488,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-01-22T22:20:29Z +issue,32511,TorchScript C++ API Tracking Issue,https://github.com/pytorch/pytorch/issues/32511,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-01-22T21:41:40Z +issue,32419,MAGMA libraries,https://github.com/pytorch/pytorch/issues/32419,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2020-01-22T13:42:16Z +issue,32227,"0 INTERNAL ASSERT FAILED at /pytorch/c10/util/intrusive_ptr.h:348, please report a bug to PyTorch.",https://github.com/pytorch/pytorch/issues/32227,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: assert failure,scan when time allows,2020-01-22T10:51:57Z +issue,32403,How to accelerate the compiling of pytorch,https://github.com/pytorch/pytorch/issues/32403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-01-21T23:25:36Z +issue,32416,Is there a way to use SYSTEM_INSTALLED 3rdparty libraries?,https://github.com/pytorch/pytorch/issues/32416,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-01-21T23:07:03Z +issue,32427,[C++] Don't use DeprecatedTypeProperties in torch::utils::reorder_tensors_like,https://github.com/pytorch/pytorch/issues/32427,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2020-01-21T19:19:47Z +issue,32415,DistributedStreamSampler: support stream sampler in distributed setting,https://github.com/pytorch/pytorch/issues/32415,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: distributed",scan when time allows,2020-01-21T19:16:32Z +issue,32402,torchvision,https://github.com/pytorch/pytorch/issues/32402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2020-01-21T19:14:54Z +issue,32372,Migrate processGroup::async_work to use Future,https://github.com/pytorch/pytorch/issues/32372,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2020-01-21T19:12:25Z +issue,19125,Multi-gpu via torch::nn::parallel::data_parallel,https://github.com/pytorch/pytorch/issues/19125,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: distributed",scan when time allows,2020-01-21T16:10:03Z +issue,32036,Internal assert failed using multiple GPUs with DataParallel,https://github.com/pytorch/pytorch/issues/32036,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: cuda, module: data parallel",scan when time allows,2020-01-20T18:43:38Z +issue,31937,RPC and dist_autograd should respect no_grad mode,https://github.com/pytorch/pytorch/issues/31937,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-01-20T02:23:44Z +issue,32357,Pytorch JIT Compilation Does Not Finish (Infinite Loop?) for Deeper Models,https://github.com/pytorch/pytorch/issues/32357,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-01-18T18:31:27Z +issue,32375,Make _rpc_sync_torchscript and _rpc_async_torchscript work with autograd profiler,https://github.com/pytorch/pytorch/issues/32375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-01-18T07:36:19Z +issue,32363,Possible ProcessGroup::Work::abort correctness issue,https://github.com/pytorch/pytorch/issues/32363,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: distributed",scan when time allows,2020-01-17T19:34:07Z +issue,32269,Support strategy to train large model that exceeds GPU mem and DRAM mem,https://github.com/pytorch/pytorch/issues/32269,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-01-17T03:13:32Z +issue,32264,torch.nn.functional.threshold not work with LongTensor,https://github.com/pytorch/pytorch/issues/32264,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-01-16T23:07:08Z +issue,11612,[feature request] Triangular Matrix Representation,https://github.com/pytorch/pytorch/issues/11612,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-16T17:01:41Z +issue,32237,PyTorch C++ API docs only tracks master branch,https://github.com/pytorch/pytorch/issues/32237,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: doc infra",scan when time allows,2020-01-15T20:13:35Z +issue,32132,Slighty out of tolerance for `test_mv` and `test_cholesky_solve_batched_cuda_float64`,https://github.com/pytorch/pytorch/issues/32132,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2020-01-15T18:43:11Z +issue,32088,Batch convolutional layer with 5d weight tensor that is not contiguous,https://github.com/pytorch/pytorch/issues/32088,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-15T17:25:36Z +issue,32087,Out-of-date link to pytorch ci dockerfiles in jenkins readme,https://github.com/pytorch/pytorch/issues/32087,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-01-15T17:13:34Z +issue,32219,[Feature proposal] improved algorithms for checkpointing,https://github.com/pytorch/pytorch/issues/32219,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-15T16:48:59Z +issue,32005,gradients inside gradient checkpoint,https://github.com/pytorch/pytorch/issues/32005,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: autograd",scan when time allows,2020-01-15T16:41:56Z +issue,32105,TensorRT: CheckDims() need adjustment for EXPLICIT_BATCH?,https://github.com/pytorch/pytorch/issues/32105,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-14T20:08:57Z +issue,32141,We should change DeprecationWarnings to UserWarnings in 27361,https://github.com/pytorch/pytorch/issues/32141,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-01-14T19:47:04Z +issue,32150,pin_memory may change the type of instance returned by collate_fn.,https://github.com/pytorch/pytorch/issues/32150,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-01-14T19:42:51Z +issue,32165,Serialization inconsistency with pickling tensors breaks caching,https://github.com/pytorch/pytorch/issues/32165,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2020-01-14T19:31:54Z +issue,32023,Inaccurate batched GRU results on CPU,https://github.com/pytorch/pytorch/issues/32023,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: nn, module: numerical-reproducibility, module: rnn",scan when time allows,2020-01-14T17:17:28Z +issue,32033,run detectron test test_spatial_narrow_as_op.py fails,https://github.com/pytorch/pytorch/issues/32033,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2020-01-14T17:16:12Z +issue,30569,"Assertion `index >= -sizes[i] && index < sizes[i] && ""index out of bounds""` failed.",https://github.com/pytorch/pytorch/issues/30569,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",scan when time allows,2020-01-13T23:02:21Z +issue,30387,[docs] Missing docs for Storage.from_buffer,https://github.com/pytorch/pytorch/issues/30387,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2020-01-11T07:05:58Z +issue,20967,Track running stats regardless of track_running_stats=False,https://github.com/pytorch/pytorch/issues/20967,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-01-10T14:43:30Z +issue,31907,Compile libtorch by source code failed.,https://github.com/pytorch/pytorch/issues/31907,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cpp",scan when time allows,2020-01-09T22:52:25Z +issue,31779,torch.poisson returns floating point tensor,https://github.com/pytorch/pytorch/issues/31779,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2020-01-09T18:54:31Z +issue,31979,Reuse spawned subprocesses in RPC tests,https://github.com/pytorch/pytorch/issues/31979,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2020-01-09T15:50:30Z +issue,16901,Issue with dataloader using pin_memory = True,https://github.com/pytorch/pytorch/issues/16901,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-01-08T19:18:56Z +issue,31867,The model training time is increasing between runs if the same DataLoader reused to train multiple models.,https://github.com/pytorch/pytorch/issues/31867,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2020-01-08T19:04:39Z +issue,31842,Support rpc/remote torch script call with script class/module name and class/module method name,https://github.com/pytorch/pytorch/issues/31842,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: rpc, oncall: jit",scan when time allows,2020-01-08T18:54:03Z +issue,31487,Disable PSIMD? Why Pytorch is trying to download PSimd when PSIMD_SOURCE_DIR is defined?,https://github.com/pytorch/pytorch/issues/31487,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2020-01-06T19:51:23Z +issue,31688,Parallelization: more balanced work distribution among workers,https://github.com/pytorch/pytorch/issues/31688,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: data parallel",scan when time allows,2020-01-06T18:26:15Z +issue,31107,Add torch.version.nccl,https://github.com/pytorch/pytorch/issues/31107,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: nccl, oncall: distributed",scan when time allows,2020-01-06T10:36:29Z +issue,31651,"update embedding at indices, other than those passed as input, in the case of sparse tensors",https://github.com/pytorch/pytorch/issues/31651,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sparse",scan when time allows,2020-01-04T12:30:36Z +issue,30404,affine_grid CUDA / cuDNN support for Half removed in 1.3.x,https://github.com/pytorch/pytorch/issues/30404,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2020-01-04T00:15:20Z +issue,31422,Support DataParallel with PackedSequence,https://github.com/pytorch/pytorch/issues/31422,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: data parallel, oncall: distributed",scan when time allows,2020-01-02T21:54:50Z +issue,31356,Optimizing DLRM for CPU,https://github.com/pytorch/pytorch/issues/31356,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2020-01-02T14:56:09Z +issue,31425,cuDNN convolution does not handle empty input tensor,https://github.com/pytorch/pytorch/issues/31425,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2020-01-02T11:41:51Z +issue,31697,the cmake problem with build from source ?,https://github.com/pytorch/pytorch/issues/31697,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2020-01-02T10:56:33Z +issue,6442,[feature request] dropout1d,https://github.com/pytorch/pytorch/issues/6442,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2020-01-02T10:18:42Z +issue,31259,Script method can't call a scripted function when it is decorated with `@torch.no_grad`,https://github.com/pytorch/pytorch/issues/31259,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2020-01-02T07:48:56Z +issue,31553,allow setting different batch size splits for data_parallel.py and distributed.py,https://github.com/pytorch/pytorch/issues/31553,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, oncall: distributed",scan when time allows,2020-01-01T04:34:03Z +issue,26802,Google Summer of Code,https://github.com/pytorch/pytorch/issues/26802,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-12-31T19:57:08Z +issue,31685,What is the significance of torchvision._is_tracing()?,https://github.com/pytorch/pytorch/issues/31685,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2019-12-30T21:50:08Z +issue,31657,Use of Sequence collections for abstract classes in Dataset,https://github.com/pytorch/pytorch/issues/31657,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-12-27T18:05:48Z +issue,31321,Ability to download docs HTML for offline use,https://github.com/pytorch/pytorch/issues/31321,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-12-27T09:42:49Z +issue,20006,"RuntimeError: invalid argument 10: ldb should be at least max(1, 0), but have 0 at ../aten/src/TH/generic/THBlas.cpp:36",https://github.com/pytorch/pytorch/issues/20006,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-12-26T15:45:39Z +issue,31606,No auto-suggest capacity for Transformer,https://github.com/pytorch/pytorch/issues/31606,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-12-26T04:51:26Z +issue,31557,[docs] F.ctc_loss docs to warn clearly about invalid inf-causing inputs; zero_infinity to become enabled by default,https://github.com/pytorch/pytorch/issues/31557,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-12-25T13:23:23Z +issue,31596,pytorch forward hangs in multiprocess environment,https://github.com/pytorch/pytorch/issues/31596,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2019-12-25T00:21:28Z +issue,29697,MultiStepLR does not return good lr after load_state_dict,https://github.com/pytorch/pytorch/issues/29697,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-12-23T17:07:01Z +issue,24963,torch.cuda.synchronize blocks CUDA execution on other threads using other devices.,https://github.com/pytorch/pytorch/issues/24963,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-12-23T13:18:23Z +issue,31548,"c++ PReLUFuncOptions declared, not used or valid",https://github.com/pytorch/pytorch/issues/31548,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-12-23T07:52:12Z +issue,1861,Factorized Output Layer,https://github.com/pytorch/pytorch/issues/1861,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-12-23T07:35:56Z +issue,19668,Importing matlab.engine after torch causes bad_alloc,https://github.com/pytorch/pytorch/issues/19668,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: internals",scan when time allows,2019-12-20T07:59:22Z +issue,30803,Pytorch openmp thread number tuning option for CPU trainning,https://github.com/pytorch/pytorch/issues/30803,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multithreading, module: performance",scan when time allows,2019-12-19T17:50:10Z +issue,31471,Mnasnet0_5 first layer shape incorrect,https://github.com/pytorch/pytorch/issues/31471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2019-12-19T17:05:18Z +issue,31300,torch runtime error when manual link libmkldnn.so,https://github.com/pytorch/pytorch/issues/31300,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn",scan when time allows,2019-12-19T15:48:12Z +issue,31289,Intel OMP multiprocessing assertion failure: Assertion failure at z_Linux_util.cpp(2338),https://github.com/pytorch/pytorch/issues/31289,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dependency bug, module: multiprocessing",scan when time allows,2019-12-19T09:17:29Z +issue,31228,[RPC] Support nn.Module pickling with share memory,https://github.com/pytorch/pytorch/issues/31228,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-12-19T09:10:53Z +issue,19163,LayerNorm is very slow (almost frozen) in CPU of multiprocessing,https://github.com/pytorch/pytorch/issues/19163,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2019-12-19T03:09:57Z +issue,31391,"Error ""builtin cannot be used as a value"" when add Python snippets in C++",https://github.com/pytorch/pytorch/issues/31391,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-18T20:49:44Z +issue,31360,DDP/MP not yielding nontrivial speedup,https://github.com/pytorch/pytorch/issues/31360,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: multiprocessing, module: performance",scan when time allows,2019-12-17T21:32:39Z +issue,31367,`clip_grad_norm` allows negative `max_norm` values,https://github.com/pytorch/pytorch/issues/31367,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-12-17T18:25:28Z +issue,31353,Memory management is inefficient which limits performance,https://github.com/pytorch/pytorch/issues/31353,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage, module: performance",scan when time allows,2019-12-17T16:18:53Z +issue,22954,Support serializing IValue to bytes (and deserialize from bytes),https://github.com/pytorch/pytorch/issues/22954,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-17T05:07:12Z +issue,4959,Speed up data loading for `TensorDataset` if the underlying dataset supports index by a list of indices,https://github.com/pytorch/pytorch/issues/4959,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: performance",scan when time allows,2019-12-16T22:37:22Z +issue,31308,[mac] Failure to import torch,https://github.com/pytorch/pytorch/issues/31308,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: macos,scan when time allows,2019-12-16T17:52:34Z +issue,28194,New Stochastic Optimization Algorithms in Pytorch,https://github.com/pytorch/pytorch/issues/28194,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-12-16T00:06:52Z +issue,21144,Data Parallel Implementation Improvements,https://github.com/pytorch/pytorch/issues/21144,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-12-15T22:04:46Z +issue,27144,torch::jit::script::Module has no zero_grad(),https://github.com/pytorch/pytorch/issues/27144,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-13T19:01:12Z +issue,31110,torch.nn.Softplus threshold argument bug?,https://github.com/pytorch/pytorch/issues/31110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-12-12T23:40:17Z +issue,17666,Building pytorch on ARM failed,https://github.com/pytorch/pytorch/issues/17666,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-12-12T19:59:24Z +issue,31191,[JIT] Slice with optional not supported,https://github.com/pytorch/pytorch/issues/31191,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-12T18:27:33Z +issue,27409,Installer not setting rpath for MAGMA (OS X w/ GPU),https://github.com/pytorch/pytorch/issues/27409,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda, module: macos",scan when time allows,2019-12-12T16:51:46Z +issue,30780,Remove Ops bound in Python Layer for Legacy Reasons,https://github.com/pytorch/pytorch/issues/30780,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: pybind",scan when time allows,2019-12-12T15:59:02Z +issue,23655,TorchScript GPU Fuser Doesn't Handle In-Place Operations,https://github.com/pytorch/pytorch/issues/23655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, oncall: jit",scan when time allows,2019-12-12T15:22:29Z +issue,31129,[jit] Python objects as arguments are not mutated,https://github.com/pytorch/pytorch/issues/31129,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-11T21:47:51Z +issue,19581,[jit] Do aten::values dispatch at build time instead of runtime,https://github.com/pytorch/pytorch/issues/19581,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-11T21:03:37Z +issue,31048,MathJax too small in Firefox,https://github.com/pytorch/pytorch/issues/31048,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-12-11T17:14:30Z +issue,31085,Common lookup of generic types across full script and mobile parsers,https://github.com/pytorch/pytorch/issues/31085,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-12-11T17:06:26Z +issue,30718,Method to broadcast parameters/buffers of DDP model,https://github.com/pytorch/pytorch/issues/30718,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-12-11T09:08:12Z +issue,30962,How can I add masks to parameters,https://github.com/pytorch/pytorch/issues/30962,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-12-11T07:37:43Z +issue,25783,Detaching a distribution's `log_prob` to block gradients only w.r.t its parameters,https://github.com/pytorch/pytorch/issues/25783,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2019-12-11T03:24:29Z +issue,30965,JIT breaks with postponed annotations,https://github.com/pytorch/pytorch/issues/30965,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-10T22:10:12Z +issue,31007,"GPU version of minimal example for libtorch fails with ""no kernel image is available...""",https://github.com/pytorch/pytorch/issues/31007,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-12-10T08:09:47Z +issue,30744,Issues in linking libtorch c++,https://github.com/pytorch/pytorch/issues/30744,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-12-10T06:43:01Z +issue,19248,add stable distribution in torch.distributions,https://github.com/pytorch/pytorch/issues/19248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2019-12-10T00:08:19Z +issue,30953,false CHECK FAILED at ../aten/src/ATen/core/function_schema_inl.h,https://github.com/pytorch/pytorch/issues/30953,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-12-09T18:26:37Z +issue,30947,Wrong initialization with kaiming_uniform_,https://github.com/pytorch/pytorch/issues/30947,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-12-09T18:25:45Z +issue,30903,Don't ship protoc in wheels,https://github.com/pytorch/pytorch/issues/30903,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2019-12-09T18:11:20Z +issue,30899,Adding max_norm constraint to an Embedding layer leads to an error,https://github.com/pytorch/pytorch/issues/30899,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-12-09T18:09:43Z +issue,30762,Page for `torch.