Publish Ropedia Xperience-10M derived artifacts

FOUNDATION_MODEL_PLAN.md

@@ -102,7 +102,7 @@ The foundation-model stage should add metrics beyond the current 12-task suite:
 4. Promote Cosmos 3 to the first world-model experiment if video/sensor
    preprocessing and storage fit.
 5. Promote OpenVLA/openpi/GR00T only after action targets are explicit and
-   retargeting artifacts are auditable.
+   retargeting artifacts are traceable.
 6. Update public cards only when a branch has real manifests, predictions,
    metrics, and qualitative examples.
 

PROJECT_README.md

@@ -31,7 +31,7 @@ For a first pass, use [`PROJECT_BRIEF.md`](PROJECT_BRIEF.md) or the
 machine-readable [`docs/data/project_brief.json`](docs/data/project_brief.json).
 They give the project shape in one page: what exists now, what the public
 sample can support, where the 12 tasks and baselines live, and what must happen
-before the 32-episode omni-model stage becomes a real held-out evaluation.
+before the multi-episode omni-model stage becomes a real held-out evaluation.
 
 | Reader goal | Best entry point |
 | --- | --- |
@@ -185,7 +185,7 @@ They give the current research state in one compact table:
 | Source alignment | Source facts, sample details, API-listing notes, and project coverage are consistent across repo, website, and HF cards |
 | Evaluation protocol | Verified generated protocol for windowing, split policy, leakage controls, and per-task metrics |
 | Website and HF mirrors | Verified by website reference reports, public project-surface reports, mirror parity, and live-publication checks; the public dashboard uses five top-level tabs plus subsection tabs for dataset, task-suite, method, result, and resource views |
-| Qwen3-Omni 32-episode pilot | Data-gated; prepared, with full metrics pending held-out evaluation |
+| Qwen3-Omni multi-episode pilot | Full-dataset access granted; 128-episode relay in progress, with full metrics pending completed staging and held-out evaluation |
 | Raw Xperience-10M data / full Qwen weights | Not redistributed |
 
 ## 90-Second Research Project Path
@@ -205,7 +205,7 @@ If you are reading the project cold, open these in order:
 | 9 | What is one model input? | [`windows.csv`](results/episode_task_suite/windows.csv), [`feature_manifest.json`](results/episode_task_suite/feature_manifest.json), [`available_modalities.json`](results/episode_task_suite/available_modalities.json) | The input is an aligned 8,546-d window vector with explicit feature-block boundaries. |
 | 10 | Are the task results backed by files? | [`summary_report.json`](results/episode_task_suite/summary_report.json), [`neural_mlp/`](results/episode_task_suite/neural_mlp/), [`docs/data/summary_metrics.json`](docs/data/summary_metrics.json) | Each task has minimal and neural-head evidence over the same window contracts. |
 | 11 | Is the website self-consistent? | [`docs/data/website_integrity.json`](docs/data/website_integrity.json), [`scripts/validate_website_integrity.py`](scripts/validate_website_integrity.py) | Local links, anchors, tab routing, JSON data, and referenced images are checked before publishing. |
-| 12 | What is still pending? | [`DATA_ACCESS_STATUS.md`](results/omni_finetune/DATA_ACCESS_STATUS.md), [`MULTI_EPISODE_ACCESS_STATUS.md`](results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md), [`scripts/omni/discover_xperience10m_sources.py`](scripts/omni/discover_xperience10m_sources.py) | The 32-episode Qwen3-Omni run is prepared; final model metrics require gated data and held-out evaluation. |
+| 12 | What is still pending? | [`DATA_ACCESS_STATUS.md`](results/omni_finetune/DATA_ACCESS_STATUS.md), [`MULTI_EPISODE_ACCESS_STATUS.md`](results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md), [`scripts/omni/discover_xperience10m_sources.py`](scripts/omni/discover_xperience10m_sources.py) | The multi-episode Qwen3-Omni run is prepared at the selection and relay level; final model metrics require completed staging, preprocessing, training, and held-out evaluation. |
 
 The machine-readable project packet is
 [`docs/data/project_packet.json`](docs/data/project_packet.json).
@@ -234,8 +234,8 @@ generated from committed metric artifacts. They define:
 - leakage controls for future labels, target feature blocks, caption/object
   labels, and train-only normalization,
 - current limitations, including cross-episode generalization,
-  audio-visual learning, pixel-depth reconstruction, and real 32-episode
-  Qwen3-Omni quality.
+  audio-visual learning, pixel-depth reconstruction, and real held-out
+  multi-episode Qwen3-Omni quality.
 
 ## Official Dataset Alignment
 
@@ -268,8 +268,8 @@ The public sample repo,
 is separately documented as `Xperience-10M-Sample` with sample metadata,
 `cc-by-nc-4.0` license, HOMIE Toolkit usage, and Rerun 0.29.0 `.rrd`
 visualization. This project preserves that distinction: the sample powers the
-current 5,821-frame task suite, while the full gated dataset remains the
-future source for held-out multi-episode training.
+current 5,821-frame task suite, while the full gated dataset is the source for
+the selected 128-episode held-out multi-episode relay now in progress.
 
 This repo's current verified subset is much smaller and intentionally explicit:
 
@@ -283,7 +283,7 @@ This repo's current verified subset is much smaller and intentionally explicit:
 The same alignment note also records what is outside the current implemented subset: real
 audio-visual learning, caption generation, pixel-depth estimation, SLAM
 estimation, neural rendering, policy learning, cross-episode generalization,
-and real 32-episode Qwen3-Omni model quality.
+and real held-out multi-episode Qwen3-Omni model quality.
 It also preserves the official responsible-use scope: the open-source
 dataset is limited in diversity and showcase/production quality, and it should
 not be used for identity recognition, re-identification, biometric profiling,
@@ -548,7 +548,7 @@ python scripts/train_all_modalities_model.py --workspace /path/to/workspace
 
 This repo includes a first Qwen3-Omni fine-tuning path over Xperience-10M. The
 current artifacts are setup-stage evidence, with held-out multi-episode metrics
-pending gated data access.
+pending completed staging, preprocessing, training, and evaluation.
 The useful distinction is:
 
 - direct Qwen3-Omni inputs: RGB/fisheye video, embedded MP4 audio, and language
@@ -558,9 +558,9 @@ The useful distinction is:
 
 The current scale-up artifacts show that the export, manifest, sensor-feature,
 LoRA, and evaluation scripts can run on the available sample episode. They do
-do not show a real 32-episode result. A real pilot requires at least 32 valid
+not show a real multi-episode result. A real pilot requires staged valid
 episodes, held-out episode splits, training metadata, predictions, metrics, and
-a run report.
+a run report; the current selected relay target is 128 episodes.
 
 ### Sample Count Decision
 
@@ -596,20 +596,21 @@ python scripts/omni/discover_xperience10m_sources.py \
 
 Current status in this repo:
 
-- local_valid_episodes: 1 (degraded-valid: annotation + fisheye_cam0.mp4)
-- local_complete_episodes: 0
-- ready_for_32_episode_pilot: false
-- planned 32-episode pilot: stratified across 32 top-level session UUIDs
-- full-dataset access: gated Xperience-10M approval is still pending
+- public_sample_valid_episodes: 1 (degraded-valid: annotation + fisheye_cam0.mp4)
+- gated_metadata_audit: 12,102 complete visible episodes across 802 complete sessions
+- selected_relay_plan: 128 metadata-balanced episodes, 96/16/16 train/val/test
+- selected_download_size: 277.71 GiB excluding `visualization.rrd`
+- ready_for_held_out_pilot: false until the selected episodes are fully staged and audited
+- full-dataset access: granted; raw multi-episode staging is in progress
 - source_discovery: `results/omni_finetune/source_discovery.json`
 - data_status: `results/omni_finetune/DATA_ACCESS_STATUS.md`
 - access_status: `results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md`
 
-Use this gate before scheduling any 32-episode full fine-tune run. The pilot
-should use stratified selection, not the first 32 paths in repository order.
-The current selection plan scans 64 top-level session UUIDs, filters for
-complete leaf episodes, excludes `visualization.rrd`, applies a `0.25 GB`
-minimum episode size, and selects 32 episodes from 32 different session UUIDs.
+Use this gate before scheduling any full fine-tune run. The pilot should use
+balanced held-out selection, not the first paths in repository order. The
+current 128-episode selection filters for complete leaf episodes, excludes
+`visualization.rrd`, balances episode-size bands, and preserves one selected
+episode per top-level session UUID.
 
 ### Uploading the pilot Qwen3-Omni LoRA
 
@@ -632,9 +633,9 @@ assuming one backbone solves every Xperience-10M objective.
 
 | Branch | Current role | When to use it |
 | --- | --- | --- |
-| Qwen3-Omni | First trainable multimodal LoRA pilot | Use for the 32-episode held-out baseline over video/audio/language plus sensor-bridge features. |
+| Qwen3-Omni | First trainable multimodal LoRA pilot | Use for the selected 128-episode held-out baseline over video/audio/language plus sensor-bridge features. |
 | Cosmos 3 | First world-model/action-generation branch | Use after data staging for future-window prediction, action-conditioned world modeling, and synthetic-data usefulness tests. |
-| GR00T | Humanoid/action-policy branch | Use after mocap/contact retargeting creates auditable humanoid action targets. |
+| GR00T | Humanoid/action-policy branch | Use after mocap/contact retargeting creates well-defined humanoid action targets. |
 | OpenVLA / openpi | Open VLA/policy baselines | Use after the project defines robot-compatible or action-token targets. |
 | Gemini Robotics | External reasoning reference | Use only for qualitative comparison or annotation support unless local trainable access exists. |
 

RESEARCH_ROADMAP.md

@@ -92,7 +92,7 @@ objective. The current decision is:
 - Cosmos 3 next for world modeling, action-conditioned future prediction, and
   synthetic-data experiments.
 - OpenVLA, openpi, GR00T, Octo, and SmolVLA-style policies after action-space
-  conversion and retargeting are auditable.
+  conversion and retargeting are traceable.
 - Gemini Robotics only as an external reasoning/reference surface unless local
   trainable access becomes available.
 

XPERIENCE10M_DATASET_CARD_ALIGNMENT.md

@@ -223,4 +223,4 @@ When describing Xperience-10M in this repo, keep these limitations visible:
 | Episode layout uses six MP4 streams and `annotation.hdf5` | Used by sample inspection and pilot-readiness scripts |
 | Audio exists in MP4 streams | Decoded into the current `audio_fisheye_cam0_aac` feature block |
 | 4D reconstruction/world modeling are intended research directions | Represented by proxy/diagnostic tasks only |
-| Real model quality requires held-out multi-episode evaluation | Pending the 32-episode pilot |
+| Real model quality requires held-out multi-episode evaluation | Pending selected multi-episode staging, training, and held-out evaluation |

docs/data/artifact_index.json

@@ -1,6 +1,6 @@
 {
   "title": "Ropedia Xperience-10M Task Suite Artifact Index",
-  "generated_at_utc": "2026-06-03T14:43:22+00:00",
+  "generated_at_utc": "2026-06-03T17:06:30+00:00",
   "status": "pass",
   "artifact_count": 72,
   "missing": [],
@@ -85,7 +85,7 @@
       "shows": "Defines the staged path from public-sample task development to multi-episode held-out evaluation and larger omni-model extensions.",
       "exists": true,
       "bytes": 6677,
-      "sha256": "5baa3ba1b96e0f1c70b8b9ea946aac55ec45d7543587bbfaf5120bef31533b17"
+      "sha256": "590702b8000bd143fdbcf65104b0cad835e6e4c79754c28c31727f1bdf43ae79"
     },
     {
       "id": "research_roadmap_json",
@@ -107,7 +107,7 @@
       "shows": "Defines the post-data-gate backbone choices: Qwen3-Omni first, Cosmos 3 for world modeling, and VLA/policy models after action-target conversion.",
       "exists": true,
       "bytes": 6538,
-      "sha256": "3e0794743ef949dd672b5263394da24428e0bee389e5b0157205b5832b4ad7f4"
+      "sha256": "15431b3ed368d660e777655d2e4a97f2870997f2ad38b451a73d631b5e431bcf"
     },
     {
       "id": "foundation_model_plan_json",
@@ -118,7 +118,7 @@
       "shows": "Machine-readable foundation-model selection matrix with source links, entry conditions, and evaluation additions.",
       "exists": true,
       "bytes": 8883,
-      "sha256": "9f15910ed056683c2fd918311b4a32cd46b699d7b7b7a3e1192bc3db6bc769ee"
+      "sha256": "0465df50071ab4c8e46acafdd2fa1653f23dacd33a1e1e1d65a571c82348ddbc"
     },
     {
       "id": "evidence_contract",
@@ -161,8 +161,8 @@
       "surface": "repo_hf",
       "shows": "Aligns public dataset wording with the official gated Xperience-10M card, public sample card, HF API metadata, and current project coverage.",
       "exists": true,
-      "bytes": 10719,
-      "sha256": "a0cc66e3a52d253dad4b8b19396cd3a8db14bf1835034c6c0aea6c5248fe4037"
+      "bytes": 10764,
+      "sha256": "8eaada6926194aeccf5af6742098f9be8aa611fc9bdf73cfa2df806327b89e9c"
     },
     {
       "id": "official_dataset_card_alignment_json",
@@ -172,8 +172,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable upstream dataset-card, sample-card, and HF API alignment facts for website and HF mirrors.",
       "exists": true,
-      "bytes": 7573,
-      "sha256": "a06402b0a9c23c8bacf0e79acd6bd07ba87e555f3c992d573eac17a3569eb2d7"
+      "bytes": 7585,
+      "sha256": "8bb8f22cf1cc8938f7dcfef15d11e3e544243bb56dc32c792313d224e91a8a2d"
     },
     {
       "id": "source_alignment",
@@ -195,7 +195,7 @@
       "shows": "Machine-readable source-alignment pass/fail check for repo, website, and HF surfaces.",
       "exists": true,
       "bytes": 4432,
-      "sha256": "ab8728f1f7539dd8e9d5a429e1bb01858d701412ab690ccd2f49f39ecbb645d9"
+      "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b"
     },
     {
       "id": "source_alignment_validator",
@@ -205,8 +205,8 @@
       "surface": "repo_hf",
       "shows": "Regenerates the source-alignment report from committed facts and public card text.",
       "exists": true,
-      "bytes": 15807,
-      "sha256": "66a9d0eebb4800be5d65f982ef13502b37522cd2a9eae0686c63efc121a52b0a"
+      "bytes": 15831,
+      "sha256": "6f48190ca77dd6ba9dd2f821d5ace7e33bd9b4d9d76adb7237cc51ddb88de239"
     },
     {
       "id": "hf_publisher",
@@ -426,7 +426,7 @@
       "shows": "Machine-readable release-check summary for validators, mirrors, and public project surfaces.",
       "exists": true,
       "bytes": 8147,
-      "sha256": "40c9cdc9d5afd15f8c69bda48a0eb36d81df14adff5555de54a481df1baed5ca"
+      "sha256": "1a6b09fdf86ad0ad6aef167de8325670d6643c5d68c35e9846ff85270dcfed7a"
     },
     {
       "id": "public_surface_qa",
@@ -448,7 +448,7 @@
       "volatile": true,
       "shows": "Machine-readable report for SEO/social metadata, accessible tab semantics, public links, project links, and reader-facing copy.",
       "exists": true,
-      "bytes": 5648,
+      "bytes": 5631,
       "hash_policy": "existence_and_size_only"
     },
     {
@@ -459,8 +459,8 @@
       "surface": "repo_hf",
       "shows": "Regenerates the public project-surface report before release.",
       "exists": true,
-      "bytes": 11894,
-      "sha256": "d0e86f45f2f23670e967c1a92024797a59106116ac9d948c35972512384f41de"
+      "bytes": 11872,
+      "sha256": "3ec688d5262d099b480bda674b858834d5d66b8085c0777a2435e58279f37d9d"
     },
     {
       "id": "task_surface_integrity",
@@ -609,7 +609,7 @@
       "volatile": true,
       "shows": "Confirms prepared GitHub/HF Space/artifact/model mirrors share the same critical data, figure, website HTML, and validator files.",
       "exists": true,
-      "bytes": 106509,
+      "bytes": 108606,
       "hash_policy": "existence_and_size_only"
     },
     {
@@ -621,7 +621,7 @@
       "volatile": true,
       "shows": "Confirms local website links, anchors, JSON data files, and referenced images resolve.",
       "exists": true,
-      "bytes": 14588,
+      "bytes": 14718,
       "hash_policy": "existence_and_size_only"
     },
     {
@@ -797,8 +797,8 @@
       "surface": "repo_hf",
       "shows": "Summarizes the data-access requirement before the 32-episode Qwen3-Omni pilot can run.",
       "exists": true,
-      "bytes": 1676,
-      "sha256": "c810259f82df1daebcaf6d03baa61c8c7dc2cc1a7d02f1b6d29776966f094a6f"
+      "bytes": 2954,
+      "sha256": "6ac14781581f06004bdc2fb0e2425c616419b738e6b858f8611bd19fab80741c"
     },
     {
       "id": "multi_episode_access_status",
@@ -808,8 +808,8 @@
       "surface": "repo_hf",
       "shows": "Documents the public multi-episode access status and 32-episode pilot selection.",
       "exists": true,
-      "bytes": 1480,
-      "sha256": "e5a3360a28ea5196cf477a648f119e9c936f9e8b427923d6f04b7589a0f63e4d"
+      "bytes": 2282,
+      "sha256": "e270cdfa4485458572a40f5c3c4bd598fe53e2b8de60af3ff1364398d96862e2"
     },
     {
       "id": "citation",