__config__` 404s,https://github.com/pytorch/pytorch/issues/30762,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-12-06T17:15:28Z +issue,28577,`Tensor.__reversed__` breaks protocol for reversible objects,https://github.com/pytorch/pytorch/issues/28577,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: pybind",scan when time allows,2019-12-05T04:00:56Z +issue,24037,Tests do not pass with the latest protobuf,https://github.com/pytorch/pytorch/issues/24037,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: protobuf,scan when time allows,2019-12-05T00:00:51Z +issue,23032,Proposal: Optional AutogradMeta for Variable,https://github.com/pytorch/pytorch/issues/23032,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: autograd,scan when time allows,2019-12-04T20:46:20Z +issue,30596,Overlapping strides not supported by cublas,https://github.com/pytorch/pytorch/issues/30596,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2019-12-04T15:26:24Z +issue,23355,[jit] add named tuple as output type to the tracer,https://github.com/pytorch/pytorch/issues/23355,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-04T00:59:17Z +issue,30589,weight norm missing p=,https://github.com/pytorch/pytorch/issues/30589,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-12-03T15:40:00Z +issue,30611,Refactor/consolidate code for generating test tensors,https://github.com/pytorch/pytorch/issues/30611,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, module: type promotion",scan when time allows,2019-12-03T15:09:27Z +issue,30633,Remove RPC internal helper that overrides the default pickler,https://github.com/pytorch/pytorch/issues/30633,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-12-03T08:37:18Z +issue,30489,Torchscript Precision Issue with PyTorch/vision pretrained model inception v3,https://github.com/pytorch/pytorch/issues/30489,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-03T02:48:16Z +issue,24834,PyTorch 1.2 'module' object has no attribute 'BFloat16StorageBase',https://github.com/pytorch/pytorch/issues/24834,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: undefined reference, module: vision",scan when time allows,2019-12-02T09:52:30Z +issue,18212,[JIT] bitwise NOT does not handle tensor shapes correctly under JIT,https://github.com/pytorch/pytorch/issues/18212,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-02T09:50:03Z +issue,18005,Support 'bytes' type in torchscript,https://github.com/pytorch/pytorch/issues/18005,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-02T09:41:50Z +issue,17970,Binary not operator causes crash when Jit module is executed on different device,https://github.com/pytorch/pytorch/issues/17970,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-12-02T09:38:59Z +issue,17797,Non-coherent result for C++ with multithreading and GPU,https://github.com/pytorch/pytorch/issues/17797,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, oncall: jit",scan when time allows,2019-12-02T09:33:29Z +issue,30533,Document memory characteristics of in-place ops,https://github.com/pytorch/pytorch/issues/30533,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: sparse, module: type promotion",scan when time allows,2019-11-27T21:50:11Z +issue,30535,[FR] nn.init.* accept None as input tensor,https://github.com/pytorch/pytorch/issues/30535,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-11-27T21:37:46Z +issue,1736,Expose optimizer options as attributes when there's a single param group,https://github.com/pytorch/pytorch/issues/1736,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, module: optimizer",scan when time allows,2019-11-27T18:47:03Z +issue,30505,Adding a function to change the process_group in DistributedDataParallel,https://github.com/pytorch/pytorch/issues/30505,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-11-27T16:11:46Z +issue,30469,Add single input/output tensor scatter/gather to ProcessGroup base class,https://github.com/pytorch/pytorch/issues/30469,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-11-27T15:38:53Z +issue,30463,cuda support,https://github.com/pytorch/pytorch/issues/30463,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-11-27T15:38:23Z +issue,30421,[jit] Module references aren't preserved,https://github.com/pytorch/pytorch/issues/30421,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-26T23:32:06Z +issue,30418,More detailed information about TensorType in error messages,https://github.com/pytorch/pytorch/issues/30418,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-26T21:08:56Z +issue,30440,Redundant counter in batchnorm impl,https://github.com/pytorch/pytorch/issues/30440,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-11-26T19:50:35Z +issue,29545,Tensorboard GPU Problems,https://github.com/pytorch/pytorch/issues/29545,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: tensorboard",scan when time allows,2019-11-26T13:22:51Z +issue,30461,`torch.multiprocessing.spawn` fails when `join=False`,https://github.com/pytorch/pytorch/issues/30461,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2019-11-26T13:12:19Z +issue,28275,nn.parallel.replicate in v1.1+ is much slower than v1.0,https://github.com/pytorch/pytorch/issues/28275,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: performance",scan when time allows,2019-11-26T12:39:55Z +issue,15291,linked error of Pytorch 1.0 release,https://github.com/pytorch/pytorch/issues/15291,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2019-11-26T12:25:37Z +issue,30413,nn.functional should maintain API parity with nn where possible,https://github.com/pytorch/pytorch/issues/30413,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-11-26T09:30:45Z +issue,30399,Conv2d: Inconsistent results on Raspberry Pi 3B,https://github.com/pytorch/pytorch/issues/30399,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2019-11-26T07:39:12Z +issue,30443,Expose bhistc to python,https://github.com/pytorch/pytorch/issues/30443,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-26T04:37:24Z +issue,30403,last updated timestamp,https://github.com/pytorch/pytorch/issues/30403,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-11-26T00:47:28Z +issue,30384,Docs missing link to torch.__config__,https://github.com/pytorch/pytorch/issues/30384,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-11-26T00:38:14Z +issue,10006,RNN gradients in eval mode in pytorch 0.4,https://github.com/pytorch/pytorch/issues/10006,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: rnn",scan when time allows,2019-11-25T20:56:13Z +issue,26119,Make GloballyUniqueId a common type for both rpc and dist autograd,https://github.com/pytorch/pytorch/issues/26119,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-25T19:04:52Z +issue,27671,Avoid RTTI in DistEngine,https://github.com/pytorch/pytorch/issues/27671,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-25T19:01:15Z +issue,22022,nn.modules.functional.h does not support optional arguments,https://github.com/pytorch/pytorch/issues/22022,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2019-11-25T16:59:20Z +issue,24273,Check PyTorch version when initializing process groups,https://github.com/pytorch/pytorch/issues/24273,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",scan when time allows,2019-11-25T09:10:38Z +issue,27843,Be able to build torch.distributed documentation easier,https://github.com/pytorch/pytorch/issues/27843,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, oncall: distributed",scan when time allows,2019-11-25T08:56:57Z +issue,30225,c10:Error: could not unlink the shared memory file,https://github.com/pytorch/pytorch/issues/30225,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing",scan when time allows,2019-11-22T20:39:12Z +issue,30150,I have reconstructed LSTM model and tested by mnist data but the loss is not changed (loss=2.3),https://github.com/pytorch/pytorch/issues/30150,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2019-11-22T09:19:53Z +issue,30182,[quantization][graph mode] SubgraphRewriter discards SourceRanges,https://github.com/pytorch/pytorch/issues/30182,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-22T07:36:25Z +issue,1591,Pad PackedSequences to original batch length,https://github.com/pytorch/pytorch/issues/1591,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nestedtensor,scan when time allows,2019-11-21T21:49:43Z +issue,30155,Transform caffe2 to trt failed,https://github.com/pytorch/pytorch/issues/30155,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-21T21:30:02Z +issue,30248,Add scripts for comprehensive benchmark TensorIterator,https://github.com/pytorch/pytorch/issues/30248,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2019-11-21T21:07:03Z +issue,30162,[docs] Newline / whitespace / comma missing in formula for PoissonNLLLoss,https://github.com/pytorch/pytorch/issues/30162,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-11-21T20:43:13Z +issue,29722,Slow (20-50x) RNN tutorial/example when torch is installed using pip comp. to conda installation,https://github.com/pytorch/pytorch/issues/29722,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: mkl, module: performance, module: rnn",scan when time allows,2019-11-21T12:41:53Z +issue,9208,[Caffe2] compiling error with gcc-6,https://github.com/pytorch/pytorch/issues/9208,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-21T07:28:13Z +issue,22911,JIT trace parameter sharing error if Module attributes happen to be the same,https://github.com/pytorch/pytorch/issues/22911,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-20T22:29:58Z +issue,30161,Support struct that is initializable / mutable in CPP,https://github.com/pytorch/pytorch/issues/30161,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-20T21:42:00Z +issue,27955,Expand Pytorch C10D backend to dynamic load third party communication library,https://github.com/pytorch/pytorch/issues/27955,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-11-20T10:44:50Z +issue,26759,[RFC] RRef Protocol,https://github.com/pytorch/pytorch/issues/26759,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-19T16:22:34Z +issue,26790,Importing tensorboard jams CUDA device selection,https://github.com/pytorch/pytorch/issues/26790,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2019-11-18T15:43:47Z +issue,29860,torch.save/load shows raw path on the pickle_module arg,https://github.com/pytorch/pytorch/issues/29860,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: serialization",scan when time allows,2019-11-15T00:16:23Z +issue,29800,unsqueeze has 'out=' option documented but not implemented(?),https://github.com/pytorch/pytorch/issues/29800,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-11-14T22:17:37Z +issue,29720,Provide a mechanism to set global state per test in thread-safe manner,https://github.com/pytorch/pytorch/issues/29720,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2019-11-14T19:12:42Z +issue,29717,CPU and CUDA error messages are divergent in type promotion,https://github.com/pytorch/pytorch/issues/29717,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, module: type promotion",scan when time allows,2019-11-14T18:33:31Z +issue,29739,"[build] gcc 7.4 needs CMAKE_CXX_FLAGS=""-std=gnu++11""",https://github.com/pytorch/pytorch/issues/29739,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: dependency bug",scan when time allows,2019-11-14T18:22:00Z +issue,29713,[FR] F.pad support syntax sugars for specifying the padding amount,https://github.com/pytorch/pytorch/issues/29713,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: convolution,scan when time allows,2019-11-14T18:14:02Z +issue,29754,[FR] general nll_loss and cross_entropy along arbitrary dimension,https://github.com/pytorch/pytorch/issues/29754,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2019-11-14T18:00:09Z +issue,11793,DataParallel: Parallel_apply assert len(modules) == len(inputs) AssertionError,https://github.com/pytorch/pytorch/issues/11793,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-11-14T08:55:22Z +issue,26344,How to get rid of zombie processes using torch.multiprocessing.Pool?,https://github.com/pytorch/pytorch/issues/26344,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dependency bug, module: multiprocessing, oncall: distributed",scan when time allows,2019-11-14T00:08:11Z +issue,8028,Build error : mpi/mpi_gpu_test.cc.o: undefined reference to symbol '_ZN3MPI8Datatype4FreeEv',https://github.com/pytorch/pytorch/issues/8028,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-13T09:27:26Z +issue,29692,[FR] add generator= kwarg support for torch.randn and torch.rand,https://github.com/pytorch/pytorch/issues/29692,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2019-11-13T08:23:31Z +issue,29662,[jit] Dict construction fails at runtime,https://github.com/pytorch/pytorch/issues/29662,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-12T22:16:59Z +issue,29235,Redo our library structure,https://github.com/pytorch/pytorch/issues/29235,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-11-12T22:07:10Z +issue,20151,RuntimeError: Given input size: (2048x1x1). Calculated output size: (2048x-5x-5). Output size is too small at /pytorch/aten/src/THNN/generic/SpatialAveragePooling.c:48,https://github.com/pytorch/pytorch/issues/20151,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: tensorboard",scan when time allows,2019-11-12T12:38:46Z +issue,20204,"SyncBatchNorm should support 2D input (B, C)",https://github.com/pytorch/pytorch/issues/20204,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, oncall: distributed",scan when time allows,2019-11-12T00:29:41Z +issue,29402,[RFC] RPC timeout,https://github.com/pytorch/pytorch/issues/29402,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-11T19:08:02Z +issue,29553,`print` uses lots of GPU memory,https://github.com/pytorch/pytorch/issues/29553,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: printing",scan when time allows,2019-11-11T18:44:20Z +issue,5737,[feature request] Callback on learning rate drop in torch.optim.lr_scheduler.ReduceLROnPlateau,https://github.com/pytorch/pytorch/issues/5737,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-11-11T03:47:00Z +issue,25297,torch.as_tensor(bytearray(...)) seems to leak memory,https://github.com/pytorch/pytorch/issues/25297,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dataloader, module: memory usage",scan when time allows,2019-11-10T22:08:36Z +issue,29429,NLLLoss reduce=True returning nan in float16,https://github.com/pytorch/pytorch/issues/29429,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-11-08T15:54:00Z +issue,29419,"IValue can't be constructed from one of `int`, `long`, or `long long`.",https://github.com/pytorch/pytorch/issues/29419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-08T14:19:48Z +issue,29377,[jit] Saving a `ScriptFunction` to a buffer doesn't work,https://github.com/pytorch/pytorch/issues/29377,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-07T19:09:56Z +issue,29172,Modules without copying in multiprocess,https://github.com/pytorch/pytorch/issues/29172,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2019-11-07T17:03:13Z +issue,29254,Add support for ivalue float scalars,https://github.com/pytorch/pytorch/issues/29254,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-07T16:59:23Z +issue,29369,nn.Conv(n)d constructor doesn't check for the number of kernel dimensions,https://github.com/pytorch/pytorch/issues/29369,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",scan when time allows,2019-11-07T16:37:44Z +issue,1487,[feature request/proposal] Relax scale_factor for nearest neighbor upsampling,https://github.com/pytorch/pytorch/issues/1487,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-07T16:32:45Z +issue,11978,[Caffe2] Attempting to install Caffe2 in Google Colab,https://github.com/pytorch/pytorch/issues/11978,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-07T13:31:48Z +issue,28745,Problem when installing Pytorch from source on CentOS 7.4,https://github.com/pytorch/pytorch/issues/28745,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-11-07T09:25:32Z +issue,27740,Scripting torchvision.models.detection.maskrcnn_resnet50_fpn,https://github.com/pytorch/pytorch/issues/27740,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2019-11-07T02:43:34Z +issue,29198,RuntimeError: CUDA out of memory with available GPU memory,https://github.com/pytorch/pytorch/issues/29198,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: memory usage",scan when time allows,2019-11-06T09:17:59Z +issue,28757,[feature request] Docs for fuse_conv_bn_eval,https://github.com/pytorch/pytorch/issues/28757,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-11-05T23:14:10Z +issue,29177,[FR] trace_module traces both eval and train graph,https://github.com/pytorch/pytorch/issues/29177,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-11-05T22:17:18Z +issue,24946,[RPC] Make ProcessGroupAgent send task non-blocking,https://github.com/pytorch/pytorch/issues/24946,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-05T17:43:04Z +issue,26446,Avoid sending zero grads over the wire in distributed autograd backward pass,https://github.com/pytorch/pytorch/issues/26446,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-05T17:40:15Z +issue,27642,Test re-entrant backward works with torch.distributed.autograd.backward(),https://github.com/pytorch/pytorch/issues/27642,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: rpc",scan when time allows,2019-11-05T17:38:55Z +issue,27641,torch.distributed.autograd.backward() should populate .grad field on Tensors by default.,https://github.com/pytorch/pytorch/issues/27641,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: rpc",scan when time allows,2019-11-05T17:38:55Z +issue,27579,Support Anomaly detection for distributed autograd.,https://github.com/pytorch/pytorch/issues/27579,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: rpc",scan when time allows,2019-11-05T17:38:54Z +issue,27099,Support implicit RRef type conversion,https://github.com/pytorch/pytorch/issues/27099,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-05T17:38:52Z +issue,28925,"Provide rpc, remote and dist autograd C++ APIs and register them as Prim::ops",https://github.com/pytorch/pytorch/issues/28925,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: cpp, module: rpc",scan when time allows,2019-11-05T17:37:07Z +issue,28882,Support RRef[T].__call__(*args) which invokes T.