docs/data/foundation_model_plan.json

@@ -138,7 +138,7 @@
     {
       "step": 5,
       "name": "Policy branch",
-      "action": "Promote OpenVLA/openpi/GR00T after action target conversion and retargeting artifacts are auditable."
+      "action": "Promote OpenVLA/openpi/GR00T after action target conversion and retargeting artifacts are traceable."
     },
     {
       "step": 6,

docs/data/mirror_parity.json

@@ -1,6 +1,6 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-03T14:46:40+00:00",
+  "generated_at_utc": "2026-06-03T17:06:49+00:00",
   "hf_root": "hf_publish",
   "summary": {
     "group_count": 101,
@@ -72,26 +72,26 @@
         "path": "repo:docs/data/artifact_index.json",
         "exists": true,
         "bytes": 32378,
-        "sha256": "f144dfb742edf120b55e3603086b926bba67100d8c2a9f12e5ae1f856cb1fad9"
+        "sha256": "414aa6f5a7b16c0b1acfb2221560a52f58cd3fbcc016cf960825b95e26473b4b"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/artifact_index.json",
           "exists": true,
           "bytes": 32378,
-          "sha256": "f144dfb742edf120b55e3603086b926bba67100d8c2a9f12e5ae1f856cb1fad9"
+          "sha256": "414aa6f5a7b16c0b1acfb2221560a52f58cd3fbcc016cf960825b95e26473b4b"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/artifact_index.json",
           "exists": true,
           "bytes": 32378,
-          "sha256": "f144dfb742edf120b55e3603086b926bba67100d8c2a9f12e5ae1f856cb1fad9"
+          "sha256": "414aa6f5a7b16c0b1acfb2221560a52f58cd3fbcc016cf960825b95e26473b4b"
         },
         "hf_model": {
           "path": "hf_model:metrics/artifact_index.json",
           "exists": true,
           "bytes": 32378,
-          "sha256": "f144dfb742edf120b55e3603086b926bba67100d8c2a9f12e5ae1f856cb1fad9"
+          "sha256": "414aa6f5a7b16c0b1acfb2221560a52f58cd3fbcc016cf960825b95e26473b4b"
         }
       },
       "failures": []
@@ -227,26 +227,26 @@
         "path": "repo:docs/data/foundation_model_plan.json",
         "exists": true,
         "bytes": 8883,
-        "sha256": "9f15910ed056683c2fd918311b4a32cd46b699d7b7b7a3e1192bc3db6bc769ee"
+        "sha256": "0465df50071ab4c8e46acafdd2fa1653f23dacd33a1e1e1d65a571c82348ddbc"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/foundation_model_plan.json",
           "exists": true,
           "bytes": 8883,
-          "sha256": "9f15910ed056683c2fd918311b4a32cd46b699d7b7b7a3e1192bc3db6bc769ee"
+          "sha256": "0465df50071ab4c8e46acafdd2fa1653f23dacd33a1e1e1d65a571c82348ddbc"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/foundation_model_plan.json",
           "exists": true,
           "bytes": 8883,
-          "sha256": "9f15910ed056683c2fd918311b4a32cd46b699d7b7b7a3e1192bc3db6bc769ee"
+          "sha256": "0465df50071ab4c8e46acafdd2fa1653f23dacd33a1e1e1d65a571c82348ddbc"
         },
         "hf_model": {
           "path": "hf_model:metrics/foundation_model_plan.json",
           "exists": true,
           "bytes": 8883,
-          "sha256": "9f15910ed056683c2fd918311b4a32cd46b699d7b7b7a3e1192bc3db6bc769ee"
+          "sha256": "0465df50071ab4c8e46acafdd2fa1653f23dacd33a1e1e1d65a571c82348ddbc"
         }
       },
       "failures": []
@@ -444,26 +444,26 @@
         "path": "repo:docs/data/publication_audit.json",
         "exists": true,
         "bytes": 7289,
-        "sha256": "5cd9531d57ee355513eb498a73b71b8d0db8dd17f8589dfbf78bc9546d5a3874"
+        "sha256": "8bd9ff61443b39b48829f80d7d4ff2193a37402e22cf842023c469c3584480af"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/publication_audit.json",
           "exists": true,
           "bytes": 7289,
-          "sha256": "5cd9531d57ee355513eb498a73b71b8d0db8dd17f8589dfbf78bc9546d5a3874"
+          "sha256": "8bd9ff61443b39b48829f80d7d4ff2193a37402e22cf842023c469c3584480af"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/publication_audit.json",
           "exists": true,
           "bytes": 7289,
-          "sha256": "5cd9531d57ee355513eb498a73b71b8d0db8dd17f8589dfbf78bc9546d5a3874"
+          "sha256": "8bd9ff61443b39b48829f80d7d4ff2193a37402e22cf842023c469c3584480af"
         },
         "hf_model": {
           "path": "hf_model:metrics/publication_audit.json",
           "exists": true,
           "bytes": 7289,
-          "sha256": "5cd9531d57ee355513eb498a73b71b8d0db8dd17f8589dfbf78bc9546d5a3874"
+          "sha256": "8bd9ff61443b39b48829f80d7d4ff2193a37402e22cf842023c469c3584480af"
         }
       },
       "failures": []
@@ -474,27 +474,27 @@
       "local": {
         "path": "repo:docs/data/public_surface_qa.json",
         "exists": true,
-        "bytes": 5648,
-        "sha256": "69b4da08a2603d9d3699125510c9bbd9982a7bf0f2f3236ee5472f60dfe04f5e"
+        "bytes": 5631,
+        "sha256": "04327d22b4210afef123a29071ec7d686854efac6ca6c1b301b58fa9a04a2624"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/public_surface_qa.json",
           "exists": true,
-          "bytes": 5648,
-          "sha256": "69b4da08a2603d9d3699125510c9bbd9982a7bf0f2f3236ee5472f60dfe04f5e"
+          "bytes": 5631,
+          "sha256": "04327d22b4210afef123a29071ec7d686854efac6ca6c1b301b58fa9a04a2624"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/public_surface_qa.json",
           "exists": true,
-          "bytes": 5648,
-          "sha256": "69b4da08a2603d9d3699125510c9bbd9982a7bf0f2f3236ee5472f60dfe04f5e"
+          "bytes": 5631,
+          "sha256": "04327d22b4210afef123a29071ec7d686854efac6ca6c1b301b58fa9a04a2624"
         },
         "hf_model": {
           "path": "hf_model:metrics/public_surface_qa.json",
           "exists": true,
-          "bytes": 5648,
-          "sha256": "69b4da08a2603d9d3699125510c9bbd9982a7bf0f2f3236ee5472f60dfe04f5e"
+          "bytes": 5631,
+          "sha256": "04327d22b4210afef123a29071ec7d686854efac6ca6c1b301b58fa9a04a2624"
         }
       },
       "failures": []
@@ -506,26 +506,26 @@
         "path": "repo:docs/data/quality_gates.json",
         "exists": true,
         "bytes": 8147,
-        "sha256": "40c9cdc9d5afd15f8c69bda48a0eb36d81df14adff5555de54a481df1baed5ca"
+        "sha256": "1a6b09fdf86ad0ad6aef167de8325670d6643c5d68c35e9846ff85270dcfed7a"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/quality_gates.json",
           "exists": true,
           "bytes": 8147,
-          "sha256": "40c9cdc9d5afd15f8c69bda48a0eb36d81df14adff5555de54a481df1baed5ca"
+          "sha256": "1a6b09fdf86ad0ad6aef167de8325670d6643c5d68c35e9846ff85270dcfed7a"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/quality_gates.json",
           "exists": true,
           "bytes": 8147,
-          "sha256": "40c9cdc9d5afd15f8c69bda48a0eb36d81df14adff5555de54a481df1baed5ca"
+          "sha256": "1a6b09fdf86ad0ad6aef167de8325670d6643c5d68c35e9846ff85270dcfed7a"
         },
         "hf_model": {
           "path": "hf_model:metrics/quality_gates.json",
           "exists": true,
           "bytes": 8147,
-          "sha256": "40c9cdc9d5afd15f8c69bda48a0eb36d81df14adff5555de54a481df1baed5ca"
+          "sha256": "1a6b09fdf86ad0ad6aef167de8325670d6643c5d68c35e9846ff85270dcfed7a"
         }
       },
       "failures": []
@@ -630,26 +630,26 @@
         "path": "repo:docs/data/research_roadmap_interactive.json",
         "exists": true,
         "bytes": 131526,
-        "sha256": "747d62be18607d718a0e9226aeb521ab110c2892dfcaf07b760a76ffdfe66679"
+        "sha256": "12b64fa9108066664839c8bc0b02e3c5ddb6d6b10311fb96f8bfc265f25d3cc0"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/research_roadmap_interactive.json",
           "exists": true,
           "bytes": 131526,
-          "sha256": "747d62be18607d718a0e9226aeb521ab110c2892dfcaf07b760a76ffdfe66679"
+          "sha256": "12b64fa9108066664839c8bc0b02e3c5ddb6d6b10311fb96f8bfc265f25d3cc0"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/research_roadmap_interactive.json",
           "exists": true,
           "bytes": 131526,
-          "sha256": "747d62be18607d718a0e9226aeb521ab110c2892dfcaf07b760a76ffdfe66679"
+          "sha256": "12b64fa9108066664839c8bc0b02e3c5ddb6d6b10311fb96f8bfc265f25d3cc0"
         },
         "hf_model": {
           "path": "hf_model:metrics/research_roadmap_interactive.json",
           "exists": true,
           "bytes": 131526,
-          "sha256": "747d62be18607d718a0e9226aeb521ab110c2892dfcaf07b760a76ffdfe66679"
+          "sha256": "12b64fa9108066664839c8bc0b02e3c5ddb6d6b10311fb96f8bfc265f25d3cc0"
         }
       },
       "failures": []
@@ -816,26 +816,26 @@
         "path": "repo:docs/data/source_alignment_audit.json",
         "exists": true,
         "bytes": 4432,
-        "sha256": "db0758394305b8b5df8baa518d1d90429c5ae8430eaa6464ed42d644a531f745"
+        "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
-          "sha256": "db0758394305b8b5df8baa518d1d90429c5ae8430eaa6464ed42d644a531f745"
+          "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
-          "sha256": "db0758394305b8b5df8baa518d1d90429c5ae8430eaa6464ed42d644a531f745"
+          "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b"
         },
         "hf_model": {
           "path": "hf_model:metrics/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
-          "sha256": "db0758394305b8b5df8baa518d1d90429c5ae8430eaa6464ed42d644a531f745"
+          "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b"
         }
       },
       "failures": []
@@ -940,26 +940,26 @@
         "path": "repo:docs/data/website_integrity.json",
         "exists": true,
         "bytes": 14718,
-        "sha256": "5e791965ff07ebacc6c139cf3d452bbf6c2ce87c5755e7ec758d6b819dd2611b"
+        "sha256": "3cb852491d4b7998cd1c2b04a644c9d97eaddf4b8bd29bb00f971ed2e880f4c5"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/website_integrity.json",
           "exists": true,
           "bytes": 14718,
-          "sha256": "5e791965ff07ebacc6c139cf3d452bbf6c2ce87c5755e7ec758d6b819dd2611b"
+          "sha256": "3cb852491d4b7998cd1c2b04a644c9d97eaddf4b8bd29bb00f971ed2e880f4c5"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/website_integrity.json",
           "exists": true,
           "bytes": 14718,
-          "sha256": "5e791965ff07ebacc6c139cf3d452bbf6c2ce87c5755e7ec758d6b819dd2611b"
+          "sha256": "3cb852491d4b7998cd1c2b04a644c9d97eaddf4b8bd29bb00f971ed2e880f4c5"
         },
         "hf_model": {
           "path": "hf_model:metrics/website_integrity.json",
           "exists": true,
           "bytes": 14718,
-          "sha256": "5e791965ff07ebacc6c139cf3d452bbf6c2ce87c5755e7ec758d6b819dd2611b"
+          "sha256": "3cb852491d4b7998cd1c2b04a644c9d97eaddf4b8bd29bb00f971ed2e880f4c5"
         }
       },
       "failures": []
@@ -970,27 +970,27 @@
       "local": {
         "path": "repo:docs/data/xperience10m_dataset_card_alignment.json",
         "exists": true,
-        "bytes": 7573,
-        "sha256": "a06402b0a9c23c8bacf0e79acd6bd07ba87e555f3c992d573eac17a3569eb2d7"
+        "bytes": 7585,
+        "sha256": "8bb8f22cf1cc8938f7dcfef15d11e3e544243bb56dc32c792313d224e91a8a2d"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/xperience10m_dataset_card_alignment.json",
           "exists": true,
-          "bytes": 7573,
-          "sha256": "a06402b0a9c23c8bacf0e79acd6bd07ba87e555f3c992d573eac17a3569eb2d7"
+          "bytes": 7585,
+          "sha256": "8bb8f22cf1cc8938f7dcfef15d11e3e544243bb56dc32c792313d224e91a8a2d"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/xperience10m_dataset_card_alignment.json",
           "exists": true,
-          "bytes": 7573,
-          "sha256": "a06402b0a9c23c8bacf0e79acd6bd07ba87e555f3c992d573eac17a3569eb2d7"
+          "bytes": 7585,
+          "sha256": "8bb8f22cf1cc8938f7dcfef15d11e3e544243bb56dc32c792313d224e91a8a2d"
         },
         "hf_model": {
           "path": "hf_model:metrics/xperience10m_dataset_card_alignment.json",
           "exists": true,
-          "bytes": 7573,
-          "sha256": "a06402b0a9c23c8bacf0e79acd6bd07ba87e555f3c992d573eac17a3569eb2d7"
+          "bytes": 7585,
+          "sha256": "8bb8f22cf1cc8938f7dcfef15d11e3e544243bb56dc32c792313d224e91a8a2d"
         }
       },
       "failures": []
@@ -1817,21 +1817,21 @@
       "local": {
         "path": "repo:scripts/build_public_surface_qa.py",
         "exists": true,
-        "bytes": 11894,
-        "sha256": "d0e86f45f2f23670e967c1a92024797a59106116ac9d948c35972512384f41de"
+        "bytes": 11872,
+        "sha256": "3ec688d5262d099b480bda674b858834d5d66b8085c0777a2435e58279f37d9d"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/build_public_surface_qa.py",
           "exists": true,
-          "bytes": 11894,
-          "sha256": "d0e86f45f2f23670e967c1a92024797a59106116ac9d948c35972512384f41de"
+          "bytes": 11872,
+          "sha256": "3ec688d5262d099b480bda674b858834d5d66b8085c0777a2435e58279f37d9d"
         },
         "hf_model": {
           "path": "hf_model:scripts/build_public_surface_qa.py",
           "exists": true,
-          "bytes": 11894,
-          "sha256": "d0e86f45f2f23670e967c1a92024797a59106116ac9d948c35972512384f41de"
+          "bytes": 11872,
+          "sha256": "3ec688d5262d099b480bda674b858834d5d66b8085c0777a2435e58279f37d9d"
         }
       },
       "failures": []
@@ -2017,21 +2017,21 @@
       "local": {
         "path": "repo:scripts/validate_publication_package.py",
         "exists": true,
-        "bytes": 19247,
-        "sha256": "55a9c51d57766d54f9eb884e06d644fd9f3d6e358b80c06e7f515ab764cc2158"
+        "bytes": 19295,
+        "sha256": "bba7c8d205b1cf2dc9b2252333532bb98acb5e6a0dc5a59c3e2d9ea4066e7687"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/validate_publication_package.py",
           "exists": true,
-          "bytes": 19247,
-          "sha256": "55a9c51d57766d54f9eb884e06d644fd9f3d6e358b80c06e7f515ab764cc2158"
+          "bytes": 19295,
+          "sha256": "bba7c8d205b1cf2dc9b2252333532bb98acb5e6a0dc5a59c3e2d9ea4066e7687"
         },
         "hf_model": {
           "path": "hf_model:scripts/validate_publication_package.py",
           "exists": true,
-          "bytes": 19247,
-          "sha256": "55a9c51d57766d54f9eb884e06d644fd9f3d6e358b80c06e7f515ab764cc2158"
+          "bytes": 19295,
+          "sha256": "bba7c8d205b1cf2dc9b2252333532bb98acb5e6a0dc5a59c3e2d9ea4066e7687"
         }
       },
       "failures": []
@@ -2067,21 +2067,21 @@
       "local": {
         "path": "repo:scripts/validate_source_alignment.py",
         "exists": true,
-        "bytes": 15807,
-        "sha256": "66a9d0eebb4800be5d65f982ef13502b37522cd2a9eae0686c63efc121a52b0a"
+        "bytes": 15831,
+        "sha256": "6f48190ca77dd6ba9dd2f821d5ace7e33bd9b4d9d76adb7237cc51ddb88de239"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/validate_source_alignment.py",
           "exists": true,
-          "bytes": 15807,
-          "sha256": "66a9d0eebb4800be5d65f982ef13502b37522cd2a9eae0686c63efc121a52b0a"
+          "bytes": 15831,
+          "sha256": "6f48190ca77dd6ba9dd2f821d5ace7e33bd9b4d9d76adb7237cc51ddb88de239"
         },
         "hf_model": {
           "path": "hf_model:scripts/validate_source_alignment.py",
           "exists": true,
-          "bytes": 15807,
-          "sha256": "66a9d0eebb4800be5d65f982ef13502b37522cd2a9eae0686c63efc121a52b0a"
+          "bytes": 15831,
+          "sha256": "6f48190ca77dd6ba9dd2f821d5ace7e33bd9b4d9d76adb7237cc51ddb88de239"
         }
       },
       "failures": []
@@ -2217,21 +2217,21 @@
       "local": {
         "path": "repo:docs/index.html",
         "exists": true,
-        "bytes": 170152,
-        "sha256": "db907d614fa5f31ee9e4c7dbb5bc3c319561cec8645b5afe4a4fdd8864c6593b"
+        "bytes": 170464,
+        "sha256": "f85f66d3551bf641c0f21a2a7581c3cc9df277c3f22403c8a361346a18c11385"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:index.html",
           "exists": true,
-          "bytes": 170152,
-          "sha256": "db907d614fa5f31ee9e4c7dbb5bc3c319561cec8645b5afe4a4fdd8864c6593b"
+          "bytes": 170464,
+          "sha256": "f85f66d3551bf641c0f21a2a7581c3cc9df277c3f22403c8a361346a18c11385"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/index.html",
           "exists": true,
-          "bytes": 170152,
-          "sha256": "db907d614fa5f31ee9e4c7dbb5bc3c319561cec8645b5afe4a4fdd8864c6593b"
+          "bytes": 170464,
+          "sha256": "f85f66d3551bf641c0f21a2a7581c3cc9df277c3f22403c8a361346a18c11385"
         }
       },
       "failures": []
@@ -2845,26 +2845,26 @@
         "path": "repo:FOUNDATION_MODEL_PLAN.md",
         "exists": true,
         "bytes": 6538,
-        "sha256": "3e0794743ef949dd672b5263394da24428e0bee389e5b0157205b5832b4ad7f4"
+        "sha256": "15431b3ed368d660e777655d2e4a97f2870997f2ad38b451a73d631b5e431bcf"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:FOUNDATION_MODEL_PLAN.md",
           "exists": true,
           "bytes": 6538,
-          "sha256": "3e0794743ef949dd672b5263394da24428e0bee389e5b0157205b5832b4ad7f4"
+          "sha256": "15431b3ed368d660e777655d2e4a97f2870997f2ad38b451a73d631b5e431bcf"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:FOUNDATION_MODEL_PLAN.md",
           "exists": true,
           "bytes": 6538,
-          "sha256": "3e0794743ef949dd672b5263394da24428e0bee389e5b0157205b5832b4ad7f4"
+          "sha256": "15431b3ed368d660e777655d2e4a97f2870997f2ad38b451a73d631b5e431bcf"
         },
         "hf_model": {
           "path": "hf_model:FOUNDATION_MODEL_PLAN.md",
           "exists": true,
           "bytes": 6538,
-          "sha256": "3e0794743ef949dd672b5263394da24428e0bee389e5b0157205b5832b4ad7f4"
+          "sha256": "15431b3ed368d660e777655d2e4a97f2870997f2ad38b451a73d631b5e431bcf"
         }
       },
       "failures": []
@@ -2938,26 +2938,26 @@
         "path": "repo:RESEARCH_ROADMAP.md",
         "exists": true,
         "bytes": 6677,
-        "sha256": "5baa3ba1b96e0f1c70b8b9ea946aac55ec45d7543587bbfaf5120bef31533b17"
+        "sha256": "590702b8000bd143fdbcf65104b0cad835e6e4c79754c28c31727f1bdf43ae79"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:RESEARCH_ROADMAP.md",
           "exists": true,
           "bytes": 6677,
-          "sha256": "5baa3ba1b96e0f1c70b8b9ea946aac55ec45d7543587bbfaf5120bef31533b17"
+          "sha256": "590702b8000bd143fdbcf65104b0cad835e6e4c79754c28c31727f1bdf43ae79"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:RESEARCH_ROADMAP.md",
           "exists": true,
           "bytes": 6677,
-          "sha256": "5baa3ba1b96e0f1c70b8b9ea946aac55ec45d7543587bbfaf5120bef31533b17"
+          "sha256": "590702b8000bd143fdbcf65104b0cad835e6e4c79754c28c31727f1bdf43ae79"
         },
         "hf_model": {
           "path": "hf_model:RESEARCH_ROADMAP.md",
           "exists": true,
           "bytes": 6677,
-          "sha256": "5baa3ba1b96e0f1c70b8b9ea946aac55ec45d7543587bbfaf5120bef31533b17"
+          "sha256": "590702b8000bd143fdbcf65104b0cad835e6e4c79754c28c31727f1bdf43ae79"
         }
       },
       "failures": []
@@ -3092,27 +3092,27 @@
       "local": {
         "path": "repo:XPERIENCE10M_DATASET_CARD_ALIGNMENT.md",
         "exists": true,
-        "bytes": 10719,
-        "sha256": "a0cc66e3a52d253dad4b8b19396cd3a8db14bf1835034c6c0aea6c5248fe4037"
+        "bytes": 10764,
+        "sha256": "8eaada6926194aeccf5af6742098f9be8aa611fc9bdf73cfa2df806327b89e9c"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:XPERIENCE10M_DATASET_CARD_ALIGNMENT.md",
           "exists": true,
-          "bytes": 10719,
-          "sha256": "a0cc66e3a52d253dad4b8b19396cd3a8db14bf1835034c6c0aea6c5248fe4037"
+          "bytes": 10764,
+          "sha256": "8eaada6926194aeccf5af6742098f9be8aa611fc9bdf73cfa2df806327b89e9c"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:XPERIENCE10M_DATASET_CARD_ALIGNMENT.md",
           "exists": true,
-          "bytes": 10719,
-          "sha256": "a0cc66e3a52d253dad4b8b19396cd3a8db14bf1835034c6c0aea6c5248fe4037"
+          "bytes": 10764,
+          "sha256": "8eaada6926194aeccf5af6742098f9be8aa611fc9bdf73cfa2df806327b89e9c"
         },
         "hf_model": {
           "path": "hf_model:XPERIENCE10M_DATASET_CARD_ALIGNMENT.md",
           "exists": true,
-          "bytes": 10719,
-          "sha256": "a0cc66e3a52d253dad4b8b19396cd3a8db14bf1835034c6c0aea6c5248fe4037"
+          "bytes": 10764,
+          "sha256": "8eaada6926194aeccf5af6742098f9be8aa611fc9bdf73cfa2df806327b89e9c"
         }
       },
       "failures": []

docs/data/public_surface_qa.json

@@ -1,7 +1,7 @@
 {
   "title": "Ropedia Xperience-10M Public Project Surface",
   "status": "pass",
-  "generated_at_utc": "2026-06-03T14:17:21+00:00",
+  "generated_at_utc": "2026-06-03T17:06:22+00:00",
   "scope": "Repo README, GitHub Pages HTML, Hugging Face Space card, artifact dataset card, and model card.",
   "checks": [
     {
@@ -18,7 +18,7 @@
         "website_integrity": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-03T13:07:43+00:00"
+          "generated_at_utc": "2026-06-03T17:02:57+00:00"
         },
         "rendered_site_check": {
           "exists": true,
@@ -28,27 +28,27 @@
         "task_surface_integrity": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-03T13:08:46+00:00"
+          "generated_at_utc": "2026-06-03T14:43:40+00:00"
         },
         "source_alignment": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-03T13:08:46+00:00"
+          "generated_at_utc": "2026-06-03T17:02:07+00:00"
         },
         "scale_up_status": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-03T13:07:42+00:00"
+          "generated_at_utc": "2026-06-03T14:43:40+00:00"
         },
         "publication_package": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-03T13:54:27+00:00"
+          "generated_at_utc": "2026-06-03T17:04:31+00:00"
         },
         "mirror_parity": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-03T13:54:27+00:00"
+          "generated_at_utc": "2026-06-03T17:04:48+00:00"
         },
         "live_publication": {
           "exists": true,
@@ -101,10 +101,10 @@
       "reason": "Public copy should consistently present the project as Ropedia Xperience-10M, with the Qwen3-Omni scale-up status.",
       "marker_counts": {
         "Ropedia Xperience-10M Task Suite": 14,
-        "Xperience-10M": 93,
-        "12-task": 20,
-        "Qwen3-Omni": 37,
-        "one public Xperience-10M sample episode": 1
+        "Xperience-10M": 128,
+        "12-task": 33,
+        "Qwen3-Omni": 81,
+        "128-episode relay": 6
       }
     },
     {
@@ -112,11 +112,11 @@
       "status": "pass",
       "reason": "Public cards should link the repo, Space, artifacts, model baselines, upstream dataset, and Ropedia dataset page.",
       "marker_counts": {
-        "https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite": 63,
-        "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite": 5,
+        "https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite": 66,
+        "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite": 8,
         "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts": 4,
-        "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines": 3,
-        "https://huggingface.co/datasets/ropedia-ai/xperience-10m": 16,
+        "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines": 4,
+        "https://huggingface.co/datasets/ropedia-ai/xperience-10m": 26,
         "https://ropedia.com/dataset": 4
       }
     },
@@ -125,14 +125,14 @@
       "status": "pass",
       "reason": "Readers should be able to find website reference, release package, mirror, and project-surface files from public copy.",
       "marker_counts": {
-        "data/project_brief.json": 7,
-        "data/website_integrity.json": 7,
-        "data/rendered_site_check.json": 5,
-        "data/task_surface_integrity.json": 10,
-        "data/publication_audit.json": 8,
-        "data/mirror_parity.json": 6,
-        "data/public_surface_qa.json": 7,
-        "data/research_roadmap.json": 10
+        "data/project_brief.json": 9,
+        "data/website_integrity.json": 13,
+        "data/rendered_site_check.json": 8,
+        "data/task_surface_integrity.json": 23,
+        "data/publication_audit.json": 18,
+        "data/mirror_parity.json": 12,
+        "data/public_surface_qa.json": 12,
+        "data/research_roadmap.json": 23
       }
     },
     {