__call__(*args) on owner,https://github.com/pytorch/pytorch/issues/28882,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rpc,scan when time allows,2019-11-05T17:37:06Z +issue,29202,Add instructions for building torch.distributed on macOS,https://github.com/pytorch/pytorch/issues/29202,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, oncall: distributed",scan when time allows,2019-11-05T16:04:44Z +issue,29057,Docker issue for Pytorch 1.3,https://github.com/pytorch/pytorch/issues/29057,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build, module: docker",scan when time allows,2019-11-05T15:14:31Z +issue,14370,[Caffe2] Check failed: output->size() == values_.size() output size: 1 given size: 1563551,https://github.com/pytorch/pytorch/issues/14370,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-11-05T03:17:39Z +issue,28938,Cannot select version in the tutorials page,https://github.com/pytorch/pytorch/issues/28938,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-11-04T21:33:12Z +issue,28515,Get rid of libc10.so,https://github.com/pytorch/pytorch/issues/28515,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-11-04T18:18:17Z +issue,28791,[CUDA-MEMCHECK] TestTorchDeviceTypeCUDA.test_pin_memory_from_constructor_cuda fails,https://github.com/pytorch/pytorch/issues/28791,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-11-01T23:40:05Z +issue,29028,Inconsistent documentation for in-place functions at https://pytorch.org/docs/stable/torch.html,https://github.com/pytorch/pytorch/issues/29028,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-11-01T22:28:47Z +issue,1359,"Conjugate Gradient mentioned in docs, but not implemented",https://github.com/pytorch/pytorch/issues/1359,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-11-01T18:25:21Z +issue,29033,Move allgather_coalesced functionality to comm.cpp.,https://github.com/pytorch/pytorch/issues/29033,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-11-01T17:31:54Z +issue,29021,Named Tensors: size of a named dimension,https://github.com/pytorch/pytorch/issues/29021,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: named tensor",scan when time allows,2019-11-01T17:26:29Z +issue,12535,[CMake] Linking against Intel OpenMP,https://github.com/pytorch/pytorch/issues/12535,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn, module: openmp",scan when time allows,2019-10-31T17:56:15Z +issue,28962,[feature request] Some activations modules missing inplace argument,https://github.com/pytorch/pytorch/issues/28962,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-10-31T17:06:36Z +issue,28845,Implement operator<< for bfloat16,https://github.com/pytorch/pytorch/issues/28845,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bfloat16,scan when time allows,2019-10-31T00:53:20Z +issue,28859,Remove bce_with_logits in derivatives.yaml,https://github.com/pytorch/pytorch/issues/28859,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-10-31T00:36:21Z +issue,28930,Add support for multidimensional input to `at::tensor`,https://github.com/pytorch/pytorch/issues/28930,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-10-31T00:30:22Z +issue,16405,error on cmake_version from tools/build_pytorch_libs.py,https://github.com/pytorch/pytorch/issues/16405,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-10-29T19:03:56Z +issue,28780,'import torch` fails with Illegal instruction,https://github.com/pytorch/pytorch/issues/28780,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-10-28T18:20:35Z +issue,28756,cmake allows both MKL and MKLDNN to be OFF; aten/src/ATen/CMakeLists.txt then ignores c++ sources,https://github.com/pytorch/pytorch/issues/28756,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-10-28T17:40:39Z +issue,28721,[docs] Improve docs of nn.MultiheadAttention,https://github.com/pytorch/pytorch/issues/28721,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-10-28T17:28:47Z +issue,28417,Patch: Fix for using `clang` to compile CUDA,https://github.com/pytorch/pytorch/issues/28417,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2019-10-28T17:10:17Z +issue,20864,Import warning when using the wrong version of CUDA,https://github.com/pytorch/pytorch/issues/20864,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-10-27T08:04:45Z +issue,28552,Doing rendezvous twice can cause hangs,https://github.com/pytorch/pytorch/issues/28552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-10-25T16:42:41Z +issue,28655,Cmake fails due to bad python call,https://github.com/pytorch/pytorch/issues/28655,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-10-25T15:46:21Z +issue,11451,[caffe2] caffe2 openmp linking error with xcode 9.0(AppleClang 9.0) on mac,https://github.com/pytorch/pytorch/issues/11451,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-10-25T11:29:25Z +issue,26409,Generated file not getting cleaned up by clean,https://github.com/pytorch/pytorch/issues/26409,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-10-24T21:56:32Z +issue,15070,In-place operations on `.data` or `.detach()` of sparse tensor doesn't update the original tensor,https://github.com/pytorch/pytorch/issues/15070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2019-10-24T19:23:01Z +issue,28558,We should suggest using as_strided_ instead of set_,https://github.com/pytorch/pytorch/issues/28558,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-10-24T15:30:02Z +issue,28441,gradient of Dirichlet.log_prob gives nan,https://github.com/pytorch/pytorch/issues/28441,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2019-10-24T15:23:44Z +issue,28532,TestTorch.test_c10_layer_norm fails if you run it on a build of PyTorch with BUILD_CAFFE2_OPS=0,https://github.com/pytorch/pytorch/issues/28532,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dispatch,scan when time allows,2019-10-24T09:16:53Z +issue,14277,Provide Protobuf library if libtorch was built with included version,https://github.com/pytorch/pytorch/issues/14277,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: protobuf",scan when time allows,2019-10-23T22:13:22Z +issue,21061,Misleading Error when doing Large Batch Matrix Multiplication,https://github.com/pytorch/pytorch/issues/21061,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-10-23T21:40:07Z +issue,28390,torch.multinominal ignores elements from cumulative distribution,https://github.com/pytorch/pytorch/issues/28390,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: random",scan when time allows,2019-10-23T12:40:10Z +issue,26990,torch.quantized.modules.floatFunctional -- add a little color commentary to the doc,https://github.com/pytorch/pytorch/issues/26990,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-10-22T22:46:55Z +issue,16710,culibos linker errors on binary_linux_conda_3.6_cu90_build,https://github.com/pytorch/pytorch/issues/16710,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-10-22T07:19:01Z +issue,27687,`nn.Sequential.__setattr__` appends to the execution list,https://github.com/pytorch/pytorch/issues/27687,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-10-21T18:35:49Z +issue,28318,`torch.nn.Module._load_state_dict` catch-all error message can be misleading,https://github.com/pytorch/pytorch/issues/28318,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-10-19T17:52:30Z +issue,23299,Add gatherv/allgatherv primitives to support non-equal contribution,https://github.com/pytorch/pytorch/issues/23299,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",scan when time allows,2019-10-18T20:48:17Z +issue,26379,DistAutogradContext should be cleaned up in case of node failures.,https://github.com/pytorch/pytorch/issues/26379,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: memory usage, module: rpc",scan when time allows,2019-10-18T20:43:03Z +issue,28249,Add scopes to autograd profiler,https://github.com/pytorch/pytorch/issues/28249,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-10-17T23:21:29Z +issue,28142,Named Tensor: Support -1 in `unflatten`.,https://github.com/pytorch/pytorch/issues/28142,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2019-10-17T23:14:29Z +issue,28110,Stable docs show dead __config__ section link,https://github.com/pytorch/pytorch/issues/28110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-10-17T22:14:11Z +issue,24354,Default warning handler in C++ doesn't seem to unique warnings,https://github.com/pytorch/pytorch/issues/24354,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: error checking,scan when time allows,2019-10-17T19:44:50Z +issue,28146,Named tensor: align_to align_as error messages,https://github.com/pytorch/pytorch/issues/28146,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2019-10-17T19:20:16Z +issue,28221,TorchScript doesn't support torch.channels_last or any other memory format constants,https://github.com/pytorch/pytorch/issues/28221,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: memory format, oncall: jit",scan when time allows,2019-10-17T15:05:26Z +issue,28220,TorchScript custom ops like `cuda` `byte` etc. doesn't support memory_format argument,https://github.com/pytorch/pytorch/issues/28220,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-17T15:04:28Z +issue,9699,[Build error] libcudnn.so: error adding symbols: File in wrong format,https://github.com/pytorch/pytorch/issues/9699,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-10-17T07:19:16Z +issue,20315,libtorch+opencv Mat result error: different from the python ones,https://github.com/pytorch/pytorch/issues/20315,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-10-16T07:05:02Z +issue,27708,Documentation makefile should include torchvision,https://github.com/pytorch/pytorch/issues/27708,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2019-10-15T23:56:51Z +issue,27926,`num_batches_tracked` update in `_BatchNorm` forward should be a single scalar update on host regardless of the residence of the layer,https://github.com/pytorch/pytorch/issues/27926,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn, module: performance",scan when time allows,2019-10-15T18:44:05Z +issue,21738,symbol lookup error: libmkl_intel_lp64.so: undefined symbol: mkl_blas_dsyrk (binaries built with static linking -DBUILD_SHARED_LIBS=OFF fail due to dynamic linker problem),https://github.com/pytorch/pytorch/issues/21738,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: static linking, module: third_party",scan when time allows,2019-10-15T14:31:07Z +issue,17796,[Caffe2] install error,https://github.com/pytorch/pytorch/issues/17796,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-10-15T08:45:12Z +issue,25314,ConstQuantizerPtr is misnamed,https://github.com/pytorch/pytorch/issues/25314,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: quantization,scan when time allows,2019-10-14T23:10:18Z +issue,25821,Conv2D 2x~20x slower than Tensorflow when channel count is small,https://github.com/pytorch/pytorch/issues/25821,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cuda, module: cudnn, module: performance",scan when time allows,2019-10-11T16:48:19Z +issue,27175,Easier way to create tensors with names,https://github.com/pytorch/pytorch/issues/27175,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: named tensor,scan when time allows,2019-10-11T13:04:44Z +issue,27015,Missing bin and include when building with torchvision on CentOS,https://github.com/pytorch/pytorch/issues/27015,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp, module: vision",scan when time allows,2019-10-10T20:49:37Z +issue,27608,Unify warning logging mechanism,https://github.com/pytorch/pytorch/issues/27608,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-10-10T00:10:16Z +issue,27540,CTCLoss cuda backend large batch handling takes up to 1.8x more memory,https://github.com/pytorch/pytorch/issues/27540,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: loss, module: memory usage",scan when time allows,2019-10-08T15:05:13Z +issue,26534,MKLDNN+AMD BLIS path for PyTorch,https://github.com/pytorch/pytorch/issues/26534,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: mkldnn",scan when time allows,2019-10-08T14:12:43Z +issue,25478,Delete TensorOptions::operator==,https://github.com/pytorch/pytorch/issues/25478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-10-08T10:14:24Z +issue,18850,Allow tracing of models which output `None`,https://github.com/pytorch/pytorch/issues/18850,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-07T22:51:42Z +issue,25892,load_state_dict on CPU first,https://github.com/pytorch/pytorch/issues/25892,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2019-10-07T20:33:57Z +issue,15864,Deformable Convolution,https://github.com/pytorch/pytorch/issues/15864,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2019-10-05T16:17:11Z +issue,27297,CUDA error: device-side assert triggered(insert_events at /pytorch/c10/cuda/CUDACachingAllocator.cpp:569),https://github.com/pytorch/pytorch/issues/27297,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-10-04T17:26:27Z +issue,27218,pytorch data_parallel oom on gpu:0,https://github.com/pytorch/pytorch/issues/27218,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: data parallel, module: memory usage",scan when time allows,2019-10-04T08:10:56Z +issue,26730,[jit] Bad error when instantiating TorchScript class with incorrect types,https://github.com/pytorch/pytorch/issues/26730,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-03T21:27:12Z +issue,25410,[jit] NamedTuples don't respect `__new__`,https://github.com/pytorch/pytorch/issues/25410,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-03T21:24:47Z +issue,25329,[jit] Bad error for incorrect container type,https://github.com/pytorch/pytorch/issues/25329,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-03T21:24:47Z +issue,24786,[jit] Dict iterator invalidation doesn't match Python,https://github.com/pytorch/pytorch/issues/24786,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-03T21:24:45Z +issue,21544,"[jit] In pickler, don't memoize if not necessary",https://github.com/pytorch/pytorch/issues/21544,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-03T21:16:31Z +issue,21442,[JIT] kwarg with default doesn't work for class instantiation,https://github.com/pytorch/pytorch/issues/21442,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-03T21:16:31Z +issue,20119,[JIT] Source highlighting doesn't line up when tabs are used for indentation,https://github.com/pytorch/pytorch/issues/20119,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-10-03T21:15:42Z +issue,27138,CUDAPytorchToCaffe2.MutualResizes is flaky,https://github.com/pytorch/pytorch/issues/27138,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: flaky-tests, module: tests",scan when time allows,2019-10-03T17:22:41Z +issue,27046,The Gather problem in DataParallel: dimension are not matched.,https://github.com/pytorch/pytorch/issues/27046,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2019-10-02T08:41:13Z +issue,24234,torch.utils.tensorboard.SummaryWriter fails to flush at program exit,https://github.com/pytorch/pytorch/issues/24234,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2019-10-01T21:25:21Z +issue,24102,[JIT] script doesn't convert dtypes back to torch.dtype from long,https://github.com/pytorch/pytorch/issues/24102,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-09-30T21:23:42Z +issue,27055,torch::NoGradGuard no_grad get wrong when I use batchsize!=1,https://github.com/pytorch/pytorch/issues/27055,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-09-30T18:18:29Z +issue,27021,Cmake warnings during build,https://github.com/pytorch/pytorch/issues/27021,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2019-09-30T18:15:04Z +issue,27034,RuntimeError:[enforce fail at context.h:48] option.device_type() ==PROTO_CPU. 1vs0,https://github.com/pytorch/pytorch/issues/27034,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-09-30T18:11:38Z +issue,27024,Is it an incompleted dst tensor synchronization in CUDA device to device copy ?,https://github.com/pytorch/pytorch/issues/27024,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-09-30T18:10:29Z +issue,26950,conv2d Memory usage is too large； pytorch 1.1.0,https://github.com/pytorch/pytorch/issues/26950,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn, module: dependency bug, module: memory usage",scan when time allows,2019-09-30T17:11:46Z +issue,25691,[dataloader] Hang because of too many open files (and probably some process dead),https://github.com/pytorch/pytorch/issues/25691,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-09-29T23:03:31Z +issue,26967,Lint rule to test for creation of tensor in native/ without options(),https://github.com/pytorch/pytorch/issues/26967,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: lint",scan when time allows,2019-09-27T14:58:51Z +issue,26964,AnyValueTest.CorrectlyAccessesIntWhenCorrectType UBSAN failure: owncast of address 0x60300105d750 which does not point to an object of type 'Holder' Sep 27 00:01:03 0x60300105d750: note: object is of type 'torch::nn::AnyModule::Value::Holder',https://github.com/pytorch/pytorch/issues/26964,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2019-09-27T13:43:47Z +issue,26863,[BUG Report]Integrate libtorch to ffmpeg but memory leak happened!,https://github.com/pytorch/pytorch/issues/26863,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: vision",scan when time allows,2019-09-26T21:24:14Z +issue,9029,DataWorkersTest::testRNNInput timeout,https://github.com/pytorch/pytorch/issues/9029,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-09-26T14:31:20Z +issue,26796,setup.py install error,https://github.com/pytorch/pytorch/issues/26796,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-09-25T22:12:27Z +issue,14487,TestAdagrad.test_row_wise_sparse_adagrad intermittently fails health check,https://github.com/pytorch/pytorch/issues/14487,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-09-25T19:47:09Z +issue,26659,torch::nn::Sequential not compatible with torch::nn::RNN,https://github.com/pytorch/pytorch/issues/26659,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2019-09-25T13:47:24Z +issue,15149,Use std::variant to represent C++ side enumerations (with binding support),https://github.com/pytorch/pytorch/issues/15149,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-09-24T17:00:54Z +issue,26396,NNPACK condition should be changed (ARM processors),https://github.com/pytorch/pytorch/issues/26396,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nnpack",scan when time allows,2019-09-24T14:06:11Z +issue,26567,Make `torch.save` serialize a zip file,https://github.com/pytorch/pytorch/issues/26567,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2019-09-24T14:05:53Z +issue,26258,"""git describe"" shows incorrect version 1.0 instead of 1.2",https://github.com/pytorch/pytorch/issues/26258,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-09-23T17:05:46Z +issue,22682,CMAKE_PARSE_IMPLICIT_LINK_INFO Function invoked with incorrect arguments,https://github.com/pytorch/pytorch/issues/22682,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: flaky-tests",scan when time allows,2019-09-23T13:54:25Z +issue,12869,caffe2: Unsupported type of tensor: nullptr (uninitialized),https://github.com/pytorch/pytorch/issues/12869,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-09-23T12:56:25Z +issue,12983,Fail to Run Caffe2 with Successful Build: This caffe2 python run does not have GPU support,https://github.com/pytorch/pytorch/issues/12983,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-09-22T21:02:05Z +issue,17458,Overflow in fbgemm,https://github.com/pytorch/pytorch/issues/17458,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-09-22T07:41:46Z +issue,21745,torch.save also saves docstrings into pickle for some reason,https://github.com/pytorch/pytorch/issues/21745,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2019-09-20T21:22:54Z +issue,26551,No way to disable mse_loss broadcasting warning,https://github.com/pytorch/pytorch/issues/26551,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-09-20T18:48:05Z +issue,24478,Port CPU_tensor_apply functions to TensorIterator (umbrella issue),https://github.com/pytorch/pytorch/issues/24478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu_tensor_apply,scan when time allows,2019-09-20T18:09:28Z +issue,12117,"Error: Internal Compiler Error (codegen): ""there was an error in verifying the lgenfe output!""",https://github.com/pytorch/pytorch/issues/12117,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2019-09-17T19:33:52Z +issue,26228,Higher dimension support for `MultiLableSoftMarginLoss`,https://github.com/pytorch/pytorch/issues/26228,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2019-09-17T19:31:29Z +issue,26338,Behavior of F.dropout in eval mode,https://github.com/pytorch/pytorch/issues/26338,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-09-17T19:05:52Z +issue,26110,Access data_ptr in RNN.cpp,https://github.com/pytorch/pytorch/issues/26110,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: rnn,scan when time allows,2019-09-17T15:39:48Z +issue,24016,Multiplying a very large CUDA tensor with another tensor yields unexpected result,https://github.com/pytorch/pytorch/issues/24016,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: dependency bug",scan when time allows,2019-09-17T14:02:43Z +issue,17354,testConvnetBenchmarks intermittently segfaults,https://github.com/pytorch/pytorch/issues/17354,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests,scan when time allows,2019-09-13T15:21:26Z +issue,26136,There should be gating around BFloat16,https://github.com/pytorch/pytorch/issues/26136,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bfloat16, module: performance",scan when time allows,2019-09-13T15:20:38Z +issue,25991,Why doc building isn't failing us for referring to a non-existent method?,https://github.com/pytorch/pytorch/issues/25991,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-09-13T10:54:45Z +issue,26058,sccache stats can cause whole build to fail,https://github.com/pytorch/pytorch/issues/26058,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-09-13T10:48:23Z +issue,25548,The inference speed of the torch c++ dynamic library compiled manually is slower than the torch library officially provided,https://github.com/pytorch/pytorch/issues/25548,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: cuda, module: performance",scan when time allows,2019-09-12T16:55:35Z +issue,25880,Incorrect lable read with ImageInput Op of Caffe2,https://github.com/pytorch/pytorch/issues/25880,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-09-11T01:53:36Z +issue,25752,torch.cuda.empty_cache() write data to gpu0,https://github.com/pytorch/pytorch/issues/25752,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-09-10T21:39:08Z +issue,23657,Accelerate PyTorch just-in-time compilation using MKL-DNN,https://github.com/pytorch/pytorch/issues/23657,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: mkldnn, oncall: jit",scan when time allows,2019-09-10T19:19:14Z +issue,25820,Pytorch master can not build with computer capability 3.0 under Mac OS X 10.13.16 with Nvidia GT 750m,https://github.com/pytorch/pytorch/issues/25820,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: macos",scan when time allows,2019-09-10T01:52:29Z +issue,25743,Custom sampler for Seq2Seq models to avoid padding,https://github.com/pytorch/pytorch/issues/25743,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-09-09T21:20:11Z +issue,25747,Int32 overflow in bincount indexing,https://github.com/pytorch/pytorch/issues/25747,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cuda",scan when time allows,2019-09-09T21:09:06Z +issue,25417,Add a mode to check input tensor sizes in allreduce_coalesced,https://github.com/pytorch/pytorch/issues/25417,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-09-09T09:50:48Z +issue,14318,No code example for AdaptiveLogSoftmaxWithLoss,https://github.com/pytorch/pytorch/issues/14318,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-09-06T18:18:41Z +issue,25460,Build link not right,https://github.com/pytorch/pytorch/issues/25460,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2019-09-06T05:34:05Z +issue,25648,"Torch.jit.trace unexpected error with `torch.cat(…, dim=-1)`",https://github.com/pytorch/pytorch/issues/25648,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-09-05T21:07:46Z +issue,19172,[Feature Request] Common constants in the torch.* namespace,https://github.com/pytorch/pytorch/issues/19172,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: numpy,scan when time allows,2019-09-05T18:50:03Z +issue,25635,[C++] Support negative index in `torch::TensorAccessor::size()`,https://github.com/pytorch/pytorch/issues/25635,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-09-05T11:47:10Z +issue,25514,Multithreaded backpropagation with custom autograd.Functions,https://github.com/pytorch/pytorch/issues/25514,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: data parallel,scan when time allows,2019-09-04T22:02:56Z +issue,25552,Usage of DDP on a module that doesn't require gradients,https://github.com/pytorch/pytorch/issues/25552,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-09-04T20:19:20Z +issue,21700,(LLD 8.0.0) ld: error: can't create dynamic relocation R_X86_64_DTPOFF32 against symbol: ideep::utils::computation_cache,https://github.com/pytorch/pytorch/issues/21700,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: mkldnn, module: static linking",scan when time allows,2019-09-04T01:02:53Z +issue,25601,Avoid non-POD data in thread_local,https://github.com/pytorch/pytorch/issues/25601,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: multithreading, module: performance",scan when time allows,2019-09-04T00:27:23Z +issue,25520,pytorch c++ api cannot call operator() on torch::nn::Sequential,https://github.com/pytorch/pytorch/issues/25520,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: nn",scan when time allows,2019-09-03T18:19:17Z +issue,25522,[dataloader] Problem in exception reraise mechanism,https://github.com/pytorch/pytorch/issues/25522,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-09-03T15:41:02Z +issue,24831,subprocess.CalledProcessError: Compile source in NVIDIA TX2,https://github.com/pytorch/pytorch/issues/24831,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cuda",scan when time allows,2019-09-03T13:23:23Z +issue,16268,Complete dtype support for torch.norm,https://github.com/pytorch/pytorch/issues/16268,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-08-30T21:51:38Z +issue,12795,C++ frontend: how to debug nan gradients,https://github.com/pytorch/pytorch/issues/12795,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-08-30T21:12:31Z +issue,25480,Vectorize bool operations,https://github.com/pytorch/pytorch/issues/25480,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: boolean tensor,scan when time allows,2019-08-30T20:57:03Z +issue,25381,regarding builtin_function_or_method,https://github.com/pytorch/pytorch/issues/25381,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-08-29T14:50:59Z +issue,23940,Improve the performance of linear algebra operations in CUDA for small problem sizes,https://github.com/pytorch/pytorch/issues/23940,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: linear algebra, module: performance",scan when time allows,2019-08-28T22:02:29Z +issue,6402,[caffe2] How to freeze a layer?,https://github.com/pytorch/pytorch/issues/6402,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-08-28T21:08:12Z +issue,9701,[request] speed-up multidim slicing backward,https://github.com/pytorch/pytorch/issues/9701,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: performance",scan when time allows,2019-08-28T15:46:29Z +issue,24090,`suggest_memory_format` has ambiguity & cannot represent intended layout format for corner cases,https://github.com/pytorch/pytorch/issues/24090,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-08-28T09:03:18Z +issue,15994,More data type support for gather_map,https://github.com/pytorch/pytorch/issues/15994,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-08-27T16:10:25Z +issue,17041,Errors running distributed example,https://github.com/pytorch/pytorch/issues/17041,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-08-26T02:32:44Z +issue,13224,Caffe2: Causes error when using flag remove_legacy_pad while converting from caffe to caffe2,https://github.com/pytorch/pytorch/issues/13224,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-08-24T08:51:22Z +issue,9310,/usr/bin/ld: cannot find -lpthreads,https://github.com/pytorch/pytorch/issues/9310,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-08-23T20:29:26Z +issue,25057,ctc_loss computes different losses and gradients on batched utterances vs. individual utterances,https://github.com/pytorch/pytorch/issues/25057,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: loss,scan when time allows,2019-08-23T19:40:04Z +issue,25070,[FR] torch.(Generator|random).seed allows specifying the seed value,https://github.com/pytorch/pytorch/issues/25070,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2019-08-23T19:33:25Z +issue,25071,Generator objects should not always use the same seed,https://github.com/pytorch/pytorch/issues/25071,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2019-08-23T19:32:43Z +issue,25112,[RFC] NestedTensor - 0.0.2,https://github.com/pytorch/pytorch/issues/25112,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-08-23T19:12:14Z +issue,24985,Data worker should fetch a sample instead of a batch.,https://github.com/pytorch/pytorch/issues/24985,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-08-23T19:04:26Z +issue,25034,"make add_module accept tuples as well or change containers(ModuleList, Sequential, etc) to allow this",https://github.com/pytorch/pytorch/issues/25034,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-08-23T18:36:16Z +issue,23720,segmentation faults when using multiprocessing_context='spawn' with large number of processes,https://github.com/pytorch/pytorch/issues/23720,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: multiprocessing, oncall: distributed",scan when time allows,2019-08-23T04:03:40Z +issue,25014,Benchmark cuDNN affine_grid_generator vs native,https://github.com/pytorch/pytorch/issues/25014,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: nn, module: performance",scan when time allows,2019-08-23T03:55:30Z +issue,18805,Running custom operator tests manually is too difficult,https://github.com/pytorch/pytorch/issues/18805,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp-extensions, module: tests",scan when time allows,2019-08-22T20:30:04Z +issue,24956,[RPC] Fix logging initialization warning in ProcessGroupAgent,https://github.com/pytorch/pytorch/issues/24956,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-08-21T15:28:55Z +issue,22128,Automatic rank selection when using file:// initialization method,https://github.com/pytorch/pytorch/issues/22128,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-08-21T06:20:18Z +issue,24836,Gloo scatter gives wrong result for stride != 1,https://github.com/pytorch/pytorch/issues/24836,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: bootcamp, oncall: distributed",scan when time allows,2019-08-20T15:20:11Z +issue,21318,"Undefined symbols for architecture x86_64: ""testing::internal::UntypedFunctionMockerBase::UntypedInvokeWith(void const*)"" on Mac OS X",https://github.com/pytorch/pytorch/issues/21318,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: internals, module: macos",scan when time allows,2019-08-20T15:18:31Z +issue,24353,Visual Studio Code not providing autosuggestions for submodules,https://github.com/pytorch/pytorch/issues/24353,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-08-20T15:13:12Z +issue,24336,Auto tuner takes too much time in serialized model,https://github.com/pytorch/pytorch/issues/24336,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-08-19T18:29:01Z +issue,24419,Building Python bits separate from C++ bits and making one play well with the other,https://github.com/pytorch/pytorch/issues/24419,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-08-19T16:24:39Z +issue,24798,Installs empty directories under Python's sitelibdir,https://github.com/pytorch/pytorch/issues/24798,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-08-19T16:23:34Z +issue,24500,Failed to compile PyTorch on IBM Power 9 architecture with CUDA 10,https://github.com/pytorch/pytorch/issues/24500,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-08-19T16:21:41Z +issue,24456,Build PyTorch 1.2.0 occur `recipe for target bin/test_parallel' failed,https://github.com/pytorch/pytorch/issues/24456,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-08-19T16:11:41Z +issue,24432,Use a ScriptModule on GPU that was saved from CPU,https://github.com/pytorch/pytorch/issues/24432,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: serialization, oncall: jit",scan when time allows,2019-08-19T16:10:42Z +issue,24343,Unified representation for enum types,https://github.com/pytorch/pytorch/issues/24343,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-08-19T16:10:07Z +issue,24417,1.0rc0-6216 installs empty directories under include and duplicates under /,https://github.com/pytorch/pytorch/issues/24417,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-08-19T16:08:36Z +issue,24498,Migrate CPU_tensor_apply to TensorIterator in aten/src/ATen/native/TensorCompare.cpp:30,https://github.com/pytorch/pytorch/issues/24498,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu_tensor_apply,scan when time allows,2019-08-16T17:27:42Z +issue,24167,Failed to build pytorch with NanoPi M4,https://github.com/pytorch/pytorch/issues/24167,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-08-16T15:56:18Z +issue,24079,[Caffe2] build android in v1.1.0 with headfile error,https://github.com/pytorch/pytorch/issues/24079,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-08-14T22:46:23Z +issue,24264,"TensorIterator ""builder"" options should be documented.",https://github.com/pytorch/pytorch/issues/24264,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-08-14T18:31:16Z +issue,8098,Please provide wheel package for windows on PyPI,https://github.com/pytorch/pytorch/issues/8098,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: binaries,scan when time allows,2019-08-13T23:00:11Z +issue,24193,TensorIterator stubs are designed for merge conflicts.,https://github.com/pytorch/pytorch/issues/24193,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-08-13T08:41:39Z +issue,23749,Wrong device in graph - Tensorboard SummaryWriter,https://github.com/pytorch/pytorch/issues/23749,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2019-08-12T19:02:50Z +issue,23771,SummaryWriter doesn't read comment if log_dir precised,https://github.com/pytorch/pytorch/issues/23771,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tensorboard,scan when time allows,2019-08-12T17:58:21Z +issue,23787,Better version of chrome://tracing,https://github.com/pytorch/pytorch/issues/23787,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-08-12T17:20:04Z +issue,23768,Better documentation about PyTorch's dependencies,https://github.com/pytorch/pytorch/issues/23768,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: docs, module: third_party",scan when time allows,2019-08-12T17:08:54Z +issue,5857,Change THCudaCheck to suggest that device-side asserts likely mean that you have out of bound indices,https://github.com/pytorch/pytorch/issues/5857,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: assert failure, module: cuda, module: error checking",scan when time allows,2019-08-12T13:54:34Z +issue,23732,nn.Module.forward signature with **kwargs,https://github.com/pytorch/pytorch/issues/23732,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: activation checkpointing, module: nn",scan when time allows,2019-08-12T12:49:08Z +issue,23865,model->to(device) costs over a millisecond when doing nothing,https://github.com/pytorch/pytorch/issues/23865,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2019-08-07T06:56:53Z +issue,23896,Remove USE_C10D flag,https://github.com/pytorch/pytorch/issues/23896,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, oncall: distributed",scan when time allows,2019-08-06T20:48:04Z +issue,21554,Pytorch hangs when dataloader multiprocessing workers are killed,https://github.com/pytorch/pytorch/issues/21554,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-08-06T01:04:00Z +issue,23721,Using PyTorch on AWS EFA network,https://github.com/pytorch/pytorch/issues/23721,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dependency bug, oncall: distributed",scan when time allows,2019-08-05T17:13:48Z +issue,23490,upcoming PEP 554: how much effort we need to support sub-interpreter,https://github.com/pytorch/pytorch/issues/23490,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-08-02T16:16:00Z +issue,19682,View in Sequential - reasoned case for it,https://github.com/pytorch/pytorch/issues/19682,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-08-01T21:51:57Z +issue,23676,"Bogus ""Your compiler (clang++) is not compatible"" message",https://github.com/pytorch/pytorch/issues/23676,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: macos",scan when time allows,2019-08-01T19:59:21Z +issue,23072,Inconsistent axis argument names in torch.diagonal and torch.transpose,https://github.com/pytorch/pytorch/issues/23072,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-08-01T02:55:31Z +issue,23540,Error from PyTorch when finalizing Python embedded in C++,https://github.com/pytorch/pytorch/issues/23540,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: pybind,scan when time allows,2019-07-31T17:42:49Z +issue,21454,[JIT] Memory Leak during tracing?,https://github.com/pytorch/pytorch/issues/21454,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-07-30T15:36:37Z +issue,23512,Build reconfiguration should consistently honor env variables,https://github.com/pytorch/pytorch/issues/23512,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-07-29T18:51:05Z +issue,23094,Versioning for libtorch nightlies,https://github.com/pytorch/pytorch/issues/23094,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2019-07-29T18:25:01Z +issue,11633,DataLoader: Could not wrapper a exception in threads,https://github.com/pytorch/pytorch/issues/11633,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: dataloader, module: error checking",scan when time allows,2019-07-29T17:19:48Z +issue,6135,Test suite should test implementations,https://github.com/pytorch/pytorch/issues/6135,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2019-07-29T15:30:36Z +issue,18849,More efficient STFT on CUDA,https://github.com/pytorch/pytorch/issues/18849,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: performance",scan when time allows,2019-07-29T04:20:32Z +issue,12855,Model with Caffe2 runs much slower than it with pytorch in GPU mode !!!!,https://github.com/pytorch/pytorch/issues/12855,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-29T00:35:30Z +issue,23429,Unreachable code in tanh,https://github.com/pytorch/pytorch/issues/23429,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-26T22:32:39Z +issue,21015,How to use Infiniband for cpu-cluster with backend