docs/data/publication_audit.json

@@ -1,6 +1,6 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-03T14:46:16+00:00",
+  "generated_at_utc": "2026-06-03T17:06:24+00:00",
   "checks": [
     {
       "name": "required_publication_assets_present",
@@ -182,8 +182,8 @@
     "github_repo": {
       "root": "repo",
       "exists": true,
-      "file_count": 365,
-      "text_file_count": 300,
+      "file_count": 379,
+      "text_file_count": 314,
       "largest_file": {
         "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
         "bytes": 55702978
@@ -193,8 +193,8 @@
     "hf_space_bundle": {
       "root": "hf_publish/space",
       "exists": true,
-      "file_count": 313,
-      "text_file_count": 248,
+      "file_count": 322,
+      "text_file_count": 257,
       "largest_file": {
         "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
         "bytes": 55702978
@@ -204,8 +204,8 @@
     "hf_artifact_bundle": {
       "root": "hf_publish/artifacts",
       "exists": true,
-      "file_count": 395,
-      "text_file_count": 310,
+      "file_count": 409,
+      "text_file_count": 324,
       "largest_file": {
         "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
         "bytes": 55702978
@@ -215,8 +215,8 @@
     "hf_model_bundle": {
       "root": "hf_publish/model",
       "exists": true,
-      "file_count": 620,
-      "text_file_count": 498,
+      "file_count": 634,
+      "text_file_count": 512,
       "largest_file": {
         "path": "artifacts/episode_task_suite/modality_reconstruction/predictions.npz",
         "bytes": 55702978

docs/data/quality_gates.json

@@ -1,7 +1,7 @@
 {
   "title": "Ropedia Xperience-10M Release Checks",
   "status": "pass",
-  "generated_at_utc": "2026-06-03T13:04:23+00:00",
+  "generated_at_utc": "2026-06-03T17:06:30+00:00",
   "rule": "A release is current when the automated reports pass and the live GitHub/Hugging Face mirrors are verified after publishing.",
   "automated_gates": [
     {

docs/data/research_roadmap_interactive.json

@@ -1913,7 +1913,7 @@
         "step": 4
       },
       {
-        "action": "Promote OpenVLA/openpi/GR00T after action target conversion and retargeting artifacts are auditable.",
+        "action": "Promote OpenVLA/openpi/GR00T after action target conversion and retargeting artifacts are traceable.",
         "name": "Policy branch",
         "step": 5
       },

docs/data/source_alignment_audit.json

@@ -1,7 +1,7 @@
 {
   "title": "Ropedia Xperience-10M Source Alignment Note",
   "status": "pass",
-  "generated_at_utc": "2026-06-03T14:46:14+00:00",
+  "generated_at_utc": "2026-06-03T17:02:07+00:00",
   "alignment_json": "docs/data/xperience10m_dataset_card_alignment.json",
   "alignment_summary": {
     "full_dataset_repo": "ropedia-ai/xperience-10m",

docs/data/website_integrity.json

@@ -1,6 +1,6 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-03T14:43:40+00:00",
+  "generated_at_utc": "2026-06-03T17:06:23+00:00",
   "docs_root": "docs",
   "site_base": "/ropedia-xperience-10m-task-suite/",
   "summary": {
@@ -75,7 +75,7 @@
       "status": "pass",
       "reason": "The project overview should appear before the deeper progress ledger.",
       "overview_index": 66066,
-      "evidence_index": 81829
+      "evidence_index": 81914
     },
     {
       "name": "project_status_links_json",
@@ -150,8 +150,8 @@
       "status": "pass",
       "reason": "The evaluation protocol should appear before the deeper evidence ledger.",
       "overview_index": 66066,
-      "protocol_index": 78688,
-      "evidence_index": 81829
+      "protocol_index": 78750,
+      "evidence_index": 81914
     },
     {
       "name": "evaluation_protocol_links_json",
@@ -284,7 +284,7 @@
     },
     {
       "path": "data/mirror_parity.json",
-      "bytes": 106509,
+      "bytes": 108606,
       "top_level_type": "dict"
     },
     {
@@ -314,7 +314,7 @@
     },
     {
       "path": "data/public_surface_qa.json",
-      "bytes": 5648,
+      "bytes": 5631,
       "top_level_type": "dict"
     },
     {
@@ -394,12 +394,12 @@
     },
     {
       "path": "data/website_integrity.json",
-      "bytes": 14588,
+      "bytes": 14719,
       "top_level_type": "dict"
     },
     {
       "path": "data/xperience10m_dataset_card_alignment.json",
-      "bytes": 7573,
+      "bytes": 7585,
       "top_level_type": "dict"
     }
   ],

docs/data/xperience10m_dataset_card_alignment.json

@@ -194,7 +194,7 @@
       "neural rendering",
       "policy learning",
       "cross-episode generalization",
-      "real 32-episode Qwen3-Omni model quality"
+      "real held-out multi-episode Qwen3-Omni model quality"
     ]
   },
   "responsible_use_boundary": [

docs/index.html

@@ -2265,12 +2265,12 @@
             </div>
           </article>
           <article class="snapshot-card gated">
-            <span class="status-pill">data-gated</span>
+            <span class="status-pill">staging</span>
             <h3>Omni-model scale-up path</h3>
-            <p>The 32-episode LoRA path is prepared; full training results require gated data access, held-out splits, training, and evaluation.</p>
+            <p>Full-dataset access is granted, a 128-episode relay is in progress, and full training results require completed staging, held-out splits, training, and evaluation.</p>
             <div class="snapshot-meta">
-              <span>current stage <strong>setup checked</strong></span>
-              <span>target gate <strong>32 episodes</strong></span>
+              <span>current stage <strong>relay started</strong></span>
+              <span>selected set <strong>128 episodes</strong></span>
               <span>held-out eval <strong>pending</strong></span>
             </div>
           </article>
@@ -2336,7 +2336,7 @@
             <h3>Foundation-Model Selection Matrix</h3>
             <p>Keep Qwen3-Omni as the first trainable held-out pilot, add Cosmos 3 for world modeling, and stage policy candidates after action targets are explicit.</p>
             <div class="roadmap-meta">
-              <strong>Entry</strong><p>32-episode data gate or a 3-8 episode preprocessing dry run.</p>
+              <strong>Entry</strong><p>Completed 128-episode staging or a smaller 3-8 episode preprocessing dry run.</p>
               <strong>Evidence</strong><p>Foundation model plan, source links, model-specific entry conditions, and evaluation additions.</p>
             </div>
           </article>
@@ -2345,7 +2345,7 @@
             <h3>64-128 Episode Robustness Run</h3>
             <p>Test whether pilot conclusions survive broader sessions, missing modalities, and stronger ablations.</p>
             <div class="roadmap-meta">
-              <strong>Entry</strong><p>32-episode pilot trains and evaluates cleanly.</p>
+              <strong>Entry</strong><p>Selected multi-episode pilot trains and evaluates cleanly.</p>
               <strong>Evidence</strong><p>Metrics by session, task, modality, ablation, and failure type.</p>
             </div>
           </article>
@@ -2384,7 +2384,7 @@
           <article class="artifact"><h3>Leakage controls</h3><p>Scalers fit on train windows only; future labels, target feature blocks, caption/object labels, and contact labels stay on the target side unless explicitly queried.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/scripts/build_evaluation_protocol.py">builder script</a></article>
           <article class="artifact"><h3>Audio ablation</h3><p>The current AAC block and a 588-d raw log-mel replacement are evaluated across all 12 task contracts under the same chronological split.</p><a href="data/audio_ablation_summary.json">audio summary</a></article>
           <article class="artifact"><h3>Foundation branch selection</h3><p>Qwen3-Omni is the first trainable baseline, Cosmos 3 becomes the world-model branch, and policy models wait for explicit action targets.</p><a href="data/foundation_model_plan.json">backbone plan</a></article>
-          <article class="artifact"><h3>Next evaluation stage</h3><p>This public-sample run covers single-episode task development. Cross-episode generalization, audio-visual learning, world modeling, policy targets, and full 32-episode Qwen3-Omni training move to the multi-episode stage.</p><a href="data/scope_claims_audit.json">pilot status</a></article>
+          <article class="artifact"><h3>Next evaluation stage</h3><p>This public-sample run covers single-episode task development. Cross-episode generalization, audio-visual learning, world modeling, policy targets, and held-out Qwen3-Omni training move to the multi-episode stage after selected data is staged.</p><a href="data/scope_claims_audit.json">pilot status</a></article>
           <article class="artifact"><h3>Scale-up requirement</h3><p>The Omni pilot requires at least 32 valid episodes, held-out episode splits, no train/test episode leakage, training metadata, predictions, metrics, and a run report.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">data status</a></article>
         </div>
       </div>
@@ -2444,9 +2444,9 @@
             </div>
           </article>
           <article class="evidence-card">
-            <span class="status-pill">data-gated</span>
+            <span class="status-pill">staging</span>
             <h3>Qwen3-Omni pilot setup</h3>
-            <p>The current Qwen3-Omni artifacts use one episode and 128 train windows. The 32-episode evaluation is still pending.</p>
+            <p>The current Qwen3-Omni artifacts use one episode and 128 train windows. A 128-episode selected relay is in progress for held-out evaluation.</p>
             <div class="evidence-links">
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/EVIDENCE_CONTRACT.md">evidence contract</a>
               <a href="data/evidence_contract.json">machine JSON</a>
@@ -2581,7 +2581,7 @@
           <article class="reading-card">
             <span class="step-index">04</span>
             <h3>Check the scale-up gate</h3>
-            <p>The multi-episode Qwen3-Omni path is prepared. The 32-episode result will be added after the data gate and held-out evaluation pass.</p>
+            <p>The multi-episode Qwen3-Omni path is prepared. The selected 128-episode result will be added after staging, preprocessing, training, and held-out evaluation pass.</p>
             <div class="reading-links">
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">data status</a>
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md">access status</a>
@@ -2591,7 +2591,7 @@
         </div>
         <div class="boundary-strip">
           <div class="boundary-item"><strong>Verified now</strong><span>One public episode, 5,821 frames, 1,161 windows, 8,546 current features, 12 minimal heads, 12 neural heads, and 4 direction-extension probes.</span></div>
-          <div class="boundary-item"><strong>Next: multi-episode</strong><span>A 32-episode held-out Qwen3-Omni LoRA pilot is gated on Xperience-10M access and must pass manifest, training, and evaluation checks.</span></div>
+          <div class="boundary-item"><strong>Next: multi-episode</strong><span>A selected 128-episode held-out Qwen3-Omni LoRA pilot is being staged and must pass manifest, training, and evaluation checks before metrics are reported.</span></div>
           <div class="boundary-item"><strong>Not redistributed</strong><span>Raw videos, raw annotations, full Qwen weights, and private gated Xperience-10M data are not included in the public repo or HF bundles.</span></div>
         </div>
       </div>
@@ -2614,7 +2614,7 @@
           <article class="artifact"><h3>Current project subset</h3><p>One public sample episode, 5,821 frames, 1,161 windows, 8,546 current features including AAC audio, and no raw-data redistribution.</p><a href="data/modality_atlas.json">modality atlas</a></article>
           <article class="artifact"><h3>Covered now</h3><p>Action/subtask labels, next-action prediction, temporal diagnostics, hand trajectory, contact, object relevance, caption grounding, retrieval, reconstruction, and misalignment.</p><a href="data/summary_metrics.json">summary metrics</a></article>
           <article class="artifact"><h3>Responsible use</h3><p>The official card notes limited diversity and showcase/production quality. This project excludes identity, surveillance, biometric, sensitive-attribute, and safety-critical uses.</p><a href="data/xperience10m_dataset_card_alignment.json">use notes</a></article>
-          <article class="artifact"><h3>Later milestones</h3><p>Full audio-visual learning, caption generation, depth-pixel prediction, SLAM estimation, neural rendering, policy learning, cross-episode generalization, and 32-episode Qwen3-Omni evaluation.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">data status</a></article>
+          <article class="artifact"><h3>Later milestones</h3><p>Full audio-visual learning, caption generation, depth-pixel prediction, SLAM estimation, neural rendering, policy learning, cross-episode generalization, and held-out Qwen3-Omni evaluation.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">data status</a></article>
         </div>
       </div>
     </section>
@@ -2729,7 +2729,7 @@
           </article>
           <article class="artifact">
             <h3>Scale means held-out episodes</h3>
-            <p>The next credible model-quality unit is a 32-episode held-out pilot across 32 sessions, not more adjacent windows from one sample.</p>
+            <p>The next credible model-quality unit is a held-out multi-episode pilot across different sessions, not more adjacent windows from one sample.</p>
             <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md">scale-up status</a>
           </article>
         </div>
@@ -2997,7 +2997,7 @@
             </button>
             <button type="button" class="content-tab" id="artifact-tab-scale-up" role="tab" data-panel-target="artifact-panel-scale-up" aria-selected="false" aria-pressed="false" aria-controls="artifact-panel-scale-up" tabindex="-1">
               <strong>Scale-Up</strong>
-              <span>data gate and Omni path</span>
+              <span>relay and Omni path</span>
             </button>
             <button type="button" class="content-tab" id="artifact-tab-checks" role="tab" data-panel-target="artifact-panel-checks" aria-selected="false" aria-pressed="false" aria-controls="artifact-panel-checks" tabindex="-1">
               <strong>Checks</strong>
@@ -3043,14 +3043,14 @@
           <section class="artifact-group tabbed-panel" id="artifact-panel-scale-up" role="tabpanel" aria-labelledby="artifact-tab-scale-up" hidden>
             <div class="artifact-group-head">
               <div><span>Scale-up path</span><h3>Prepared for multi-episode training</h3></div>
-              <p>The multi-episode Qwen3-Omni path is documented and scripted. Full-pilot metrics come after the data gate and held-out evaluation pass.</p>
+              <p>The multi-episode Qwen3-Omni path is documented and scripted. Full-pilot metrics come after selected data is staged and held-out evaluation passes.</p>
             </div>
             <div class="artifact-grid">
-              <article class="artifact primary-artifact"><div><h3>Project scope</h3><p>Connects implemented single-episode artifacts, setup-stage Omni work, pending data access, and later multi-episode milestones.</p></div><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/EVIDENCE_CONTRACT.md">EVIDENCE_CONTRACT.md</a></article>
+              <article class="artifact primary-artifact"><div><h3>Project scope</h3><p>Connects implemented single-episode artifacts, setup-stage Omni work, current 128-episode relay, and later multi-episode milestones.</p></div><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/EVIDENCE_CONTRACT.md">EVIDENCE_CONTRACT.md</a></article>
               <article class="artifact"><h3>Foundation-model plan</h3><p>Backbone selection matrix covering Qwen3-Omni, Cosmos 3, GR00T, OpenVLA/openpi, Gemini Robotics, Octo, and SmolVLA-style policy candidates.</p><a href="data/foundation_model_plan.json">foundation_model_plan.json</a></article>
-              <article class="artifact"><h3>Multi-episode access status</h3><p>Public data-access path, selected 32-episode pilot plan, and data requirements.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md">MULTI_EPISODE_ACCESS_STATUS.md</a></article>
+              <article class="artifact"><h3>Multi-episode access status</h3><p>Public data-access path, selected 128-episode relay plan, and data requirements.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md">MULTI_EPISODE_ACCESS_STATUS.md</a></article>
               <article class="artifact"><h3>Qwen3-Omni setup artifacts</h3><p>Manifests, metadata, metrics, and progress logs from the current setup run.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/episode_manifest.json">episode_manifest.json</a></article>
-              <article class="artifact"><h3>32-episode data requirement</h3><p>The data status file defines what must be available before full pilot training and held-out metrics.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">DATA_ACCESS_STATUS.md</a></article>
+              <article class="artifact"><h3>Multi-episode data requirement</h3><p>The data status file defines what must be available before full pilot training and held-out metrics.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">DATA_ACCESS_STATUS.md</a></article>
             </div>
           </section>
 
@@ -3079,14 +3079,14 @@
     <section id="omni-relay" data-project-tab="resources" role="tabpanel" aria-labelledby="tab-resources" tabindex="-1">
       <div class="wrap">
         <div class="section-head">
-          <h2>Qwen3-Omni pilot is approval-ready.</h2>
-          <p>The full Xperience-10M Hugging Face dataset is gated. While access is pending, the public plan has selected a 32-episode pilot across 32 different session UUIDs.</p>
+          <h2>Qwen3-Omni pilot is in data staging.</h2>
+          <p>Full Xperience-10M access is granted. The current plan selects 128 metadata-balanced episodes across 128 different session UUIDs, with raw staging in progress and no held-out metrics reported yet.</p>
         </div>
         <div class="artifact-grid">
-          <article class="artifact"><h3>Selection</h3><p>Stratified round-robin over 64 top-level sessions; 680 complete candidates scanned; 32 sessions selected.</p></article>
+          <article class="artifact"><h3>Selection</h3><p>128 complete episodes selected from 128 unique top-level sessions, balanced across episode-size bands and split 96/16/16 for train/val/test.</p></article>
           <article class="artifact"><h3>Transfer</h3><p>Download raw episodes only from official gated sources, exclude visualization.rrd, validate files, then stage them for training.</p></article>
-          <article class="artifact"><h3>Current LoRA artifact</h3><p>The current LoRA artifact uses the locally available sample data. The 32-episode result begins after gated data is staged and held-out evaluation runs.</p></article>
-          <article class="artifact"><h3>Backbone branches</h3><p>Qwen3-Omni is the immediate LoRA path; Cosmos 3 is the first world-model branch; GR00T/OpenVLA/openpi become policy branches after action targets are auditable.</p><a href="data/foundation_model_plan.json">backbone plan</a></article>
+          <article class="artifact"><h3>Current LoRA artifact</h3><p>The current LoRA artifact uses the locally available sample data. The multi-episode result begins after selected data is staged, preprocessed, trained, and evaluated on held-out sessions.</p></article>
+          <article class="artifact"><h3>Backbone branches</h3><p>Qwen3-Omni is the immediate LoRA path; Cosmos 3 is the first world-model branch; GR00T/OpenVLA/openpi become policy branches after action targets are well-defined.</p><a href="data/foundation_model_plan.json">backbone plan</a></article>
         </div>
       </div>
     </section>
@@ -3102,7 +3102,7 @@
           <article class="artifact"><h3>Reproducibility matrix</h3><p>Machine-readable command matrix covering sample download, baselines, 12 tasks, figures, and validation.</p><a href="data/reproducibility_matrix.json">reproducibility_matrix.json</a></article>
           <article class="artifact"><h3>Exact-match reproduction record</h3><p>The last metric rebuild reproduced the public-sample outputs from a fresh cache and matched the committed metrics.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/notes/reproducibility_audit.md">reproducibility_audit.md</a></article>
           <article class="artifact"><h3>Website reference report</h3><p>Local HTML references, anchors, JSON bundles, and image dimensions are validated before publishing.</p><a href="data/website_integrity.json">website_integrity.json</a></article>
-          <article class="artifact"><h3>32-Episode pilot status</h3><p>The 32-episode Qwen3-Omni pilot is prepared at the code and selection-plan level; final metrics follow gated data access and held-out evaluation.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">DATA_ACCESS_STATUS.md</a></article>
+          <article class="artifact"><h3>Multi-episode pilot status</h3><p>The Qwen3-Omni pilot is prepared at the code and selection-plan level; final metrics follow completed staging, preprocessing, training, and held-out evaluation.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">DATA_ACCESS_STATUS.md</a></article>
         </div>
         <p class="repro-note">Minimal path: install the toolkit dependencies, download the official sample, run the 12-task suite with neural heads, regenerate visualizations, then run the artifact index and publication validator.</p>
         <pre class="code-panel"><button type="button" data-copy="setup">Copy</button><code id="setup">git clone https://github.com/Ropedia/HOMIE-toolkit.git

results/omni_finetune/ANNOTATION_RECORD_PROBE.md

@@ -0,0 +1,140 @@
+# Xperience-10M Annotation Record Probe
+
+Minimal-cost probe. Downloaded only `annotation.hdf5`; no MP4 or `visualization.rrd` files were downloaded.
+
+- Repo: `ropedia-ai/xperience-10m`
+- Probe count: 3
+- Raw annotation cache: outside the published repo
+- Local files only: `False`
+
+## 9cecac72-8874-4b97-9541-18d4858f8e43/ep10/annotation.hdf5
+
+- Downloaded annotation size: 6.38 MiB (6,687,192 bytes)
+- HDF5 top-level keys: `calibration, caption, depth, full_body_mocap, hand_mocap, imu, metadata, slam, video`
+- HDF5 dataset count: 65
+- Largest first-dimension dataset: `imu/accel_xyz` with first dimension `190`
+
+### Caption JSON Summary
+
+| Measure | Value |
+| --- | --- |
+| Parse status | ok |
+| JSON bytes | 1,178 |
+| Segment count | 1 |
+| Current-action count | 1 |
+| Object-frame count | 1 |
+| Interaction-frame count | 1 |
+| Sampled-frame count | 1 |
+| Unique subtasks | 1 |
+| Unique action labels | 1 |
+| Unique objects | 3 |
+| Action labels | ["Arrange items in bin"] |
+| Objects | ["cardboard box", "hand", "plastic storage bin"] |
+
+### Top Groups
+
+| Group | Dataset count | Max first dimension | First-dim histogram top values |
+| --- | --- | --- | --- |
+| calibration | 23 | 4 | {"4": 14} |
+| caption | 1 | 0 | {} |
+| depth | 5 | 20 | {"20": 2} |
+| full_body_mocap | 9 | 20 | {"20": 9} |
+| hand_mocap | 10 | 20 | {"20": 10} |
+| imu | 4 | 190 | {"190": 3, "20": 1} |
+| metadata | 6 | 0 | {} |
+| slam | 4 | 47 | {"20": 3, "47": 1} |
+| video | 3 | 20 | {"20": 2} |
+
+### Caption / Action / Interaction Related Datasets
+
+| Dataset | Shape | Dtype | First dim | Sample values |
+| --- | --- | --- | --- | --- |
+| caption | [] | object | None | ["{\"config\": {\"segment_sec\": 20, \"sample_fps\": 0.5, \"total_tokens\": 2047, \"Main Task\": \"Packing items into a plastic bin. The person is placing va... |
+
+## cdc1ae12-a460-48ac-a892-7d314095c4b1/ep23/annotation.hdf5
+
+- Downloaded annotation size: 6.38 MiB (6,687,256 bytes)
+- HDF5 top-level keys: `calibration, caption, depth, full_body_mocap, hand_mocap, imu, metadata, slam, video`
+- HDF5 dataset count: 65
+- Largest first-dimension dataset: `imu/accel_xyz` with first dimension `188`
+
+### Caption JSON Summary
+
+| Measure | Value |
+| --- | --- |
+| Parse status | ok |
+| JSON bytes | 1,051 |
+| Segment count | 1 |
+| Current-action count | 1 |
+| Object-frame count | 1 |
+| Interaction-frame count | 1 |
+| Sampled-frame count | 1 |
+| Unique subtasks | 1 |
+| Unique action labels | 1 |
+| Unique objects | 4 |
+| Action labels | ["Pulling up sock"] |
+| Objects | ["bathroom floor", "feet", "sock", "toilet"] |
+
+### Top Groups
+
+| Group | Dataset count | Max first dimension | First-dim histogram top values |
+| --- | --- | --- | --- |
+| calibration | 23 | 4 | {"4": 14} |
+| caption | 1 | 0 | {} |
+| depth | 5 | 20 | {"20": 2} |
+| full_body_mocap | 9 | 20 | {"20": 9} |
+| hand_mocap | 10 | 20 | {"20": 10} |
+| imu | 4 | 188 | {"188": 3, "20": 1} |
+| metadata | 6 | 0 | {} |
+| slam | 4 | 128 | {"20": 3, "128": 1} |
+| video | 3 | 20 | {"20": 2} |
+
+### Caption / Action / Interaction Related Datasets
+
+| Dataset | Shape | Dtype | First dim | Sample values |
+| --- | --- | --- | --- | --- |
+| caption | [] | object | None | ["{\"config\": {\"segment_sec\": 20, \"sample_fps\": 0.5, \"total_tokens\": 2035, \"Main Task\": \"Putting on socks. The person is standing in a bathroom and... |
+
+## 10282b64-a955-461e-9ef9-a1ddf8dc619a/ep5/annotation.hdf5
+
+- Downloaded annotation size: 6.40 MiB (6,706,448 bytes)
+- HDF5 top-level keys: `calibration, caption, depth, full_body_mocap, hand_mocap, imu, metadata, slam, video`
+- HDF5 dataset count: 65
+- Largest first-dimension dataset: `slam/point_cloud` with first dimension `837`
+
+### Caption JSON Summary
+
+| Measure | Value |
+| --- | --- |
+| Parse status | ok |
+| JSON bytes | 1,299 |
+| Segment count | 1 |
+| Current-action count | 1 |
+| Object-frame count | 1 |
+| Interaction-frame count | 1 |
+| Sampled-frame count | 1 |
+| Unique subtasks | 1 |
+| Unique action labels | 1 |
+| Unique objects | 4 |
+| Action labels | ["Walk down retail aisle"] |
+| Objects | ["person seated", "product packaging", "retail shelf", "shopping bags"] |
+
+### Top Groups
+
+| Group | Dataset count | Max first dimension | First-dim histogram top values |
+| --- | --- | --- | --- |
+| calibration | 23 | 4 | {"4": 14} |
+| caption | 1 | 0 | {} |
+| depth | 5 | 20 | {"20": 2} |
+| full_body_mocap | 9 | 20 | {"20": 9} |
+| hand_mocap | 10 | 20 | {"20": 10} |
+| imu | 4 | 190 | {"190": 3, "20": 1} |
+| metadata | 6 | 0 | {} |
+| slam | 4 | 837 | {"20": 3, "837": 1} |
+| video | 3 | 20 | {"20": 2} |
+
+### Caption / Action / Interaction Related Datasets
+
+| Dataset | Shape | Dtype | First dim | Sample values |
+| --- | --- | --- | --- | --- |
+| caption | [] | object | None | ["{\"config\": {\"segment_sec\": 20, \"sample_fps\": 0.5, \"total_tokens\": 2060, \"Main Task\": \"walking through a retail store. The video shows a first-pe... |

results/omni_finetune/DATA_ACCESS_STATUS.md

@@ -9,10 +9,18 @@ held-out multi-episode experiment.
 | --- | --- |
 | Target pilot size | 32 valid Xperience-10M episodes |
 | Current public local sample | 1 episode |
-| Full dataset access | Pending gated-dataset approval |
+| Full dataset access | Granted; metadata-only Hugging Face audit completed |
+| Current full-dataset metadata snapshot | 12,102 complete visible HF episodes across 802 complete sessions |
+| Current staged multi-episode data | 128-episode relay started; staging not complete yet |
+| Recommended small fine-tune selection | 128 metadata-balanced episodes, 96/16/16 train/val/test |
 | Current Qwen3-Omni artifacts | Setup-stage sample run, not held-out multi-episode model metrics |
 | Public raw-data redistribution | Not included |
 
+The current 128-episode relay is an operational data-staging step. It should
+not be described as a completed fine-tune or evaluated model until all selected
+episodes are staged, audited, preprocessed, trained, and evaluated on held-out
+sessions.
+
 ## Episode Requirement
 
 A valid training episode needs `annotation.hdf5` and at least
@@ -35,14 +43,25 @@ The 32-episode pilot should only be reported after:
 | --- | ---: |
 | Local public sample | 1 |
 | ModelScope discovery | 0 |
-| Hugging Face discovery | 0 |
+| Hugging Face gated metadata audit | 12,102 complete visible episodes |
 
-These counts describe the current staged/project-visible data, not the full
-scale of Xperience-10M.
+The Hugging Face count is a metadata-only availability result. It does not mean
+that the raw files have been downloaded, staged, or used for multi-episode
+training yet.
 