gloo?,https://github.com/pytorch/pytorch/issues/21015,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-07-26T02:26:22Z +issue,23364,dyndep function not working,https://github.com/pytorch/pytorch/issues/23364,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-25T04:20:47Z +issue,23301,Error while using Libtorch + OpenCV + Qt Creator,https://github.com/pytorch/pytorch/issues/23301,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-07-24T21:43:25Z +issue,23240,UBSAN failure in test_simple_model (__main__.TestTensorBoardNumpy): runtime error: call to function pybind11::class_::dealloc(pybind11::detail::value_and_holder&) through pointer to incorrect function type 'void (*)(pybind11::detail::value_and_holder &)',https://github.com/pytorch/pytorch/issues/23240,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2019-07-23T18:44:59Z +issue,23103,Test utility for non-contiguous tensors,https://github.com/pytorch/pytorch/issues/23103,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: tests,scan when time allows,2019-07-23T15:38:33Z +issue,23230,Consolidate multiprocessing helpers in distributed tests,https://github.com/pytorch/pytorch/issues/23230,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-07-23T15:05:54Z +issue,23079,update docs that sorting is not needed in,https://github.com/pytorch/pytorch/issues/23079,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rnn",scan when time allows,2019-07-22T17:36:36Z +issue,23068,[c++] torch::conv2d() expected output_padding to be a single integer value or a list of 3 values,https://github.com/pytorch/pytorch/issues/23068,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cpp, module: docs, module: nn",scan when time allows,2019-07-19T19:48:21Z +issue,23054,ConcatDataset returns different error messages setting out of range plus index and minus index.,https://github.com/pytorch/pytorch/issues/23054,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-07-19T18:53:16Z +issue,6794,[Caffe2] TensorProtosDBInput AttributeError,https://github.com/pytorch/pytorch/issues/6794,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-19T00:36:08Z +issue,23026,BatchNorm1d fails on first run through GPU,https://github.com/pytorch/pytorch/issues/23026,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: nn",scan when time allows,2019-07-18T21:21:56Z +issue,21818,C++ ABI - Coupling different libraries issue,https://github.com/pytorch/pytorch/issues/21818,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-07-18T06:14:15Z +issue,22687,DispatchStub should report what operator it failed to find kernel for,https://github.com/pytorch/pytorch/issues/22687,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: internals",scan when time allows,2019-07-17T15:27:09Z +issue,22868,Make it easier to bisect on PyTorch,https://github.com/pytorch/pytorch/issues/22868,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: ci",scan when time allows,2019-07-17T11:53:05Z +issue,22859,CRITICAL:root:Cannot load caffe2.python. Error: DLL load failed: The specified module could not be found.,https://github.com/pytorch/pytorch/issues/22859,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-17T06:22:46Z +issue,22879,Difference between dropout2d and dropout3d,https://github.com/pytorch/pytorch/issues/22879,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-07-17T01:04:31Z +issue,22785,Getting cuda runtime error (48) with Jetson TX2 when running simple program,https://github.com/pytorch/pytorch/issues/22785,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-07-13T00:22:57Z +issue,22766,assert_no_internal_overlap should pass const char*,https://github.com/pytorch/pytorch/issues/22766,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-07-13T00:17:56Z +issue,14478,Flaky download from files.pythonhosted.org when installing botocore,https://github.com/pytorch/pytorch/issues/14478,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests,scan when time allows,2019-07-12T19:10:05Z +issue,20488,RoiAlignTest.CheckCPUGPUEqual is still flaky,https://github.com/pytorch/pytorch/issues/20488,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests,scan when time allows,2019-07-12T18:21:29Z +issue,22495,`binary_linux_libtorch_2.7m_cu100_devtoolset3_build` times out after running for 5 hours,https://github.com/pytorch/pytorch/issues/22495,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-07-11T22:17:33Z +issue,22604,"""CrossEntropyLoss"" should mention in its name that it takes softmax for target",https://github.com/pytorch/pytorch/issues/22604,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: loss, module: nn",scan when time allows,2019-07-11T00:12:56Z +issue,22414,CPU random number generator is slow,https://github.com/pytorch/pytorch/issues/22414,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: performance, module: random",scan when time allows,2019-07-10T19:34:05Z +issue,22669,BFloat16 numeric limits should contain more info,https://github.com/pytorch/pytorch/issues/22669,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: bfloat16,scan when time allows,2019-07-10T16:29:29Z +issue,21780,cumsum cuda numerical instability,https://github.com/pytorch/pytorch/issues/21780,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: numerical-stability",scan when time allows,2019-07-10T13:35:07Z +issue,22609,caffe_translator TranslateCrop fails when more than one dimensions is cropped,https://github.com/pytorch/pytorch/issues/22609,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-10T08:01:21Z +issue,5280,BatchNorm1d raises RuntimeError (CUDNN_STATUS_BAD_PARAM) on 3D input.,https://github.com/pytorch/pytorch/issues/5280,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cudnn,scan when time allows,2019-07-10T07:59:44Z +issue,13304,ASSERT FAILED at /opt/conda/conda-bld/pytorch-nightly_1539602533843/work/aten/src/ATen/core/blob.h:79,https://github.com/pytorch/pytorch/issues/13304,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-10T06:09:18Z +issue,22580,Pytorch compilation error on Mac OS,https://github.com/pytorch/pytorch/issues/22580,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-07-09T21:01:49Z +issue,20682,[distribution] Support for various domain for AffineTransform,https://github.com/pytorch/pytorch/issues/20682,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2019-07-08T17:39:28Z +issue,22585,[dataloader] Mysterious error when using spawn start_method,https://github.com/pytorch/pytorch/issues/22585,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-07-08T16:44:13Z +issue,22578,torch.gels runs 100 time slower on gpu than on cpu,https://github.com/pytorch/pytorch/issues/22578,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2019-07-08T14:58:39Z +issue,22557,CPU torch.exponential_ function may generate 0 which can cause downstream NaN,https://github.com/pytorch/pytorch/issues/22557,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: random,scan when time allows,2019-07-08T14:36:40Z +issue,22494,Error in equation,https://github.com/pytorch/pytorch/issues/22494,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-07-07T10:46:33Z +issue,22536,libtorch new op,https://github.com/pytorch/pytorch/issues/22536,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: custom-operators, module: docs",scan when time allows,2019-07-06T21:12:05Z +issue,22532,Know which function is used by conv and force to use a function,https://github.com/pytorch/pytorch/issues/22532,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: nn",scan when time allows,2019-07-05T10:38:02Z +issue,22274,second derivatives of unfold,https://github.com/pytorch/pytorch/issues/22274,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: derivatives,scan when time allows,2019-07-04T16:37:43Z +issue,18011,caffe2 Segmentation fault (core dumped),https://github.com/pytorch/pytorch/issues/18011,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-04T07:17:07Z +issue,22040,SigAbort while running the Caffe2 unit test - thread_init_test - built on Clang7 +glibc 2.23,https://github.com/pytorch/pytorch/issues/22040,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-04T04:39:52Z +issue,22277,Handle all IntArrayRef expansions in ATen,https://github.com/pytorch/pytorch/issues/22277,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-07-03T09:01:41Z +issue,22406,Build failure with setup.py,https://github.com/pytorch/pytorch/issues/22406,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-07-03T06:16:06Z +issue,22455,"Move csrc/distributed/c10d/{comm,reducer} to libtorch.so",https://github.com/pytorch/pytorch/issues/22455,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-07-02T17:39:28Z +issue,22383,Storage operation failing on second GPU,https://github.com/pytorch/pytorch/issues/22383,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-07-01T21:38:37Z +issue,22375,Label.dim Enforcement Check in AccuracyOp,https://github.com/pytorch/pytorch/issues/22375,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-01T21:17:00Z +issue,22370,returned non-zero exit status 2.,https://github.com/pytorch/pytorch/issues/22370,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-07-01T20:56:44Z +issue,22358,"When I run python setup.py install to install Caffe2, I have an error: ""No such file or directory: 'nvcc': 'nvcc'""",https://github.com/pytorch/pytorch/issues/22358,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-07-01T20:45:17Z +issue,12308,nn.functional.linear() for sparse tensor,https://github.com/pytorch/pytorch/issues/12308,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: sparse",scan when time allows,2019-07-01T17:25:42Z +issue,22340,Reducer bucketing based on autograd profile,https://github.com/pytorch/pytorch/issues/22340,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-06-28T14:03:57Z +issue,17249,Proposal: Add __tensor_wrap__ method similar to numpy __array_wrap__,https://github.com/pytorch/pytorch/issues/17249,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: numpy",scan when time allows,2019-06-27T18:25:00Z +issue,7471,Make c10d/FileStore cache file descriptor,https://github.com/pytorch/pytorch/issues/7471,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-06-27T14:33:09Z +issue,22252,[libtorch] header warning suppression,https://github.com/pytorch/pytorch/issues/22252,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build warnings,scan when time allows,2019-06-27T03:53:38Z +issue,22220,[doc] nn.Module.forward documentation unclear,https://github.com/pytorch/pytorch/issues/22220,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: nn",scan when time allows,2019-06-26T14:40:33Z +issue,12819,Initialization error when moving data to the GPU,https://github.com/pytorch/pytorch/issues/12819,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-06-26T14:36:59Z +issue,15309,libtorch without cmake,https://github.com/pytorch/pytorch/issues/15309,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: cpp",scan when time allows,2019-06-26T02:56:09Z +issue,21899,Crash when using tensor.set_data() function in libtorch on windows,https://github.com/pytorch/pytorch/issues/21899,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-06-26T02:33:03Z +issue,21902,contradictory output values,https://github.com/pytorch/pytorch/issues/21902,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-06-26T02:28:14Z +issue,22232,[FYI] Introducing Quantized Tensor,https://github.com/pytorch/pytorch/issues/22232,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-06-25T21:30:01Z +issue,21777,typo and missing return statements,https://github.com/pytorch/pytorch/issues/21777,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: third_party",scan when time allows,2019-06-25T18:22:30Z +issue,8044,[PyTorch] Windows CI CUDA mem leak check on BN tests are flaky,https://github.com/pytorch/pytorch/issues/8044,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: memory usage",scan when time allows,2019-06-24T19:04:07Z +issue,9563,batch_sampler/test_worker_seed intermittently fails with address already in use on OS X,https://github.com/pytorch/pytorch/issues/9563,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: flaky-tests, module: serialization",scan when time allows,2019-06-24T19:03:02Z +issue,22136,[dataloader] SIGCHLD handler should poll the queue for exception first,https://github.com/pytorch/pytorch/issues/22136,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-06-24T18:30:33Z +issue,22035,Have a different way to check if gradient was computed in the optimizer (not checking for None),https://github.com/pytorch/pytorch/issues/22035,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: optimizer",scan when time allows,2019-06-24T18:21:50Z +issue,21847,Logging mode for saying when tensor broadcast occurs,https://github.com/pytorch/pytorch/issues/21847,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: molly-guard",scan when time allows,2019-06-24T18:17:29Z +issue,3931,Suppress hidden state output of RNNs?,https://github.com/pytorch/pytorch/issues/3931,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: memory usage,scan when time allows,2019-06-24T16:18:43Z +issue,21682,C++ module API footgun: assigning to parameter doesn't update `parameters()` list,https://github.com/pytorch/pytorch/issues/21682,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-06-23T14:36:11Z +issue,21673,[FR] Diagonal Transform for Distributions,https://github.com/pytorch/pytorch/issues/21673,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: distributions,scan when time allows,2019-06-23T12:07:17Z +issue,21731,Improve multithreaded random number generation (RNG),https://github.com/pytorch/pytorch/issues/21731,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multithreading, module: random",scan when time allows,2019-06-20T07:00:24Z +issue,19016,Clean up and consolidate DDP tests,https://github.com/pytorch/pytorch/issues/19016,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",scan when time allows,2019-06-18T22:13:21Z +issue,21828,torch::zeros is slow for small tensors (C++),https://github.com/pytorch/pytorch/issues/21828,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: performance",scan when time allows,2019-06-18T21:02:47Z +issue,20273,Linker errors when linking statically (avx perfkernels) [Caffe2],https://github.com/pytorch/pytorch/issues/20273,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-06-16T19:29:54Z +issue,21824,How about add torch::end for slicing in c++ frontend,https://github.com/pytorch/pytorch/issues/21824,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-06-15T22:32:04Z +issue,21766,Missing header files for prebuilt libtorch,https://github.com/pytorch/pytorch/issues/21766,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-06-14T20:56:59Z +issue,21277,[caffe2] check Range Operator inputs with bug,https://github.com/pytorch/pytorch/issues/21277,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-06-14T06:53:35Z +issue,8430,batchnorm2d track_running_stats,https://github.com/pytorch/pytorch/issues/8430,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2019-06-13T23:08:18Z +issue,21684,The cuda problem in caffe2,https://github.com/pytorch/pytorch/issues/21684,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-06-13T21:55:12Z +issue,21602,Strange latency overhead of F.conv2d,https://github.com/pytorch/pytorch/issues/21602,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: performance",scan when time allows,2019-06-12T15:26:06Z +issue,21477,Not obvious how to install torchvision with PyTorch source build,https://github.com/pytorch/pytorch/issues/21477,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2019-06-11T22:10:40Z +issue,21587,RuntimeError: cublas runtime error,https://github.com/pytorch/pytorch/issues/21587,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2019-06-10T22:14:56Z +issue,3468,improve performance of common CPU clone / contiguous calls with HPTT,https://github.com/pytorch/pytorch/issues/3468,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2019-06-10T19:33:16Z +issue,21459,RuntimeError: cublas runtime error,https://github.com/pytorch/pytorch/issues/21459,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cublas,scan when time allows,2019-06-10T17:41:01Z +issue,18206,Rename ignore_index to ignore_target in CrossEntropyLoss,https://github.com/pytorch/pytorch/issues/18206,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-06-09T00:35:50Z +issue,16655,"CUDA cache doubles on the second batch and causes OOM, `empty_cache` doesn't empty it",https://github.com/pytorch/pytorch/issues/16655,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cudnn, module: dependency bug",scan when time allows,2019-06-06T18:02:14Z +issue,21405,how libtorch can work with tensor data as same as pytorch,https://github.com/pytorch/pytorch/issues/21405,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpp, module: docs",scan when time allows,2019-06-06T15:25:06Z +issue,21399,collect_env ignores conda environment,https://github.com/pytorch/pytorch/issues/21399,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: collect_env.py, module: docs",scan when time allows,2019-06-05T14:58:53Z +issue,16257,why check ArgumentInfo is_pod? suffer bugs,https://github.com/pytorch/pytorch/issues/16257,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-06-05T14:32:32Z +issue,21290,"Title of docs page includes ""PyTorch master documentation"" even for non-master branches.",https://github.com/pytorch/pytorch/issues/21290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-06-04T18:31:15Z +issue,8913,Multiprocessing Self Test Error,https://github.com/pytorch/pytorch/issues/8913,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multiprocessing,scan when time allows,2019-06-04T18:22:45Z +issue,21165,Getting Access to Blob/Tensor reference in jit::script::Module,https://github.com/pytorch/pytorch/issues/21165,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-06-04T18:22:06Z +issue,14131,running caffe2 float16 tensors results in aten runtime error,https://github.com/pytorch/pytorch/issues/14131,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-06-02T20:19:36Z +issue,21135,[cmake build] can't build pytorch with install mkl library,https://github.com/pytorch/pytorch/issues/21135,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-05-31T07:59:26Z +issue,9204,/usr/bin/ld: cannot find -lpthreads,https://github.com/pytorch/pytorch/issues/9204,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-30T10:00:51Z +issue,21050,output values not same and much slower than Python API,https://github.com/pytorch/pytorch/issues/21050,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-05-30T00:10:32Z +issue,20975,"build caffe2 operators failed, 'sorry, unimplemented: non-trivial designated initializers not supported'",https://github.com/pytorch/pytorch/issues/20975,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-29T23:32:20Z +issue,16843,Allow to build pytorch for a *specific* architecture,https://github.com/pytorch/pytorch/issues/16843,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-05-29T23:29:33Z +issue,3990,Raise an error when using magma built against wrong version of cuda,https://github.com/pytorch/pytorch/issues/3990,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: binaries, module: build",scan when time allows,2019-05-29T23:24:27Z +issue,21012,"Error when creating new caffe2::Predictor(_initNet, _predictNet)",https://github.com/pytorch/pytorch/issues/21012,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-29T23:18:06Z +issue,8741,[Feature Request] Add to() method for