 ## Related Files
 
 - `results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md`
+- `results/omni_finetune/FULL_DATASET_METADATA_AUDIT.md`
+- `results/omni_finetune/full_dataset_metadata_audit.json`
+- `results/omni_finetune/XPERIENCE10M_128_EPISODE_SELECTION.md`
+- `results/omni_finetune/XPERIENCE10M_128_RELAY_AND_FINETUNE_PLAN.md`
+- `results/omni_finetune/xperience10m_128_episode_selection.json`
+- `results/omni_finetune/xperience10m_128_episode_download_files.txt`
 - `results/omni_finetune/source_discovery.json`
 - `scripts/omni/discover_xperience10m_sources.py`
+- `scripts/omni/analyze_xperience10m_hf_metadata.py`
+- `scripts/omni/select_xperience10m_pilot_episodes.py`
+- `scripts/omni/relay_xperience10m_selection.py`
+- `scripts/omni/audit_staged_xperience10m_content.py`
 - `scripts/omni/build_episode_manifest.py`

results/omni_finetune/FULL_DATASET_METADATA_AUDIT.md

@@ -0,0 +1,115 @@
+# Xperience-10M HF Metadata Audit
+
+Metadata-only analysis of the gated Hugging Face dataset. No MP4, HDF5, RRD, or model files were downloaded.
+
+## Access and Source
+
+- Repo: `ropedia-ai/xperience-10m`
+- Repo SHA: `ce943cf271a758b60240084892d05cf6dc12dd90`
+- Last modified: `2026-04-21T05:03:45+00:00`
+- Gated mode: `manual`
+- Pretty name: `Xperience-10M`
+- License field: `other`
+- HF size category: `1M<n<10M`
+- Tags: `egocentric, first-person, multimodal, 3d, 4d, embodied-ai, robotics, human-motion, mocap, imu, audio, depth, captions, video`
+
+## Current Hub File Metadata
+
+| Measure | Value |
+| --- | --- |
+| Files listed by API | 85,257 |
+| Total bytes from file metadata | 25.52 TiB (28,057,584,187,079 bytes) |
+| Bytes excluding visualization.rrd | 24.63 TiB (27,083,292,060,675 bytes) |
+| visualization.rrd bytes | 907.38 GiB (974,292,126,404 bytes) |
+| Top-level session folders | 804 |
+| Episode-like folders | 12,103 |
+
+## File Composition
+
+| File type | Count |
+| --- | --- |
+| .hdf5 | 12,103 |
+| .md | 1 |
+| .mp4 | 72,612 |
+| .rrd | 541 |
+
+## Episode Completeness
+
+| Measure | Value |
+| --- | --- |
+| annotation.hdf5 files | 12,103 |
+| MP4 files | 72,612 |
+| visualization.rrd files | 541 |
+| Complete episodes: annotation + all six MP4 views | 12,102 (99.9917%) |
+| Degraded-valid episodes: annotation + fisheye_cam0 | 12,102 (99.9917%) |
+| Sessions with complete episodes | 802 |
+| Video-count histogram per episode | {"0": 1, "6": 12102} |
+
+## Episode Size Distribution
+
+| Statistic | Training bytes per complete episode, excluding visualization.rrd |
+| --- | --- |
+| Min | 7.78 MiB |
+| P25 | 2.13 GiB |
+| Median | 2.20 GiB |
+| P75 | 2.25 GiB |
+| Mean | 2.08 GiB |
+| Max | 2.53 GiB |
+
+## Annotation File Size Distribution
+
+| Statistic | annotation.hdf5 size |
+| --- | --- |
+| Min | 6.38 MiB |
+| P25 | 1.74 GiB |
+| Median | 1.83 GiB |
+| P75 | 1.85 GiB |
+| Mean | 1.70 GiB |
+| Max | 1.86 GiB |
+
+## Pilot Scale Estimates
+
+| Pilot | Episodes | Max windows at 256/episode | Storage estimate |
+| --- | --- | --- | --- |
+| 32-episode smallest one-per-session | 32 | 8192 | 35.35 GiB |
+| 32-episode median-sized estimate | 32 | 8192 | 70.51 GiB |
+| 32-episode mean-sized estimate | 32 | 8192 | 66.69 GiB |
+| 100-episode pilot | 100 | 25600 | roughly 220.34 GiB at median episode size |
+| 500-episode pilot | 500 | 128000 | roughly 1.08 TiB at median episode size |
+| All complete visible HF episodes | 12102 | 3098112 | 24.63 TiB |
+
+## Incomplete Episode Records
+
+[
+  {
+    "episode_path": "dc3f4139-f499-4de7-b057-e25b7dfb2d83/ep1",
+    "episode_id": "ep1",
+    "top_level_session": "dc3f4139-f499-4de7-b057-e25b7dfb2d83",
+    "file_count": 1,
+    "total_bytes": 1418232696,
+    "training_bytes_excluding_visualization_rrd": 1418232696,
+    "has_annotation": true,
+    "has_fisheye_cam0": false,
+    "video_count": 0,
+    "has_all_six_videos": false,
+    "is_degraded_valid": false,
+    "is_complete": false,
+    "has_visualization_rrd": false,
+    "missing_required_files": [
+      "fisheye_cam0.mp4",
+      "fisheye_cam1.mp4",
+      "fisheye_cam2.mp4",
+      "fisheye_cam3.mp4",
+      "stereo_left.mp4",
+      "stereo_right.mp4"
+    ]
+  }
+]
+
+## Download and Compute Recommendation
+
+- This metadata audit can run on any machine with Hugging Face access.
+- If the training host cannot reach Hugging Face, download on an HF-reachable relay host, then transfer staged episode folders to the training host.
+- For training downloads, include `annotation.hdf5` plus the six MP4 streams; exclude `visualization.rrd` unless Rerun visualization is specifically needed.
+- For the first real training pilot, prefer 32 complete episodes from different top-level sessions and avoid selecting only the tiny outlier episodes.
+- The training host is used after staged data exists: manifest validation, preprocessing, LoRA training, and held-out evaluation.

results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md

@@ -1,7 +1,9 @@
 # Multi-Episode Access Status
 
 Current status: access to the gated full `ropedia-ai/xperience-10m` dataset is
-still pending approval from the dataset authors.
+granted, and a metadata-only Hugging Face audit has been completed. A
+128-episode metadata-balanced relay has started, but the selected multi-episode
+data has not been fully staged, audited, trained, or evaluated yet.
 
 This file records the public data-access status and pilot requirements. It does
 not include local-machine aliases, private paths, SSH hosts, or token locations.
@@ -13,18 +15,21 @@ not include local-machine aliases, private paths, SSH hosts, or token locations.
 | Dataset | `ropedia-ai/xperience-10m` |
 | Target | 32 complete leaf episodes |
 | Strategy | stratified round-robin across top-level session UUIDs |
-| Candidate scan | first 64 top-level session UUIDs |
-| Valid candidates | 680 |
-| Selected sessions | 32 |
-| Minimum episode size | 0.25 GB |
-| Estimated bytes | 72,031,620,552 |
+| Metadata-audited visible complete episodes | 12,102 |
+| Metadata-audited complete sessions | 802 |
+| Recommended next selection | 128 metadata-balanced episodes |
+| Recommended split | 96 train / 16 val / 16 test |
+| Recommended estimated download | 277.71 GiB excluding `visualization.rrd` |
+| Representative 32-episode estimate | ~70.5 GiB at median episode size |
+| Smallest one-per-session 32-episode estimate | 35.35 GiB |
 | Excluded file | `visualization.rrd` |
 
 ## Current Stage
 
 The current Qwen3-Omni artifacts come from the locally available sample data.
-The 32-episode held-out model-quality run starts after the selected episodes
-are available locally.
+The held-out model-quality run starts after selected complete episodes are
+downloaded, transferred if needed, validated locally, audited for content
+balance, and preprocessed into train/val/test examples.
 
 A real 32-episode pilot can be claimed only after:
 
@@ -38,6 +43,15 @@ The reader-facing data access summary is:
 
 `results/omni_finetune/DATA_ACCESS_STATUS.md`
 
-The machine-generated discovery report remains:
+The current metadata-only full dataset audit is:
+
+`results/omni_finetune/FULL_DATASET_METADATA_AUDIT.md`
+
+The current 128-episode metadata-balanced download plan is:
+
+`results/omni_finetune/XPERIENCE10M_128_EPISODE_SELECTION.md`
+
+The older machine-generated source discovery blocker remains a pre-access local
+staging record:
 
 `results/omni_finetune/DATA_BLOCKER_REPORT.md`

results/omni_finetune/XPERIENCE10M_128_EPISODE_SELECTION.md

@@ -0,0 +1,55 @@
+# Xperience-10M 128-Episode Metadata-Balanced Selection
+
+This is a download plan, not a trained model result. It uses Hugging Face file metadata only and downloads no raw episode data.
+
+## Why This Selection
+
+- Use only complete episodes: `annotation.hdf5` plus six MP4 streams.
+- Exclude `visualization.rrd` from the training download plan.
+- Avoid tiny annotation outliers that are likely one-segment examples.
+- Use one episode per top-level session to reduce leakage and overfitting to one capture session.
+- Balance across four annotation-size bands as a proxy for duration/content richness before category labels are available.
+- Split by session into train/val/test.
+
+## Selection Summary
+
+| Measure | Value |
+| --- | --- |
+| Selected episodes | 128 |
+| Unique sessions | 128 |
+| Split counts | {"test": 16, "train": 96, "val": 16} |
+| Size-band counts | {"long": 32, "lower_mid": 32, "short": 32, "upper_mid": 32} |
+| Estimated training download, no RRD | 277.71 GiB |
+| Estimated annotation bytes | 226.53 GiB |
+| Estimated windows at 256/episode | 32768 |
+| Session leakage train/val | 0 |
+| Session leakage train/test | 0 |
+| Session leakage val/test | 0 |
+
+## Filters
+
+| Rule | Value |
+| --- | --- |
+| Available complete episodes | 12102 |
+| Candidates after filters | 11478 |
+| Minimum annotation size | 992.76 MiB |
+| Minimum training size | 1.22 GiB |
+| Rejected counts | {"annotation_too_small": 606, "training_too_small": 18} |
+
+## Split x Size Band
+
+| Split | short | lower_mid | upper_mid | long |
+| --- | --- | --- | --- | --- |
+| train | 24 | 24 | 24 | 24 |
+| val | 4 | 4 | 4 | 4 |
+| test | 4 | 4 | 4 | 4 |
+
+## Important Limitation
+
+HF metadata does not expose semantic content categories. This selection is the best first-pass balance before downloading. After the selected annotations are staged, parse `Main Task`, `Sub Task`, `Current Action`, objects, and interaction text; then swap episodes if one content cluster dominates.
+
+## Output Files
+
+- JSON: `results/omni_finetune/xperience10m_128_episode_selection.json`
+- CSV: `results/omni_finetune/xperience10m_128_episode_selection.csv`
+- Download file list: `results/omni_finetune/xperience10m_128_episode_download_files.txt`

results/omni_finetune/XPERIENCE10M_128_RELAY_AND_FINETUNE_PLAN.md

@@ -0,0 +1,195 @@
+# Xperience-10M 128-Episode Relay and Fine-Tune Plan
+
+This is the executable plan for moving from metadata selection to real
+multi-episode training. It does not claim model-quality results until data is
+downloaded, staged, audited, trained, and evaluated on held-out sessions.
+
+## Current Preflight
+
+| Host | Role | Status |
+| --- | --- | --- |
+| HF-reachable relay host | Dataset download relay | Needs Hugging Face access and enough scratch storage for one batch |
+| Private training host | Persistent data + training | Needs enough storage for the staged selection and the training/eval stack |
+
+Conclusion: use a Hugging Face reachable machine as the download relay and the
+private training machine as the persistent data store. Even when the relay has
+enough free space for the full selection, the safer execution downloads one
+batch, transfers it to the training host, then deletes the relay-local batch.
+
+Private execution status:
+
+- a 128-episode relay job has been launched on a private HF-reachable host,
+- the first batch is downloading,
+- no multi-episode model-quality training result is claimed yet.
+
+## Selected Data
+
+- Selection file: `results/omni_finetune/xperience10m_128_episode_selection.json`
+- Download list: `results/omni_finetune/xperience10m_128_episode_download_files.txt`
+- Episodes: 128
+- Sessions: 128 unique sessions
+- Split: 96 train / 16 val / 16 test
+- Files: 896 training files
+- Excluded: `visualization.rrd`
+- Estimated training-host storage: 277.71 GiB excluding RRD
+
+## Relay Setup
+
+Define host-specific paths outside the public artifact:
+
+```bash
+export RELAY_WORKDIR=/path/to/ropedia-episode-task-suite
+export RELAY_ROOT=/path/to/xperience10m_relay
+export TRAINING_HOST=<training-user>@<training-host>
+export TRAINING_REPO=/path/to/ropedia-episode-task-suite
+export TRAINING_DATA_ROOT=/path/to/xperience10m_128
+```
+
+Create a dedicated relay-to-training SSH key:
+
+```bash
+ssh <relay-host> 'mkdir -p ~/.ssh && chmod 700 ~/.ssh && test -f ~/.ssh/xperience10m_relay_ed25519 || ssh-keygen -t ed25519 -N "" -f ~/.ssh/xperience10m_relay_ed25519 -C xperience10m-relay-to-training'
+ssh <relay-host> 'cat ~/.ssh/xperience10m_relay_ed25519.pub'
+```
+
+Append that public key to the training host `~/.ssh/authorized_keys`, then verify from the relay:
+
+```bash
+ssh <relay-host> 'ssh -i ~/.ssh/xperience10m_relay_ed25519 -o BatchMode=yes -o StrictHostKeyChecking=accept-new <training-user>@<training-host> hostname'
+```
+
+## Copy Minimal Repo Files to Relay
+
+```bash
+ssh <relay-host> 'mkdir -p "$RELAY_WORKDIR"'
+rsync -av \
+  scripts/omni/relay_xperience10m_selection.py \
+  results/omni_finetune/xperience10m_128_episode_selection.json \
+  <relay-host>:"$RELAY_WORKDIR"/
+```
+
+## Relay Dry Run
+
+```bash
+ssh <relay-host> '
+cd "$RELAY_WORKDIR" &&
+python3 relay_xperience10m_selection.py \
+  --selection-json xperience10m_128_episode_selection.json \
+  --relay-root "$RELAY_ROOT" \
+  --batch-max-gib 40 \
+  --batch-max-episodes 16 \
+  --transfer-host "$TRAINING_HOST" \
+  --transfer-root "$TRAINING_DATA_ROOT" \
+  --ssh-key ~/.ssh/xperience10m_relay_ed25519 \
+  --delete-after-transfer \
+  --dry-run
+'
+```
+
+## Start Relay
+
+Run in a persistent terminal or `tmux` session on the relay:
+
+```bash
+export HF_TOKEN=...
+cd "$RELAY_WORKDIR"
+python3 relay_xperience10m_selection.py \
+  --selection-json xperience10m_128_episode_selection.json \
+  --relay-root "$RELAY_ROOT" \
+  --batch-max-gib 40 \
+  --batch-max-episodes 16 \
+  --transfer-host "$TRAINING_HOST" \
+  --transfer-root "$TRAINING_DATA_ROOT" \
+  --ssh-key ~/.ssh/xperience10m_relay_ed25519 \
+  --delete-after-transfer
+```
+
+Batch sizing is intentionally conservative. A 40 GiB batch size keeps restarts
+and partial-transfer cleanup cheaper than treating the full 277.71 GiB selection
+as one unit.
+
+## Training-Host Data Validation
+
+After transfer completes:
+
+```bash
+cd "$TRAINING_REPO"
+python3 scripts/omni/discover_xperience10m_sources.py \
+  --workspace "$TRAINING_REPO" \
+  --data-root "$TRAINING_DATA_ROOT" \
+  --output results/omni_finetune/source_discovery_128.json \
+  --report-output results/omni_finetune/DATA_BLOCKER_REPORT_128.md \
+  --target-episodes 128 \
+  --skip-modelscope \
+  --skip-huggingface
+```
+
+Then build the episode manifest:
+
+```bash
+python3 scripts/omni/build_episode_manifest.py \
+  --workspace "$TRAINING_REPO" \
+  --data-root "$TRAINING_DATA_ROOT" \
+  --max-episodes 128 \
+  --train-fraction 0.75 \
+  --val-fraction 0.125 \
+  --test-fraction 0.125 \
+  --split-seed 7 \
+  --output results/omni_finetune/episode_manifest_128.json
+```
+
+## Content Rebalance Gate
+
+Parse staged annotations before training:
+
+```bash
+python3 scripts/omni/audit_staged_xperience10m_content.py \
+  --data-root "$TRAINING_DATA_ROOT" \
+  --selection-json results/omni_finetune/xperience10m_128_episode_selection.json \
+  --output-json results/omni_finetune/staged_content_audit_128.json \
+  --output-csv results/omni_finetune/staged_content_audit_128.csv \
+  --report-output results/omni_finetune/STAGED_CONTENT_AUDIT_128.md
+```
+
+If a category dominates train, val, or test, swap episodes before training.
+
+## Training Order
+
+### 1. Qwen3-Omni LoRA Baseline
+
+Use this as the first real multi-episode SFT run because the repo already has
+working Qwen3-Omni training/eval scripts.
+
+Expected dataset:
+
+- 128 episodes
+- 32,768 max windows at 256 windows per episode
+- held-out sessions in val/test
+
+### 2. Cosmos3-Nano Compatibility
+
+Cosmos3-Nano should be treated as a second branch:
+
+- first run inference compatibility on a few staged clips,
+- then adapt data format for Cosmos video/action tasks,
+- then run post-training only after Qwen3-Omni and content audit pass.
+
+Good Cosmos tasks:
+
+- video + text -> physical reasoning text,
+- video + text -> future state/action label,
+- video + action/text -> future video,
+- video + text -> action trajectory proxy.
+
+Do not start with Cosmos3-Super. Cosmos3-Nano is the practical first target;
+Super is for a later run after data format, metrics, and compute are stable.
+
+## Acceptance Gates
+
+- 128 selected episodes staged on the training host.
+- No `visualization.rrd` in training data.
+- 128 unique sessions preserved.
+- Train/val/test session leakage is zero.
+- Content audit reviewed before training.
+- Qwen3-Omni eval runs on held-out sessions.
+- Cosmos3-Nano branch starts with compatibility, not immediate full fine-tune.

results/omni_finetune/annotation_record_probe.json

@@ -0,0 +1,455 @@
+{
+  "status": "pass",
+  "generated_at_utc": "2026-06-03T15:11:33+00:00",
+  "repo_id": "ropedia-ai/xperience-10m",
+  "download_policy": "annotation.hdf5 only; no videos or visualization.rrd downloaded",
+  "cache_note": "raw annotation files were cached outside the published repo",
+  "probes": [
+    {
+      "repo_filename": "9cecac72-8874-4b97-9541-18d4858f8e43/ep10/annotation.hdf5",
+      "inspection": {
+        "cache_note": "annotation file cached outside the published repo",
+        "local_bytes": 6687192,
+        "local_human": "6.38 MiB",
+        "top_level_keys": [
+          "calibration",
+          "caption",
+          "depth",
+          "full_body_mocap",
+          "hand_mocap",
+          "imu",
+          "metadata",
+          "slam",
+          "video"
+        ],
+        "dataset_count": 65,
+        "dataset_first_dim_histogram_top20": {
+          "20": 27,
+          "4": 14,
+          "190": 3,
+          "47": 1
+        },
+        "top_group_stats": {
+          "calibration": {
+            "dataset_count": 23,
+            "max_first_dim": 4,
+            "first_dim_values": {
+              "4": 14
+            }
+          },
+          "caption": {
+            "dataset_count": 1,
+            "max_first_dim": 0,
+            "first_dim_values": {}
+          },
+          "depth": {
+            "dataset_count": 5,
+            "max_first_dim": 20,
+            "first_dim_values": {
+              "20": 2
+            }
+          },
+          "full_body_mocap": {
+            "dataset_count": 9,
+            "max_first_dim": 20,
+            "first_dim_values": {
+              "20": 9
+            }
+          },
+          "hand_mocap": {
+            "dataset_count": 10,
+            "max_first_dim": 20,
+            "first_dim_values": {
+              "20": 10
+            }
+          },
+          "imu": {
+            "dataset_count": 4,
+            "max_first_dim": 190,
+            "first_dim_values": {
+              "190": 3,
+              "20": 1
+            }
+          },
+          "metadata": {
+            "dataset_count": 6,
+            "max_first_dim": 0,
+            "first_dim_values": {}
+          },
+          "slam": {
+            "dataset_count": 4,
+            "max_first_dim": 47,
+            "first_dim_values": {
+              "20": 3,
+              "47": 1
+            }
+          },
+          "video": {
+            "dataset_count": 3,
+            "max_first_dim": 20,
+            "first_dim_values": {
+              "20": 2
+            }
+          }
+        },
+        "max_first_dim_dataset": {
+          "path": "imu/accel_xyz",
+          "shape": [
+            190,
+            3
+          ],
+          "dtype": "float64",
+          "first_dim": 190,
+          "storage_bytes": 4560,
+          "storage_human": "4.45 KiB"
+        },
+        "text_action_interaction_related_datasets": [
+          {
+            "path": "caption",
+            "shape": [],
+            "dtype": "object",
+            "first_dim": null,
+            "storage_bytes": 16,
+            "storage_human": "16.00 B",
+            "sample_values": [
+              "{\"config\": {\"segment_sec\": 20, \"sample_fps\": 0.5, \"total_tokens\": 2047, \"Main Task\": \"Packing items into a plastic bin. The person is placing various items into a clear plastic storage container.\"}, \"segments\": [{\"segment_id\": 0, \"start_frame\": \"82777404821554\", \"end_frame\": \"frame_0000021\", \"Sub Task\": \"Packing items into a plastic bin\", \"Current Action\": [{\"label\": \"Arrange items in bin\", \"description\": \"The person adjusts the position of items inside the plastic storage container to ensure they are organized.\", \"start_frame\": 82777404821554, \"end_frame\": 82777404821554}], \"sampled_frames\": {\"Image 1\": 82777404821554}, \"objects\": {\"82777404821554\": [\"plastic storage bin\", \"hand\", \"cardboard box\"]}, \"interaction\": {\"82777404821554\": \"The hand is reaching into and organizing items inside the plastic storage bin.\"}, \"api_call_start\": \"2026-03-12T19:33:43.280472\", \"api_call_end\": \"2026-03-12T19:33:44.979810\", \"tokens_in\": 1842, \"tokens_out\": 205}], \"global_summary\": \"The video depicts the process of organizing and packing various personal items into a plastic storage container. It focuses on the practical task of tidying up or preparing belongings for storage.\"}"
+            ]
+          }
+        ],
+        "caption_json_summary": {
+          "parse_status": "ok",
+          "json_bytes": 1178,
+          "top_keys": [
+            "config",
+            "segments",
+            "global_summary"
+          ],
+          "config": {
+            "segment_sec": 20,
+            "sample_fps": 0.5,
+            "total_tokens": 2047,
+            "Main Task": "Packing items into a plastic bin. The person is placing various items into a clear plastic storage container."
+          },
+          "segment_count": 1,
+          "current_action_count": 1,
+          "unique_sub_task_count": 1,
+          "unique_action_label_count": 1,
+          "object_frame_count": 1,
+          "interaction_frame_count": 1,
+          "sampled_frame_count": 1,
+          "unique_object_count": 3,
+          "sub_tasks": [
+            "Packing items into a plastic bin"
+          ],
+          "action_labels": [
+            "Arrange items in bin"
+          ],
+          "objects": [
+            "cardboard box",
+            "hand",
+            "plastic storage bin"
+          ],
+          "global_summary_preview": "The video depicts the process of organizing and packing various personal items into a plastic storage container. It focuses on the practical task of tidying up or preparing belongings for storage."
+        }
+      }
+    },
+    {
+      "repo_filename": "cdc1ae12-a460-48ac-a892-7d314095c4b1/ep23/annotation.hdf5",
+      "inspection": {
+        "cache_note": "annotation file cached outside the published repo",
+        "local_bytes": 6687256,
+        "local_human": "6.38 MiB",
+        "top_level_keys": [
+          "calibration",
+          "caption",
+          "depth",
+          "full_body_mocap",
+          "hand_mocap",
+          "imu",
+          "metadata",
+          "slam",
+          "video"
+        ],
+        "dataset_count": 65,
+        "dataset_first_dim_histogram_top20": {
+          "20": 27,
+          "4": 14,
+          "188": 3,
+          "128": 1
+        },
+        "top_group_stats": {
+          "calibration": {
+            "dataset_count": 23,
+            "max_first_dim": 4,
+            "first_dim_values": {
+              "4": 14
+            }
+          },
+          "caption": {
+            "dataset_count": 1,
+            "max_first_dim": 0,
+            "first_dim_values": {}
+          },
+          "depth": {
+            "dataset_count": 5,
+            "max_first_dim": 20,
+            "first_dim_values": {
+              "20": 2
+            }
+          },
+          "full_body_mocap": {
+            "dataset_count": 9,
+            "max_first_dim": 20,
+            "first_dim_values": {
+              "20": 9
+            }
+          },
+          "hand_mocap": {
+            "dataset_count": 10,
+            "max_first_dim": 20,
+            "first_dim_values": {
+              "20": 10
+            }
+          },
+          "imu": {
+            "dataset_count": 4,
+            "max_first_dim": 188,
+            "first_dim_values": {
+              "188": 3,
+              "20": 1
+            }
+          },
+          "metadata": {
+            "dataset_count": 6,
+            "max_first_dim": 0,
+            "first_dim_values": {}
+          },
+          "slam": {
+            "dataset_count": 4,
+            "max_first_dim": 128,
+            "first_dim_values": {
+              "20": 3,
+              "128": 1
+            }
+          },
+          "video": {
+            "dataset_count": 3,
+            "max_first_dim": 20,
+            "first_dim_values": {
+              "20": 2
+            }
+          }
+        },
+        "max_first_dim_dataset": {
+          "path": "imu/accel_xyz",
+          "shape": [
+            188,
+            3
+          ],
+          "dtype": "float64",
+          "first_dim": 188,
+          "storage_bytes": 4512,
+          "storage_human": "4.41 KiB"
+        },
+        "text_action_interaction_related_datasets": [
+          {
+            "path": "caption",
+            "shape": [],
+            "dtype": "object",
+            "first_dim": null,
+            "storage_bytes": 16,
+            "storage_human": "16.00 B",
+            "sample_values": [
+              "{\"config\": {\"segment_sec\": 20, \"sample_fps\": 0.5, \"total_tokens\": 2035, \"Main Task\": \"Putting on socks. The person is standing in a bathroom and putting on their socks.\"}, \"segments\": [{\"segment_id\": 0, \"start_frame\": \"78968504788029\", \"end_frame\": \"78969405629613\", \"Sub Task\": \"Putting on socks\", \"Current Action\": [{\"label\": \"Pulling up sock\", \"description\": \"The person is manually adjusting and pulling up a sock on their foot.\", \"start_frame\": 78968504788029, \"end_frame\": 78968504788029}], \"sampled_frames\": {\"Image 1\": 78968504788029}, \"objects\": {\"78968504788029\": [\"sock\", \"feet\", \"bathroom floor\", \"toilet\"]}, \"interaction\": {\"78968504788029\": \"The person's hands are gripping and pulling on the fabric of the sock to adjust it over their foot.\"}, \"api_call_start\": \"2026-03-11T07:58:15.838321\", \"api_call_end\": \"2026-03-11T07:58:17.411279\", \"tokens_in\": 1839, \"tokens_out\": 196}], \"global_summary\": \"The video focuses on the simple, everyday task of putting on socks. It provides a brief look at this routine action as the central theme.\"}"
+            ]
+          }
+        ],
+        "caption_json_summary": {
+          "parse_status": "ok",
+          "json_bytes": 1051,
+          "top_keys": [
+            "config",
+            "segments",
+            "global_summary"
+          ],
+          "config": {
+            "segment_sec": 20,
+            "sample_fps": 0.5,
+            "total_tokens": 2035,
+            "Main Task": "Putting on socks. The person is standing in a bathroom and putting on their socks."
+          },
+          "segment_count": 1,
+          "current_action_count": 1,
+          "unique_sub_task_count": 1,
+          "unique_action_label_count": 1,
+          "object_frame_count": 1,
+          "interaction_frame_count": 1,
+          "sampled_frame_count": 1,
+          "unique_object_count": 4,
+          "sub_tasks": [
+            "Putting on socks"
+          ],
+          "action_labels": [
+            "Pulling up sock"
+          ],
+          "objects": [
+            "bathroom floor",
+            "feet",
+            "sock",
+            "toilet"
+          ],
+          "global_summary_preview": "The video focuses on the simple, everyday task of putting on socks. It provides a brief look at this routine action as the central theme."
+        }
+      }
+    },
+    {
+      "repo_filename": "10282b64-a955-461e-9ef9-a1ddf8dc619a/ep5/annotation.hdf5",
+      "inspection": {
+        "cache_note": "annotation file cached outside the published repo",
+        "local_bytes": 6706448,
+        "local_human": "6.40 MiB",
+        "top_level_keys": [
+          "calibration",
+          "caption",
+          "depth",
+          "full_body_mocap",
+          "hand_mocap",
+          "imu",
+          "metadata",
+          "slam",
+          "video"
+        ],
+        "dataset_count": 65,
+        "dataset_first_dim_histogram_top20": {
+          "20": 27,
+          "4": 14,
+          "190": 3,
+          "837": 1
+        },
+        "top_group_stats": {
+          "calibration": {
+            "dataset_count": 23,
+            "max_first_dim": 4,
+            "first_dim_values": {
+              "4": 14
+            }
+          },
+          "caption": {
+            "dataset_count": 1,
+            "max_first_dim": 0,
+            "first_dim_values": {}
+          },
+          "depth": {
+            "dataset_count": 5,
+            "max_first_dim": 20,
+            "first_dim_values": {
+              "20": 2
+            }
+          },
+          "full_body_mocap": {
+            "dataset_count": 9,
+            "max_first_dim": 20,
+            "first_dim_values": {
+              "20": 9
+            }
+          },
+          "hand_mocap": {
+            "dataset_count": 10,
+            "max_first_dim": 20,
+            "first_dim_values": {
+              "20": 10
+            }
+          },
+          "imu": {
+            "dataset_count": 4,
+            "max_first_dim": 190,
+            "first_dim_values": {
+              "190": 3,
+              "20": 1
+            }
+          },
+          "metadata": {
+            "dataset_count": 6,
+            "max_first_dim": 0,
+            "first_dim_values": {}
+          },
+          "slam": {
+            "dataset_count": 4,
+            "max_first_dim": 837,
+            "first_dim_values": {
+              "20": 3,
+              "837": 1
+            }
+          },
+          "video": {
+            "dataset_count": 3,
+            "max_first_dim": 20,
+            "first_dim_values": {
+              "20": 2
+            }
+          }
+        },
+        "max_first_dim_dataset": {
+          "path": "slam/point_cloud",
+          "shape": [
+            837,
+            3
+          ],
+          "dtype": "float64",
+          "first_dim": 837,
+          "storage_bytes": 20088,
+          "storage_human": "19.62 KiB"
+        },
+        "text_action_interaction_related_datasets": [
+          {
+            "path": "caption",
+            "shape": [],
+            "dtype": "object",
+            "first_dim": null,
+            "storage_bytes": 16,
+            "storage_human": "16.00 B",
+            "sample_values": [
+              "{\"config\": {\"segment_sec\": 20, \"sample_fps\": 0.5, \"total_tokens\": 2060, \"Main Task\": \"walking through a retail store. The video shows a first-person perspective of someone walking through a retail aisle lined with shelves of products, while other people are seated nearby.\"}, \"segments\": [{\"segment_id\": 0, \"start_frame\": \"78307554787048\", \"end_frame\": \"frame_0000021\", \"Sub Task\": \"walking through a retail store\", \"Current Action\": [{\"label\": \"Walk down retail aisle\", \"description\": \"The camera operator is walking along a retail aisle while observing merchandise on the shelves and other people in the store.\", \"start_frame\": \"78307554787048\", \"end_frame\": \"78307554787048\"}], \"sampled_frames\": {\"Image 1\": 78307554787048}, \"objects\": {\"78307554787048\": [\"retail shelf\", \"product packaging\", \"shopping bags\", \"person seated\"]}, \"interaction\": {\"78307554787048\": \"The individual is walking through the store environment, passing by shelved products and people.\"}, \"api_call_start\": \"2026-03-12T13:50:32.723710\", \"api_call_end\": \"2026-03-12T13:50:34.677914\", \"tokens_in\": 1856, \"tokens_out\": 204}], \"global_summary\": \"The video captures a casual, observational stroll through a retail store environment. It focuses on the experience of navigating the aisles and browsing the products on display.\"}"
+            ]
+          }
+        ],
+        "caption_json_summary": {
+          "parse_status": "ok",
+          "json_bytes": 1299,
+          "top_keys": [
+            "config",
+            "segments",
+            "global_summary"
+          ],
+          "config": {
+            "segment_sec": 20,
+            "sample_fps": 0.5,
+            "total_tokens": 2060,
+            "Main Task": "walking through a retail store. The video shows a first-person perspective of someone walking through a retail aisle lined with shelves of products, while other people are seated nearby."
+          },
+          "segment_count": 1,
+          "current_action_count": 1,
+          "unique_sub_task_count": 1,
+          "unique_action_label_count": 1,
+          "object_frame_count": 1,
+          "interaction_frame_count": 1,
+          "sampled_frame_count": 1,
+          "unique_object_count": 4,
+          "sub_tasks": [
+            "walking through a retail store"
+          ],
+          "action_labels": [
+            "Walk down retail aisle"
+          ],
+          "objects": [
+            "person seated",
+            "product packaging",
+            "retail shelf",
+            "shopping bags"
+          ],
+          "global_summary_preview": "The video captures a casual, observational stroll through a retail store environment. It focuses on the experience of navigating the aisles and browsing the products on display."
+        }
+      }
+    }
+  ]
+}