optimizers/schedulers,https://github.com/pytorch/pytorch/issues/8741,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-05-29T23:00:35Z +issue,18677,How to compile/install caffe2 with cuda 9.0?,https://github.com/pytorch/pytorch/issues/18677,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-27T12:43:18Z +issue,20416,Class based Sampler for Class Incremental/Continual Learning research,https://github.com/pytorch/pytorch/issues/20416,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: dataloader,scan when time allows,2019-05-27T05:19:48Z +issue,20531,c++ torch::nn::Sequential increments count on name errors,https://github.com/pytorch/pytorch/issues/20531,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-05-26T19:42:49Z +issue,13836,"Pytorch-Caffe2 export: ""Arrays are not almost equal to 3 decimals""",https://github.com/pytorch/pytorch/issues/13836,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-26T06:37:24Z +issue,13664,Caffe2 install,https://github.com/pytorch/pytorch/issues/13664,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-24T03:05:05Z +issue,20753,cmake for Torch unusable in archlinux,https://github.com/pytorch/pytorch/issues/20753,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: dependency bug",scan when time allows,2019-05-22T02:15:11Z +issue,20776,Caffe2: executor_test.py::ExecutorGPUResNetTest fails with an AssertionError,https://github.com/pytorch/pytorch/issues/20776,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-21T22:08:19Z +issue,20777,"Caffe2: operator_test/instance_norm_test.py::TestInstanceNorm::test_instance_norm_gradients test case fails with ""RuntimeError: dim() called on undefined TensorError from operator""",https://github.com/pytorch/pytorch/issues/20777,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-21T22:07:47Z +issue,20562,Performance issue when accessing an extremely large (10GB) longtensor,https://github.com/pytorch/pytorch/issues/20562,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn",scan when time allows,2019-05-21T00:07:12Z +issue,20734,[jit] set up views for Autodiff and autograd hooks,https://github.com/pytorch/pytorch/issues/20734,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-05-20T23:34:38Z +issue,20641,Main page broadcasting (?) example image bug,https://github.com/pytorch/pytorch/issues/20641,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-05-20T19:25:48Z +issue,10232,[Caffe2] AVX not enabled for pytorch/caffe2,https://github.com/pytorch/pytorch/issues/10232,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-17T17:50:20Z +issue,20576,NVRTC_ERROR unknown when using self-built libtorch,https://github.com/pytorch/pytorch/issues/20576,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-05-17T09:39:18Z +issue,7035,[Compilation] how to disable caffe2?,https://github.com/pytorch/pytorch/issues/7035,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: build,scan when time allows,2019-05-16T23:18:58Z +issue,20272,Redundantly saving sizes of SavedVariables in autograd Function,https://github.com/pytorch/pytorch/issues/20272,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: autograd, module: bootcamp",scan when time allows,2019-05-15T19:50:00Z +issue,20481,[docs] Automatically detect docs missing in rst,https://github.com/pytorch/pytorch/issues/20481,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: tests",scan when time allows,2019-05-14T21:20:33Z +issue,20466,Lint rule to prevent direct use of #pragma omp,https://github.com/pytorch/pytorch/issues/20466,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: lint,scan when time allows,2019-05-14T21:05:22Z +issue,20412,String in tensor,https://github.com/pytorch/pytorch/issues/20412,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-05-13T21:34:39Z +issue,10577,[BUG]: unstable happend in saving model.,https://github.com/pytorch/pytorch/issues/10577,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: serialization,scan when time allows,2019-05-13T21:30:52Z +issue,20405,CosineAnnealingLR has unexpected behavior with large step,https://github.com/pytorch/pytorch/issues/20405,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-05-13T18:26:43Z +issue,20375,Better documentation / molly-guards around use of multiprocessing with spawn in Jupyter/ipython notebooks,https://github.com/pytorch/pytorch/issues/20375,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-05-13T17:59:13Z +issue,20363,"Define portable M_PI replacement, use it instead of non-standard M_PI in math.h",https://github.com/pytorch/pytorch/issues/20363,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-05-10T18:35:34Z +issue,20342,TensorIterator resizes output to a scalar if there are no inputs,https://github.com/pytorch/pytorch/issues/20342,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: internals,scan when time allows,2019-05-10T17:40:14Z +issue,5388,Perf regression: indexing 1-d tensor,https://github.com/pytorch/pytorch/issues/5388,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: performance,scan when time allows,2019-05-10T17:29:40Z +issue,13118,"warning: attribute namespace ""clang"" is unrecognized; High Sierra / Fedora compilation with clang results in spurious clang errors in nvcc",https://github.com/pytorch/pytorch/issues/13118,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: build, module: build warnings, module: cuda",scan when time allows,2019-05-09T15:18:20Z +issue,20183,Latex Errors when Compiling documentation to latexpdf,https://github.com/pytorch/pytorch/issues/20183,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: doc infra, module: docs",scan when time allows,2019-05-08T15:12:51Z +issue,20268,[Caffe2] Convert caffe to caffe2 with solverstate,https://github.com/pytorch/pytorch/issues/20268,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-08T13:48:20Z +issue,12401,AttributeError: Method VideoInput is not a registered operator,https://github.com/pytorch/pytorch/issues/12401,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-08T06:38:37Z +issue,19335,CUDA optimization: using `__restrict__` whenever possible,https://github.com/pytorch/pytorch/issues/19335,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-05-08T02:02:17Z +issue,20245,ProcessGroupMPI tests in test_c10d.py,https://github.com/pytorch/pytorch/issues/20245,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: tests, oncall: distributed",scan when time allows,2019-05-07T21:14:13Z +issue,20111,Split libtorch binary build CI job into separate variants,https://github.com/pytorch/pytorch/issues/20111,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2019-05-07T17:40:48Z +issue,20189,Unify Caffe2 and PyTorch OpenMP initialization,https://github.com/pytorch/pytorch/issues/20189,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: multithreading,scan when time allows,2019-05-07T16:48:27Z +issue,18436,UnpicklingError when trying to load multiple objects from a file,https://github.com/pytorch/pytorch/issues/18436,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: pickle, module: serialization",scan when time allows,2019-05-07T10:32:59Z +issue,20165,torch.nn.threshold cannot accept tensor as a threshold,https://github.com/pytorch/pytorch/issues/20165,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: nn,scan when time allows,2019-05-06T21:14:34Z +issue,17495,RuntimeError: storage_.IsType() ASSERT FAILED,https://github.com/pytorch/pytorch/issues/17495,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-06T19:31:08Z +issue,17350,"torch.nn.CrossEntropyLoss with ""reduction"" sum/mean is not deterministic on segmentation outputs / labels",https://github.com/pytorch/pytorch/issues/17350,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cuda,scan when time allows,2019-05-06T19:30:18Z +issue,20138,Inconsistant values of lr_scheduler.get_lr and lr in optimizer.param_groups,https://github.com/pytorch/pytorch/issues/20138,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-05-06T16:00:07Z +issue,20086,CosineAnnealingLR giving unexpected learning rates on PyTorch 1.1.,https://github.com/pytorch/pytorch/issues/20086,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: optimizer,scan when time allows,2019-05-03T21:28:24Z +issue,18000,Unsupported type of tensor c10::half when running resnet50_trainer.py in Caffe2,https://github.com/pytorch/pytorch/issues/18000,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-03T05:07:52Z +issue,19504,[jit] Bad error when calling `ScriptModule`s with attributes/parameters,https://github.com/pytorch/pytorch/issues/19504,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-05-03T00:06:39Z +issue,20022,Precision of sparse float embeddings differs from dense embeddings on CPU,https://github.com/pytorch/pytorch/issues/20022,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: nn, module: numerical-reproducibility",scan when time allows,2019-05-02T16:31:38Z +issue,19825,Performance issue master (a25b79531),https://github.com/pytorch/pytorch/issues/19825,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: multithreading, module: performance",scan when time allows,2019-05-01T23:31:58Z +issue,4632,CUDNN_STATUS_INTERNAL_ERROR when training with conv3d,https://github.com/pytorch/pytorch/issues/4632,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: convolution, module: cudnn",scan when time allows,2019-05-01T03:30:15Z +issue,19960,"On first construction, CUDAContext changes default CPU allocator behavior",https://github.com/pytorch/pytorch/issues/19960,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-05-01T00:42:59Z +issue,19826,vectorized convert_to_int_of_same_size can't handle nan,https://github.com/pytorch/pytorch/issues/19826,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cpu, module: error checking, module: nans and infs",scan when time allows,2019-05-01T00:18:08Z +issue,19930,caffe2/resnet50 assert in fetch_blob() when base_learning_rate = 0,https://github.com/pytorch/pytorch/issues/19930,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-30T16:19:34Z +issue,19408,Massive memory overhead over NumPy,https://github.com/pytorch/pytorch/issues/19408,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: internals, module: memory usage",scan when time allows,2019-04-30T11:13:53Z +issue,19263,GCP Base Image Wrong CUDA Version,https://github.com/pytorch/pytorch/issues/19263,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: doc infra,scan when time allows,2019-04-29T23:28:27Z +issue,19708,Test failure for depthwise3x3_conv,https://github.com/pytorch/pytorch/issues/19708,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-26T08:14:32Z +issue,19652,[docs] torch.set_flush_denormal(...) to mention default mode,https://github.com/pytorch/pytorch/issues/19652,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-04-24T22:12:14Z +issue,19637,"[jit] Traced {zeros,empty}()/{zeros,empty}_like() calls do not respect device args.",https://github.com/pytorch/pytorch/issues/19637,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: jit,scan when time allows,2019-04-24T21:59:18Z +issue,8153,GRU is implementation of GRU v1 draft rather than final GRU paper algo,https://github.com/pytorch/pytorch/issues/8153,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: docs, module: rnn",scan when time allows,2019-04-24T20:52:40Z +issue,19672,torch.set_flush_denormal not working on some (old) OSX machines,https://github.com/pytorch/pytorch/issues/19672,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: macos,scan when time allows,2019-04-24T20:46:08Z +issue,12498,Support calculating grad for dense in sparse @ dense,https://github.com/pytorch/pytorch/issues/12498,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2019-04-23T19:37:01Z +issue,19599,cudnn conv doesn't check batch_size > 0,https://github.com/pytorch/pytorch/issues/19599,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: cudnn, module: error checking",scan when time allows,2019-04-23T19:27:43Z +issue,19244,[Caffe2] Retraining saved model,https://github.com/pytorch/pytorch/issues/19244,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-22T14:00:15Z +issue,19416,[caffe2]Reloading model gives segmentation fault,https://github.com/pytorch/pytorch/issues/19416,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-19T21:15:20Z +issue,19499,C++ API 'nn::Sequential' has inconsistent behavior with python conterpart,https://github.com/pytorch/pytorch/issues/19499,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-04-19T20:21:51Z +issue,8290,[feature request] MPI init_method for torch.distributed,https://github.com/pytorch/pytorch/issues/8290,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-04-19T18:58:04Z +issue,19437,The `unary_kernel` call should be pushed into CopyKernel.cpp and it should completely replace the current `copy_kernel`.,https://github.com/pytorch/pytorch/issues/19437,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpu,scan when time allows,2019-04-18T21:54:21Z +issue,19330,Multiprocessing on distributed Multi-nodes shutdown error: ‘spawn’ on slave node leads to semaphore_tracker leaked,https://github.com/pytorch/pytorch/issues/19330,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-04-18T19:45:58Z +issue,6083,[feature request] DC-ASGD (Delay Compensated Asynchronous Stochastic Gradient Descent),https://github.com/pytorch/pytorch/issues/6083,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: optimizer, oncall: distributed",scan when time allows,2019-04-15T17:56:57Z +issue,19229,Different behavior of torch.nn.MultiMarginLoss on CPU/GPU Tensors,https://github.com/pytorch/pytorch/issues/19229,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",scan when time allows,2019-04-14T04:14:10Z +issue,16804,Wrong description of positive class weight in BCEWithLogitsLoss,https://github.com/pytorch/pytorch/issues/16804,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: docs,scan when time allows,2019-04-12T22:00:47Z +issue,12509,"The text design (color, type) makes it hard to read",https://github.com/pytorch/pytorch/issues/12509,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-11T16:27:17Z +issue,14632,[c10d] Check that allgather/gather output tensors point to different storage,https://github.com/pytorch/pytorch/issues/14632,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-04-11T16:20:27Z +issue,15491,LibTorch: include cmake files for all distributed headers,https://github.com/pytorch/pytorch/issues/15491,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: cpp,scan when time allows,2019-04-11T16:18:02Z +issue,19107,pytorch/caffe2/onnx/backend.cc:1668:57: error: invalid conversion from ‘google::protobuf::int32 {aka int}’ to ‘onnx_torch::TensorProto::DataType {aka onnx_torch::TensorProto_DataType}’ [-fpermissive],https://github.com/pytorch/pytorch/issues/19107,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-10T22:45:53Z +issue,19120,Embedding layer does not check input range,https://github.com/pytorch/pytorch/issues/19120,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: cuda, module: error checking",scan when time allows,2019-04-10T20:17:33Z +issue,18933,Completion of error handling,https://github.com/pytorch/pytorch/issues/18933,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: error checking, module: internals",scan when time allows,2019-04-09T00:32:00Z +issue,6408,[caffe2] Prefetching blobs for memory optimization,https://github.com/pytorch/pytorch/issues/6408,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-08T21:12:47Z +issue,18920,[c10d] CUDA tests for C++ reducer,https://github.com/pytorch/pytorch/issues/18920,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: oncall: distributed,scan when time allows,2019-04-08T16:45:51Z +issue,18998,torch.from_PIL() Request ?,https://github.com/pytorch/pytorch/issues/18998,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: vision,scan when time allows,2019-04-08T16:26:10Z +issue,18944,Caffe2 on google colab:,https://github.com/pytorch/pytorch/issues/18944,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-06T16:33:02Z +issue,16291,Confusing documentation with distributions.Categorical about logits,https://github.com/pytorch/pytorch/issues/16291,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: distributions, module: docs",scan when time allows,2019-04-05T14:11:02Z +issue,17897,CUDA large matrix-vector product (torch.mv) causes illegal memory access,https://github.com/pytorch/pytorch/issues/17897,stale/low urgency,2,"old update age or low-urgency/stale label; component/oncall labels: module: 64-bit, module: cublas, module: cuda, module: dependency bug",scan when time allows,2019-04-04T16:59:15Z +issue,15120,Different implementations of upsampleBilinear between pytorch and caffe2,https://github.com/pytorch/pytorch/issues/15120,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-03T23:47:30Z +issue,1369,"""Sparsified"" mathematical operations",https://github.com/pytorch/pytorch/issues/1369,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: sparse,scan when time allows,2019-04-03T20:06:33Z +issue,18658,[FR] add CPU information in collect_env.py,https://github.com/pytorch/pytorch/issues/18658,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: collect_env.py,scan when time allows,2019-04-02T22:25:43Z +issue,18692,Add build tests for feature environment vars,https://github.com/pytorch/pytorch/issues/18692,stale/low urgency,2,old update age or low-urgency/stale label; component/oncall labels: module: ci,scan when time allows,2019-04-02T22:23:04Z +issue,6357,[feature request] Unpooling layer in Caffe2,https://github.com/pytorch/pytorch/issues/6357,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-01T19:23:09Z +issue,16299,caffe2::CudnnConvOp::RunOnDevice() fails on Squeezenet,https://github.com/pytorch/pytorch/issues/16299,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-04-01T01:42:09Z +issue,8886,[Caffe2] Dose caffe2 support the function similar to itersize of caffe,https://github.com/pytorch/pytorch/issues/8886,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-28T01:29:16Z +issue,18273,[caffe2] Broken Operators Catalog,https://github.com/pytorch/pytorch/issues/18273,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-25T13:06:58Z +issue,14731,Caffe2 C++ runs single threaded,https://github.com/pytorch/pytorch/issues/14731,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-22T00:44:48Z +issue,17936,[caffe2] resize_op_test.py::TestResize::test_nearest FAILED,https://github.com/pytorch/pytorch/issues/17936,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-18T19:44:13Z +issue,17800,[Caffe2] cudnn mismatch,https://github.com/pytorch/pytorch/issues/17800,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-15T17:20:47Z +issue,14486,"building bundled nccl fails in (caffe2, cuda 8, cudnn 7) CI environment",https://github.com/pytorch/pytorch/issues/14486,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-13T18:13:20Z +issue,17907,"[caffe2] resnet_trainer example fails with MNIST dataset, when parameter num_channels=1 is provided",https://github.com/pytorch/pytorch/issues/17907,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-12T13:23:21Z +issue,17887,Caffe2 building failure,https://github.com/pytorch/pytorch/issues/17887,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-11T21:32:55Z +issue,17872,Linking Caffe2 with sequential MKL but still calling multi-threaded mkl,https://github.com/pytorch/pytorch/issues/17872,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-11T16:48:22Z +issue,17632,Caffe2 C++ script for classification/object_detection with CMakeLists.txt,https://github.com/pytorch/pytorch/issues/17632,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-10T21:41:36Z +issue,17850,Build fails for caffe2/CMakeFiles/caffe2.dir/__/aten/src/ATen/native/mkldnn/Conv.cpp.o; possibly MKL-DNN issue,https://github.com/pytorch/pytorch/issues/17850,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-10T16:50:53Z +issue,17712,Can we to build Caffe2 custom ops independently from a Caffe2 build ?,https://github.com/pytorch/pytorch/issues/17712,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-08T04:26:20Z +issue,15630,[Caffe2] Internal compiler error for CUDA.,https://github.com/pytorch/pytorch/issues/15630,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-07T09:12:37Z +issue,17634,[Caffe2] android app crashed while linked with libcaffe2_detectron_ops.so,https://github.com/pytorch/pytorch/issues/17634,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-03-02T12:41:38Z +issue,15290,[caffe2] Installation problem on OSX,https://github.com/pytorch/pytorch/issues/15290,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-02-24T17:03:37Z +issue,17437,Automatic aggregation of a mix of sparse and dense gradients is not supported yet,https://github.com/pytorch/pytorch/issues/17437,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-02-24T04:16:24Z +issue,16666,numa_test.py fails with AssertionError,https://github.com/pytorch/pytorch/issues/16666,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-02-04T16:43:27Z +issue,16590,Benchmarking pre-trained model in Caffe2,https://github.com/pytorch/pytorch/issues/16590,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-31T02:52:52Z +issue,16280,caffe2: cudaHostRegister() and mbind()-related test failures on ppc64le,https://github.com/pytorch/pytorch/issues/16280,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-29T23:52:29Z +issue,16494,caffe2 softmaxwithloss problem,https://github.com/pytorch/pytorch/issues/16494,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-29T08:36:44Z +issue,16436,How to retrain the modelzoo model in Caffe2?,https://github.com/pytorch/pytorch/issues/16436,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-28T09:02:59Z +issue,16385,Adding new module to caffe2,https://github.com/pytorch/pytorch/issues/16385,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-25T22:37:57Z +issue,11354,caffe2 failed to build from source,https://github.com/pytorch/pytorch/issues/11354,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-24T22:07:29Z +issue,16254,I hope Caffe2's python interface add DataLoader Module Similar to pytorch's DataSet and DataLoader,https://github.com/pytorch/pytorch/issues/16254,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-23T09:22:26Z +issue,9704,"A newer protobuf should be used, so that cmake install will work when building for mobile",https://github.com/pytorch/pytorch/issues/9704,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-23T05:25:08Z +issue,6868,[Caffe2] Android NNApi integration bugs.,https://github.com/pytorch/pytorch/issues/6868,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-22T00:32:29Z +issue,16195,how to implement crf predict for segmentation task in Caffe2?,https://github.com/pytorch/pytorch/issues/16195,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-20T03:02:21Z +issue,16138,[Caffe2] resize5d stride/dim issue,https://github.com/pytorch/pytorch/issues/16138,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-18T04:45:17Z +issue,13598,caffe2: RuntimeError: [enforce fail at reshape_op.h:110] with Alexnet onnx test with cuda,https://github.com/pytorch/pytorch/issues/13598,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-16T15:51:43Z +issue,14939,caffe2/predictor missing from libtorch,https://github.com/pytorch/pytorch/issues/14939,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-16T10:03:35Z +issue,14989,[Caffe2] CNN Training on CPU is faster than GPU,https://github.com/pytorch/pytorch/issues/14989,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-14T21:11:29Z +issue,15869,RuntimeError: [enforce fail at pybind_state.cc:1111] success. Error running net train,https://github.com/pytorch/pytorch/issues/15869,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-14T14:08:18Z +issue,9676,"[caffe2]running problem, No module named caffe2_pybind11_state_hip",https://github.com/pytorch/pytorch/issues/9676,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-13T09:53:07Z +issue,15432,caffe2: multi-thread(or multi-instance) predict is much slower than the single thread or single instance,https://github.com/pytorch/pytorch/issues/15432,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-12T01:09:35Z +issue,14459,[Caffe2] Error protos.protos_size() == OutputSize() when loading dataset created by regular Caffe (datum),https://github.com/pytorch/pytorch/issues/14459,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-11T08:02:18Z +issue,14940,caffe2::Predictor: Cannot find operator schema for GivenTensorFill. Will skip schema checking.,https://github.com/pytorch/pytorch/issues/14940,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-11T02:02:26Z +issue,15431,IDEEP error:could not initialize a memory descriptor,https://github.com/pytorch/pytorch/issues/15431,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-11T00:38:28Z +issue,15771,Implicit conversion error in caffe2,https://github.com/pytorch/pytorch/issues/15771,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-10T17:18:35Z +issue,15465,Transplant caffe2 on my DNN-accelerator,https://github.com/pytorch/pytorch/issues/15465,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-10T17:16:21Z +issue,13772,Error in building Caffe2 on Windows (experimental operators),https://github.com/pytorch/pytorch/issues/13772,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-09T20:03:58Z +issue,11865,[Caffe2] Errors occured when running Cpp Predictor,https://github.com/pytorch/pytorch/issues/11865,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-08T03:03:26Z +issue,12322,[Caffe2] Segmentation fault (core dumped) while import caffe2.python.core,https://github.com/pytorch/pytorch/issues/12322,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-07T16:37:22Z +issue,15004,[Caffe2] How to link Caffe2 in cmake file for C++ compilation?,https://github.com/pytorch/pytorch/issues/15004,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-07T01:14:48Z +issue,14963,PyTorch 1.0 source build fails (error in caffe2/utils/fatal_signal_asan_no_sig_test.cc),https://github.com/pytorch/pytorch/issues/14963,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2019-01-02T08:45:14Z +issue,15608,[Caffe2] How to switch to test phase?,https://github.com/pytorch/pytorch/issues/15608,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-30T16:50:06Z +issue,9531,[caffe2]train on GPU and test on cpu failed,https://github.com/pytorch/pytorch/issues/9531,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-30T16:22:10Z +issue,15249,[caffe2] controlled forward pass,https://github.com/pytorch/pytorch/issues/15249,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-20T16:58:34Z +issue,10471,[Caffe2] Multithreading in Caffe2,https://github.com/pytorch/pytorch/issues/10471,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-20T06:31:54Z +issue,12530,[caffe2] Memory usage,https://github.com/pytorch/pytorch/issues/12530,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-20T04:28:34Z +issue,15313,AttributeError: module 'caffe2.python._import_c_extension' has no attribute 'get_cudnn_version' when Caffe2 is not built with CuDNN,https://github.com/pytorch/pytorch/issues/15313,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-17T22:10:25Z +issue,14925,Am I able to run caffe2 and gloo example: resnet50_trainer.py using pytorch v1.0rc1?,https://github.com/pytorch/pytorch/issues/14925,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-17T07:24:34Z +issue,11735,Caffe2 installation: libcaffe2_gpu.so: undefined reference to `caffe2::ClipTransformRGB,https://github.com/pytorch/pytorch/issues/11735,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-14T07:51:22Z +issue,12007,caffe2's installation is still problematic...,https://github.com/pytorch/pytorch/issues/12007,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-11T09:58:38Z +issue,15009,[Caffe2] Caffe2Config.cmake,https://github.com/pytorch/pytorch/issues/15009,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-10T20:06:00Z +issue,14726,[caffe2] Corresponding C++ API for prepare_prediction_net,https://github.com/pytorch/pytorch/issues/14726,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-10T16:23:36Z +issue,14788,[caffe2] [NNPACK] how to set thread pool for NNPACK?,https://github.com/pytorch/pytorch/issues/14788,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-10T16:21:18Z +issue,14799,which operator of caffe2 have the same function as torch.nn.Parameter,https://github.com/pytorch/pytorch/issues/14799,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-10T16:20:27Z +issue,14970,"[Caffe2] ""Same"" padding",https://github.com/pytorch/pytorch/issues/14970,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-09T23:01:32Z +issue,14796,Cannot build Caffe2 with TensorRT,https://github.com/pytorch/pytorch/issues/14796,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-05T17:51:05Z +issue,7078,Fail to import Caffe2Backend,https://github.com/pytorch/pytorch/issues/7078,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-04T20:45:12Z +issue,11861,[Caffe2 installation problem],https://github.com/pytorch/pytorch/issues/11861,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-04T12:16:59Z +issue,14652,fc_without_bias / FCWithoutBias,https://github.com/pytorch/pytorch/issues/14652,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-03T18:27:42Z +issue,14685,[Caffe2] Exception when creating gradient for [Cast] SquaredL2Distance as output layer of CNN network,https://github.com/pytorch/pytorch/issues/14685,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-12-03T18:21:47Z +issue,13493,How to do inference with float16 (or half) in caffe2?,https://github.com/pytorch/pytorch/issues/13493,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-30T10:20:48Z +issue,12913,ubuntu16.04 build from source caffe2,https://github.com/pytorch/pytorch/issues/12913,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-29T19:43:56Z +issue,14508,[Caffe2] GPU test passed. Cannot see on nvidia-smi,https://github.com/pytorch/pytorch/issues/14508,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-29T15:40:20Z +issue,14538,[Caffe2] How to fetch trainable parameters?,https://github.com/pytorch/pytorch/issues/14538,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-29T15:33:35Z +issue,14408,DeviceOption::set_device_type() doesn't accept caffe2::CPU anymore,https://github.com/pytorch/pytorch/issues/14408,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-28T09:55:44Z +issue,14431,BUILD_CAFFE2_OPS=OFF is not tested in CI,https://github.com/pytorch/pytorch/issues/14431,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-27T22:28:22Z +issue,14380,[caffe2] Caffe2 GlobalInit should be run before any other API calls,https://github.com/pytorch/pytorch/issues/14380,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T19:15:09Z +issue,14366,How to use model.net.Clip?,https://github.com/pytorch/pytorch/issues/14366,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:23:09Z +issue,14312,[Caffe2] Cannot get repeated argument in custom operator in CUDA context,https://github.com/pytorch/pytorch/issues/14312,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:20:23Z +issue,14185,[Caffe2] Non-spatial batchnorm optimization,https://github.com/pytorch/pytorch/issues/14185,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:20:21Z +issue,14222,Caffe2->ONNX conversion issue due to spatial-bn,https://github.com/pytorch/pytorch/issues/14222,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:08:10Z +issue,14262,[Caffe2] How can I use detectron with pytorch?,https://github.com/pytorch/pytorch/issues/14262,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:07:58Z +issue,14320,[caffe2] Fails to build with fbgemm enabled,https://github.com/pytorch/pytorch/issues/14320,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:07:36Z +issue,14328,[Caffe2] How to use the euclidean loss (L2) as output of a CNN model?,https://github.com/pytorch/pytorch/issues/14328,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:07:29Z +issue,14337,Caffe2 C++ tutorial is not working,https://github.com/pytorch/pytorch/issues/14337,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:07:27Z +issue,14348,[Caffe2] Error when loading a leveldb dataset using brew.db_input (Error protos.protos_size() == OutputSize().),https://github.com/pytorch/pytorch/issues/14348,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-26T18:07:22Z +issue,10940,[Caffe2] Model work Alright in Python but Failed in C++,https://github.com/pytorch/pytorch/issues/10940,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-23T15:44:26Z +issue,13226,[Caffe2] cmake3 detection error?,https://github.com/pytorch/pytorch/issues/13226,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-22T04:09:42Z +issue,8471,Does the caffe2 have the depthwise separable convolution oprater ?,https://github.com/pytorch/pytorch/issues/8471,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-21T09:28:44Z +issue,12515,how to use mask-rcnn in caffe2 c++ gpu,https://github.com/pytorch/pytorch/issues/12515,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-20T10:09:40Z +issue,14030,Cannot import caffe2_pybind11_state_gpu,https://github.com/pytorch/pytorch/issues/14030,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-19T18:45:10Z +issue,13715,caffe2 c++ load pb model fail,https://github.com/pytorch/pytorch/issues/13715,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-14T15:07:24Z +issue,13655,[caffe2] How to export onnx model trained on Detectron in Caffe2?,https://github.com/pytorch/pytorch/issues/13655,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-12T18:41:55Z +issue,13671,Caffe2 Build Static,https://github.com/pytorch/pytorch/issues/13671,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-12T18:37:35Z +issue,13667,[caffe2] Modify models in model zoo,https://github.com/pytorch/pytorch/issues/13667,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-12T18:37:34Z +issue,13684,[caffe2] Adding CUDA operators for generate proposals and NMS layers,https://github.com/pytorch/pytorch/issues/13684,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-12T18:37:13Z +issue,11791,[caffe2]Can i build caffe2 library only for cpu inference purpose and reduce the binary size?,https://github.com/pytorch/pytorch/issues/11791,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-10T15:17:24Z +issue,11636,undefined reference to caffe2,https://github.com/pytorch/pytorch/issues/11636,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-07T08:10:09Z +issue,13208,[caffe2] How to use the operators that are not included in brew through the python API?,https://github.com/pytorch/pytorch/issues/13208,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-06T22:14:05Z +issue,13297,[Caffe2]How to accelerate group convolution in Caffe2?,https://github.com/pytorch/pytorch/issues/13297,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-06T17:39:30Z +issue,13498,Caffe2 missing headers in docker/conda,https://github.com/pytorch/pytorch/issues/13498,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-05T18:52:45Z +issue,13373,cannot run mobilenet_v2_quantized on pytorch/caffe2,https://github.com/pytorch/pytorch/issues/13373,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-02T00:38:43Z +issue,12931,USE_OPENMP=OFF is ignored [Caffe2],https://github.com/pytorch/pytorch/issues/12931,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-11-01T13:56:47Z +issue,6618,[Caffe2] CUDNN_STATUS_BAD_PARAM Error with the LRN layer while trying to run the code using CUDA. The training works fine on CPU,https://github.com/pytorch/pytorch/issues/6618,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-31T22:48:09Z +issue,13405,[caffe2] How to use Split operator?,https://github.com/pytorch/pytorch/issues/13405,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-31T18:56:09Z +issue,13041,[caffe2] How to handle multiple inputs and multiple outputs in the network architecture?,https://github.com/pytorch/pytorch/issues/13041,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-26T17:57:36Z +issue,13120,[caffe2] incompatible constructor arguments,https://github.com/pytorch/pytorch/issues/13120,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-25T14:38:02Z +issue,12808,tutorial_blob ERROR,https://github.com/pytorch/pytorch/issues/12808,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-25T00:09:02Z +issue,12576,[feature request] `ignore_label` argument in Caffe2 `SoftmaxWithLoss`,https://github.com/pytorch/pytorch/issues/12576,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-23T19:39:13Z +issue,12642,fail to visualize caffe2 model,https://github.com/pytorch/pytorch/issues/12642,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-23T16:28:27Z +issue,12879,Eigen in Caffe2 doesn't produce vectorized instructions,https://github.com/pytorch/pytorch/issues/12879,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-22T19:15:19Z +issue,12755,caffe2 python custom operator could not update parameters,https://github.com/pytorch/pytorch/issues/12755,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-22T17:59:18Z +issue,7912,[Caffe2] Operators of Detectron module not registered/compiled when built on windows,https://github.com/pytorch/pytorch/issues/7912,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-20T20:49:15Z +issue,12501,C++: Calling Workspace::RunNet for a prediction on a different thread each time causes a GPU memory leak,https://github.com/pytorch/pytorch/issues/12501,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-18T23:12:23Z +issue,12763,"Caffe2: Two entries of external_input ""data_0"" in mnist_predict_net.pbtxt file",https://github.com/pytorch/pytorch/issues/12763,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-17T23:19:58Z +issue,12646,Caffe2 Installation inside Pytorch,https://github.com/pytorch/pytorch/issues/12646,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-17T19:49:45Z +issue,12640,how to store a bounding box in Tensor?,https://github.com/pytorch/pytorch/issues/12640,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-15T17:54:12Z +issue,12658,Move BigTensorSerialization tests out of default caffe2_cpu_tests,https://github.com/pytorch/pytorch/issues/12658,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-15T17:30:09Z +issue,12280,`GLIBC_2.23' not found on Ubuntu14.04.,https://github.com/pytorch/pytorch/issues/12280,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-08T17:54:51Z +issue,7127,how can i use openMP for caffe2? why caffe2 not work in multi-threads mode?,https://github.com/pytorch/pytorch/issues/7127,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-08T05:42:14Z +issue,6660,[caffe2] GANs,https://github.com/pytorch/pytorch/issues/6660,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-05T22:33:02Z +issue,12297,[caffe2] Inference using multiple GPU,https://github.com/pytorch/pytorch/issues/12297,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-03T21:34:50Z +issue,12257,[Caffe2] Relink error after installing Caffe2 from conda,https://github.com/pytorch/pytorch/issues/12257,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-02T09:27:29Z +issue,12045,Create a custom function using the functions from math_cpu.cc source code in Caffe2?,https://github.com/pytorch/pytorch/issues/12045,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-01T17:47:52Z +issue,12086,Caffe2 issues with using Glog without GFlags,https://github.com/pytorch/pytorch/issues/12086,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-01T17:43:23Z +issue,12134,[caffe2] How to set different learning rates for different layers?,https://github.com/pytorch/pytorch/issues/12134,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-01T17:39:39Z +issue,12181,Network surgery for transfer fails,https://github.com/pytorch/pytorch/issues/12181,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-10-01T17:26:23Z +issue,8944,[caffe2] UnicodeDecodeError when running LeNet,https://github.com/pytorch/pytorch/issues/8944,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-29T06:09:53Z +issue,11850,[Caffe2/Bug] Cannot enable MKL-DNN,https://github.com/pytorch/pytorch/issues/11850,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-27T20:21:56Z +issue,12136,[Caffe2] GAN No Gradients in Generator,https://github.com/pytorch/pytorch/issues/12136,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-27T13:37:39Z +issue,10675,GenerateProposals CUDA implementation,https://github.com/pytorch/pytorch/issues/10675,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-26T10:25:15Z +issue,11885,[caffe2] Bug for softmaxwithloss operator,https://github.com/pytorch/pytorch/issues/11885,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-24T17:51:07Z +issue,11890,caffe2 argmax and argmin documentation incorrect for output type,https://github.com/pytorch/pytorch/issues/11890,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-24T17:49:35Z +issue,11996,Test the Caffe2 Installation with GPU erro,https://github.com/pytorch/pytorch/issues/11996,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-24T17:25:54Z +issue,12010,Caffe2 compiled with MKLDNN doesn't have device_type = MKLDNN,https://github.com/pytorch/pytorch/issues/12010,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-24T17:19:11Z +issue,9975,python caffe2/python/operator_test/activation_ops_test.py Segmentation fault (core dumped),https://github.com/pytorch/pytorch/issues/9975,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-24T09:59:39Z +issue,11976,[caffe2] Documentation of Optimizers (Python API),https://github.com/pytorch/pytorch/issues/11976,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-22T15:57:14Z +issue,10875,how can i run two model in a simple project?,https://github.com/pytorch/pytorch/issues/10875,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-18T12:59:21Z +issue,11678,[caffe2] adam_op implementation is incorrect.,https://github.com/pytorch/pytorch/issues/11678,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-14T00:32:48Z +issue,10254,"[Caffe2] I don't knw whether GPU support is right, is there something wrong with Hypothesis?",https://github.com/pytorch/pytorch/issues/10254,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-11T02:38:45Z +issue,11363,Hypothesis operator tests in Caffe2 generate too many warnings,https://github.com/pytorch/pytorch/issues/11363,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-10T22:13:18Z +issue,11433,Assorted issues in Caffe2's Metal ops,https://github.com/pytorch/pytorch/issues/11433,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-10T17:24:59Z +issue,9708,"[caffe2]mpirun multi-node multi GPU in Distributed mode ,run resnet50_trainer.py get RuntimeError",https://github.com/pytorch/pytorch/issues/9708,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-10T13:06:41Z +issue,9484,WARNING:root:This caffe2 python run does not have GPU support. Will run in CPU only mode. WARNING:root:Debug message: libcurand.so.9.0: cannot open shared object file: No such file or directory Segmentation fault (core dumped),https://github.com/pytorch/pytorch/issues/9484,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-09T15:57:08Z +issue,11267,How to simulate multi-node using single-node with 8 GPUs,https://github.com/pytorch/pytorch/issues/11267,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-07T03:24:05Z +issue,11268,How to re-shuffle lmdb per epoch in the caffe2 training process,https://github.com/pytorch/pytorch/issues/11268,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-07T03:23:51Z +issue,10984,[caffe2] Implement Gather for any value of `axis`,https://github.com/pytorch/pytorch/issues/10984,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-04T16:52:26Z +issue,10990,[Caffe2] [Question] How to disable the bias parameter?,https://github.com/pytorch/pytorch/issues/10990,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-04T16:52:22Z +issue,11058,A bug in roi_align.cc? [Caffe2],https://github.com/pytorch/pytorch/issues/11058,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-04T16:49:15Z +issue,11180,A Caffe2 Implementation of Pose Estimation,https://github.com/pytorch/pytorch/issues/11180,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-04T16:33:01Z +issue,11197,[Caffe2] How to correctly add Caffe2 libraries to Visual Studio to write C++ programs?,https://github.com/pytorch/pytorch/issues/11197,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-09-03T08:02:18Z +issue,11119,[Caffe2] Error C2492: data with thread storage duration may not have dll interface,https://github.com/pytorch/pytorch/issues/11119,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-31T10:27:43Z +issue,10991,Error when building,https://github.com/pytorch/pytorch/issues/10991,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-31T01:56:45Z +issue,10714,[feature request][caffe2] extend FC/FCTranspose op to handle 2d bias.,https://github.com/pytorch/pytorch/issues/10714,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-30T04:17:30Z +issue,10985,Where could I see all of the caffe2 operators and their arguments in python API?,https://github.com/pytorch/pytorch/issues/10985,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-29T07:48:50Z +issue,10902,LNK2019 error when linking with MSVC [Caffe2],https://github.com/pytorch/pytorch/issues/10902,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-28T18:28:02Z +issue,10808,[Feature Request] Crop op or ConvTranspose with output_shape,https://github.com/pytorch/pytorch/issues/10808,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-23T05:43:59Z +issue,7946,how to use Softmax when do segmentation,https://github.com/pytorch/pytorch/issues/7946,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-23T02:41:34Z +issue,10738,[caffe2] the caffe2 operators document is too old,https://github.com/pytorch/pytorch/issues/10738,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-22T17:25:01Z +issue,7047,"""Undefined symbols for architecture arm64"" when linking libcaffe2.a for iOS",https://github.com/pytorch/pytorch/issues/7047,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-21T22:13:24Z +issue,10582,[Caffe2] Unable to use MPI rendezvous in Caffe2,https://github.com/pytorch/pytorch/issues/10582,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-20T17:47:35Z +issue,7789,detectron net create error,https://github.com/pytorch/pytorch/issues/7789,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-20T01:54:53Z +issue,10341,[Caffe2] Which document should we follow to install caffe2?,https://github.com/pytorch/pytorch/issues/10341,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-19T17:22:27Z +issue,10515,[Caffe2] Segmentation faults in multithreading Caffe2,https://github.com/pytorch/pytorch/issues/10515,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-17T15:28:32Z +issue,10161,error when building caffe2 from source with gcc 7 & cuda gcc 5.5,https://github.com/pytorch/pytorch/issues/10161,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-16T06:57:09Z +issue,10314,[Caffe2] ConvPoolOpBase operator,https://github.com/pytorch/pytorch/issues/10314,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-13T17:58:36Z +issue,10342,[Caffe2] which kind of database can I read from caffe2,https://github.com/pytorch/pytorch/issues/10342,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-13T17:57:10Z +issue,10461,Key already registered. Offending key: caffe2_print_stacktraces.,https://github.com/pytorch/pytorch/issues/10461,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-13T17:29:49Z +issue,10460,Type name float registered twice. This should not happen. Do you have duplicated CAFFE_KNOWN_TYPE?,https://github.com/pytorch/pytorch/issues/10460,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-13T17:25:03Z +issue,10215,[Caffe2] Build failure on Ubuntu 16.04,https://github.com/pytorch/pytorch/issues/10215,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-13T16:21:53Z +issue,8982,parallelize_bmuf_distributed_test intermittently hangs,https://github.com/pytorch/pytorch/issues/8982,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-13T15:07:47Z +issue,10070,UnicodeDecodeError while loading caffe2 model,https://github.com/pytorch/pytorch/issues/10070,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-07T08:36:36Z +issue,9761,[caffe2] Questions about conv_transpose && ConvTranspose,https://github.com/pytorch/pytorch/issues/9761,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-06T18:01:48Z +issue,10176,[Caffe2] Modelling using Ops instead of Helper Functions,https://github.com/pytorch/pytorch/issues/10176,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-06T17:35:30Z +issue,10256,[Caffe 2] How to access/remove Blobs from Workspace?,https://github.com/pytorch/pytorch/issues/10256,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-06T17:20:48Z +issue,10007,Upgrade Caffe2 Hypothesis,https://github.com/pytorch/pytorch/issues/10007,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-06T12:33:43Z +issue,9016,SimpleMetaNetDefInitializer intermittently hangs,https://github.com/pytorch/pytorch/issues/9016,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-04T18:08:02Z +issue,10170,build error,https://github.com/pytorch/pytorch/issues/10170,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-03T18:28:23Z +issue,9921,A serious problem when installing caffe2. Can anyone help me?,https://github.com/pytorch/pytorch/issues/9921,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-03T09:37:47Z +issue,10088,[Caffe2] Implementing CoordConv layer,https://github.com/pytorch/pytorch/issues/10088,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-08-02T16:23:19Z +issue,9853,TestSequenceOps.test_gather_padding failing,https://github.com/pytorch/pytorch/issues/9853,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-30T17:59:04Z +issue,9854,TestConvolution.test_conv_separate_stride_pad_gradients failing,https://github.com/pytorch/pytorch/issues/9854,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-30T17:58:57Z +issue,9881,cannot open caffe2.lib in VS2015 debug mode,https://github.com/pytorch/pytorch/issues/9881,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-30T17:57:00Z +issue,9883,NetTest.OperatorWithExecutorHelper intermittently hangs,https://github.com/pytorch/pytorch/issues/9883,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-30T17:56:46Z +issue,9945,WERROR=1 doesn't work with FULL_CAFFE2,https://github.com/pytorch/pytorch/issues/9945,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-30T17:46:27Z +issue,9150,[caffe2] Drop connections,https://github.com/pytorch/pytorch/issues/9150,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-23T20:42:37Z +issue,9028,"[caffe2]build problem, can not find caffe2_pybind11_state_hip",https://github.com/pytorch/pytorch/issues/9028,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-21T06:57:38Z +issue,9449,PredictorTest.SimpleBatchSizedMapInput intermittently hangs,https://github.com/pytorch/pytorch/issues/9449,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-20T01:08:35Z +issue,9166,nvcc fatal : A single input file is required for a non-link phase when an outputfile is specified,https://github.com/pytorch/pytorch/issues/9166,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-16T17:10:03Z +issue,7571,"the latest version of caffe2, Which file is the function loadToNCHW()",https://github.com/pytorch/pytorch/issues/7571,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-16T14:03:40Z +issue,9218,[Caffe2] Cannot load caffe2.python. Error: libcaffe2.so: cannot open shared object file: No such file or directory,https://github.com/pytorch/pytorch/issues/9218,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-14T15:07:36Z +issue,9207,Where is the include and lib path for caffe2?,https://github.com/pytorch/pytorch/issues/9207,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-14T03:58:25Z +issue,9174,[Caffe2] build_ios.sh => 'is only available on iOS 11 or newer',https://github.com/pytorch/pytorch/issues/9174,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-11T17:57:43Z +issue,9024,cmake error:Could NOT find IDEEP (missing: /usr/local/include),https://github.com/pytorch/pytorch/issues/9024,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-09T16:56:34Z +issue,8984,[Caffe2] segmentation fault,https://github.com/pytorch/pytorch/issues/8984,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-03T19:09:05Z +issue,9003,cmake error,https://github.com/pytorch/pytorch/issues/9003,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-03T19:02:13Z +issue,8979,Check failed: error == cudaSuccess unspecified launch failure [caffe2],https://github.com/pytorch/pytorch/issues/8979,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-02T17:55:20Z +issue,9068,Deployment for ios for 1.0,https://github.com/pytorch/pytorch/issues/9068,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-02T17:34:49Z +issue,9098,[caffe2] how to substract the mean value ?,https://github.com/pytorch/pytorch/issues/9098,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-02T17:29:04Z +issue,8730,[caffe2] AffineChannelOp,https://github.com/pytorch/pytorch/issues/8730,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-07-01T07:26:15Z +issue,7781,[Caffe2] Fail to build after upgrading to cuda 9.2,https://github.com/pytorch/pytorch/issues/7781,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-28T17:36:52Z +issue,8012,[Caffe2] VideoInput/LMDB Reader to Standalone Predictor Question,https://github.com/pytorch/pytorch/issues/8012,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-27T03:03:10Z +issue,8902,[Caffe 2] Does caffe2 support Galaxy 5S with Android 6?,https://github.com/pytorch/pytorch/issues/8902,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-26T18:46:52Z +issue,8650,Cannot use IterOp at runtime CUDA&CPU,https://github.com/pytorch/pytorch/issues/8650,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-20T11:16:10Z +issue,8648,Cannot allocate memory Error from operator,https://github.com/pytorch/pytorch/issues/8648,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-19T16:04:18Z +issue,8479,An error occurred while creating a new notebook.,https://github.com/pytorch/pytorch/issues/8479,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-18T17:40:45Z +issue,8548,[caffe2] [feature request]Doese caffe2 support conv_nd with group?,https://github.com/pytorch/pytorch/issues/8548,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-18T17:35:50Z +issue,8607,[Caffe2] Wrong prediction with simple FF,https://github.com/pytorch/pytorch/issues/8607,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-18T17:17:58Z +issue,8546,Q: how to generate my own pb files to C++ Predictor,https://github.com/pytorch/pytorch/issues/8546,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-15T05:25:06Z +issue,8128,Cmake is getting permission denied when installed system wide,https://github.com/pytorch/pytorch/issues/8128,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-12T18:26:30Z +issue,8186,[Caffe2] Successive in-place operators cause RuntimeError of gradient operator versions,https://github.com/pytorch/pytorch/issues/8186,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-11T14:09:27Z +issue,8328,how can I fetch the result of output's Blobs_namet after I sum two numpy array via brew.sum,https://github.com/pytorch/pytorch/issues/8328,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-11T13:56:42Z +issue,7986,[Caffe2] place some nodes of the same network on GPU and others on CPU?,https://github.com/pytorch/pytorch/issues/7986,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-07T12:40:49Z +issue,6901,[Caffe2] [feature request] Dilations in grouped convolutons,https://github.com/pytorch/pytorch/issues/6901,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-06-01T13:25:42Z +issue,7945,[Caffe2]How to convert Caffe's mean file to Caffe2's ?,https://github.com/pytorch/pytorch/issues/7945,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-31T03:17:41Z +issue,7897,"[caffe2] build from source, cannot find my cudnn7.1",https://github.com/pytorch/pytorch/issues/7897,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-29T18:22:35Z +issue,6846,Jetson TX1 Caffe2 installation fails on latest JetPack release,https://github.com/pytorch/pytorch/issues/6846,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-22T08:40:50Z +issue,7729,[Caffe2][Caffe] Caffe to Caffe2 ParseError,https://github.com/pytorch/pytorch/issues/7729,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-21T13:57:17Z +issue,7493,Build from source with anaconda [caffe2],https://github.com/pytorch/pytorch/issues/7493,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-14T18:19:58Z +issue,7480,[caffe2] How to use multiple CPUs?,https://github.com/pytorch/pytorch/issues/7480,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-14T14:01:22Z +issue,7227,[Caffe2]Install problem,https://github.com/pytorch/pytorch/issues/7227,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-10T22:08:57Z +issue,6552,[Caffe2] - Run on different Cuda streams,https://github.com/pytorch/pytorch/issues/6552,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-07T17:30:51Z +issue,7133,[Caffe2 warpctc] How to use the offered warpctc?,https://github.com/pytorch/pytorch/issues/7133,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-03T03:51:39Z +issue,7179,[caffe2]when i do CMAKE,https://github.com/pytorch/pytorch/issues/7179,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-03T01:44:59Z +issue,6273,Multithreading Scaling Issue with MKL,https://github.com/pytorch/pytorch/issues/6273,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-05-01T02:28:44Z +issue,6902,[Caffe2] Flatten Layer in caffe2,https://github.com/pytorch/pytorch/issues/6902,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-04-26T17:30:44Z +issue,6875,[Bug] VS reports unresolved external symbol from caffe2 observer,https://github.com/pytorch/pytorch/issues/6875,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-04-26T16:25:52Z +issue,6783,[caffe2] IfOp,https://github.com/pytorch/pytorch/issues/6783,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-04-25T19:49:28Z +issue,6845,[caffe2] global average pool in caffe2,https://github.com/pytorch/pytorch/issues/6845,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-04-25T19:48:49Z +issue,6857,[caffe2] Understanding WorkFlow for Training and Testing,https://github.com/pytorch/pytorch/issues/6857,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-04-24T08:29:50Z +issue,6542,"【Train issue with caffe2 detectron】Aborted at 1523501813 (unix time) try ""date -d @1523501813"" if you are using GNU date *** PC: @ 0x7fa35103733a (unknown)",https://github.com/pytorch/pytorch/issues/6542,stale/low urgency,2,old update age or low-urgency/stale label,scan when time allows,2018-04-16T17:28:45Z