results/omni_finetune/full_dataset_metadata_audit.json

@@ -0,0 +1,2255 @@
+{
+  "status": "pass",
+  "generated_at_utc": "2026-06-03T15:10:54+00:00",
+  "repo_id": "ropedia-ai/xperience-10m",
+  "repo_sha": "ce943cf271a758b60240084892d05cf6dc12dd90",
+  "gated": "manual",
+  "last_modified": "2026-04-21T05:03:45+00:00",
+  "card_data": {
+    "language": [
+      "en"
+    ],
+    "license": "other",
+    "size_categories": [
+      "1M<n<10M"
+    ],
+    "task_categories": [
+      "video-classification",
+      "image-to-text",
+      "depth-estimation",
+      "robotics"
+    ],
+    "pretty_name": "Xperience-10M",
+    "tags": [
+      "egocentric",
+      "first-person",
+      "multimodal",
+      "3d",
+      "4d",
+      "embodied-ai",
+      "robotics",
+      "human-motion",
+      "mocap",
+      "imu",
+      "audio",
+      "depth",
+      "captions",
+      "video"
+    ],
+    "extra_gated_heading": "Request controlled access to Xperience-10M",
+    "extra_gated_description": "Access is reviewed manually and is limited to approved non-commercial use. Completion of an external agreement-signing step may be required before approval. Please sign the agreement via [DocuSign](https://ropedia.docsend.com/view/ra7ej7gs6s98sw87)",
+    "extra_gated_button_content": "I have signed the access control agreement"
+  },
+  "summary": {
+    "sibling_count": 85258,
+    "file_count_excluding_gitattributes": 85257,
+    "total_bytes_from_file_metadata": 28057584187079,
+    "total_human_from_file_metadata": "25.52 TiB",
+    "training_bytes_excluding_visualization_rrd": 27083292060675,
+    "training_human_excluding_visualization_rrd": "24.63 TiB",
+    "visualization_rrd_bytes": 974292126404,
+    "visualization_rrd_human": "907.38 GiB",
+    "top_level_session_count": 804,
+    "episode_like_folder_count": 12103,
+    "annotation_hdf5_count": 12103,
+    "mp4_count": 72612,
+    "visualization_rrd_count": 541,
+    "complete_episode_count": 12102,
+    "degraded_valid_episode_count": 12102,
+    "complete_episode_pct": 99.9917,
+    "degraded_valid_episode_pct": 99.9917,
+    "complete_sessions": 802,
+    "degraded_valid_sessions": 802,
+    "all_complete_episode_training_bytes_excluding_visualization_rrd": 27081873816281,
+    "all_complete_episode_training_human_excluding_visualization_rrd": "24.63 TiB"
+  },
+  "file_type_counts": {
+    ".hdf5": 12103,
+    ".md": 1,
+    ".mp4": 72612,
+    ".rrd": 541
+  },
+  "basename_counts": {
+    "README.md": 1,
+    "annotation.hdf5": 12103,
+    "fisheye_cam0.mp4": 12102,
+    "fisheye_cam1.mp4": 12102,
+    "fisheye_cam2.mp4": 12102,
+    "fisheye_cam3.mp4": 12102,
+    "stereo_left.mp4": 12102,
+    "stereo_right.mp4": 12102,
+    "visualization.rrd": 541
+  },
+  "video_count_histogram": {
+    "0": 1,
+    "6": 12102
+  },
+  "episode_count_per_session_summary": {
+    "count": 803,
+    "min": 1,
+    "p25": 8,
+    "median": 11,
+    "p75": 13,
+    "max": 51,
+    "mean": 15.07
+  },
+  "episode_size_summary": {
+    "count": 12103,
+    "min_bytes": 8158968,
+    "p25_bytes": 2292986495,
+    "median_bytes": 2369826102,
+    "p75_bytes": 2420801956,
+    "max_bytes": 5108045286,
+    "mean_bytes": 2318233840,
+    "min_human": "7.78 MiB",
+    "median_human": "2.21 GiB",
+    "mean_human": "2.16 GiB",
+    "max_human": "4.76 GiB"
+  },
+  "annotation_file_size_summary": {
+    "count": 12103,
+    "min_bytes": 6687192,
+    "p25_bytes": 1869556164,
+    "median_bytes": 1969457752,
+    "p75_bytes": 1989975784,
+    "max_bytes": 2000432556,
+    "mean_bytes": 1830083732,
+    "min_human": "6.38 MiB",
+    "median_human": "1.83 GiB",
+    "mean_human": "1.70 GiB",
+    "max_human": "1.86 GiB"
+  },
+  "complete_episode_training_size_summary": {
+    "count": 12102,
+    "min_bytes": 8158968,
+    "p25_bytes": 2286228606,
+    "median_bytes": 2365869284,
+    "p75_bytes": 2412829381,
+    "max_bytes": 2721499520,
+    "mean_bytes": 2237801505,
+    "min_human": "7.78 MiB",
+    "median_human": "2.20 GiB",
+    "mean_human": "2.08 GiB",
+    "max_human": "2.53 GiB"
+  },
+  "incomplete_episode_records": [
+    {
+      "episode_path": "dc3f4139-f499-4de7-b057-e25b7dfb2d83/ep1",
+      "episode_id": "ep1",
+      "top_level_session": "dc3f4139-f499-4de7-b057-e25b7dfb2d83",
+      "file_count": 1,
+      "total_bytes": 1418232696,
+      "training_bytes_excluding_visualization_rrd": 1418232696,
+      "has_annotation": true,
+      "has_fisheye_cam0": false,
+      "video_count": 0,
+      "has_all_six_videos": false,
+      "is_degraded_valid": false,
+      "is_complete": false,
+      "has_visualization_rrd": false,
+      "missing_required_files": [
+        "fisheye_cam0.mp4",
+        "fisheye_cam1.mp4",
+        "fisheye_cam2.mp4",
+        "fisheye_cam3.mp4",
+        "stereo_left.mp4",
+        "stereo_right.mp4"
+      ]
+    }
+  ],
+  "pilot_scale_estimates": {
+    "windows_per_episode": 256,
+    "all_complete_episodes_windows_at_256_each": 3098112,
+    "episode_32_windows_at_256_each": 8192,
+    "episode_100_windows_at_256_each": 25600,
+    "episode_500_windows_at_256_each": 128000,
+    "median_based_32_episode_training_bytes": 75707817088,
+    "median_based_32_episode_training_human": "70.51 GiB",
+    "mean_based_32_episode_training_bytes": 71609648160,
+    "mean_based_32_episode_training_human": "66.69 GiB"
+  },
+  "selected_32_smallest_one_per_session_estimate": {
+    "episode_count": 32,
+    "estimated_training_bytes_excluding_visualization_rrd": 37955941983,
+    "estimated_training_human": "35.35 GiB",
+    "episodes": [
+      {
+        "episode_path": "003dcaf0-edba-4787-ada0-187d2748f684/ep11",
+        "episode_id": "ep11",
+        "top_level_session": "003dcaf0-edba-4787-ada0-187d2748f684",
+        "file_count": 7,
+        "total_bytes": 2163028071,
+        "training_bytes_excluding_visualization_rrd": 2163028071,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "0073a4be-ec41-47f3-92d6-f63cf89aec9b/ep8",
+        "episode_id": "ep8",
+        "top_level_session": "0073a4be-ec41-47f3-92d6-f63cf89aec9b",
+        "file_count": 8,
+        "total_bytes": 2766042581,
+        "training_bytes_excluding_visualization_rrd": 1456277585,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": true,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "008316ec-3f52-4e8f-846a-e7bc82b418d6/ep12",
+        "episode_id": "ep12",
+        "top_level_session": "008316ec-3f52-4e8f-846a-e7bc82b418d6",
+        "file_count": 7,
+        "total_bytes": 689449531,
+        "training_bytes_excluding_visualization_rrd": 689449531,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "00986a68-7248-492a-befc-daae0c3f9866/ep2",
+        "episode_id": "ep2",
+        "top_level_session": "00986a68-7248-492a-befc-daae0c3f9866",
+        "file_count": 8,
+        "total_bytes": 107130526,
+        "training_bytes_excluding_visualization_rrd": 58620463,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": true,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "021c920e-ed68-47c8-9a94-fda10d4fe4cd/ep12",
+        "episode_id": "ep12",
+        "top_level_session": "021c920e-ed68-47c8-9a94-fda10d4fe4cd",
+        "file_count": 7,
+        "total_bytes": 888600873,
+        "training_bytes_excluding_visualization_rrd": 888600873,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "02575e43-d2b0-46d1-a025-415f3d82ef26/ep9",
+        "episode_id": "ep9",
+        "top_level_session": "02575e43-d2b0-46d1-a025-415f3d82ef26",
+        "file_count": 7,
+        "total_bytes": 1906075876,
+        "training_bytes_excluding_visualization_rrd": 1906075876,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "02a859ba-0603-41a4-93c2-ba4f87a95805/ep5",
+        "episode_id": "ep5",
+        "top_level_session": "02a859ba-0603-41a4-93c2-ba4f87a95805",
+        "file_count": 7,
+        "total_bytes": 886112666,
+        "training_bytes_excluding_visualization_rrd": 886112666,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "03728b87-3d0c-4d48-b73c-3c0abc1019e4/ep2",
+        "episode_id": "ep2",
+        "top_level_session": "03728b87-3d0c-4d48-b73c-3c0abc1019e4",
+        "file_count": 8,
+        "total_bytes": 731769483,
+        "training_bytes_excluding_visualization_rrd": 372047999,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": true,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "0376a3fd-52ac-4de6-afd1-4505e235010a/ep1",
+        "episode_id": "ep1",
+        "top_level_session": "0376a3fd-52ac-4de6-afd1-4505e235010a",
+        "file_count": 8,
+        "total_bytes": 85280549,
+        "training_bytes_excluding_visualization_rrd": 38576910,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": true,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "03f00b2b-052d-4a8f-ad13-006c9dd0bcb9/ep1",
+        "episode_id": "ep1",
+        "top_level_session": "03f00b2b-052d-4a8f-ad13-006c9dd0bcb9",
+        "file_count": 8,
+        "total_bytes": 4006984788,
+        "training_bytes_excluding_visualization_rrd": 2165734690,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": true,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "0474d134-f983-4590-9cf9-29bd1c6e33cb/ep11",
+        "episode_id": "ep11",
+        "top_level_session": "0474d134-f983-4590-9cf9-29bd1c6e33cb",
+        "file_count": 7,
+        "total_bytes": 1306123462,
+        "training_bytes_excluding_visualization_rrd": 1306123462,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "0480eba9-1eb1-4527-a980-707bca52f25d/ep10",
+        "episode_id": "ep10",
+        "top_level_session": "0480eba9-1eb1-4527-a980-707bca52f25d",
+        "file_count": 7,
+        "total_bytes": 392072447,
+        "training_bytes_excluding_visualization_rrd": 392072447,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "04d9f378-2812-4c7f-a269-1ad0bc43445f/ep14",
+        "episode_id": "ep14",
+        "top_level_session": "04d9f378-2812-4c7f-a269-1ad0bc43445f",
+        "file_count": 7,
+        "total_bytes": 372353944,
+        "training_bytes_excluding_visualization_rrd": 372353944,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "0513e241-a112-42bb-943b-dc0ab1c7c205/ep1",
+        "episode_id": "ep1",
+        "top_level_session": "0513e241-a112-42bb-943b-dc0ab1c7c205",
+        "file_count": 8,
+        "total_bytes": 4341033575,
+        "training_bytes_excluding_visualization_rrd": 2325182940,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": true,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "052ad81d-80cc-48ab-998b-d54591bada1b/ep7",
+        "episode_id": "ep7",
+        "top_level_session": "052ad81d-80cc-48ab-998b-d54591bada1b",
+        "file_count": 7,
+        "total_bytes": 175364778,
+        "training_bytes_excluding_visualization_rrd": 175364778,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "05342bb9-7bb0-4594-92fb-899da50042dc/ep4",
+        "episode_id": "ep4",
+        "top_level_session": "05342bb9-7bb0-4594-92fb-899da50042dc",
+        "file_count": 7,
+        "total_bytes": 1785501295,
+        "training_bytes_excluding_visualization_rrd": 1785501295,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "0662a9d0-cf06-4ee3-892f-f545d463ce7c/ep10",
+        "episode_id": "ep10",
+        "top_level_session": "0662a9d0-cf06-4ee3-892f-f545d463ce7c",
+        "file_count": 7,
+        "total_bytes": 1917229250,
+        "training_bytes_excluding_visualization_rrd": 1917229250,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "06640ca6-c179-41c2-8216-d7a5a47a2bc5/ep12",
+        "episode_id": "ep12",
+        "top_level_session": "06640ca6-c179-41c2-8216-d7a5a47a2bc5",
+        "file_count": 7,
+        "total_bytes": 1120737093,
+        "training_bytes_excluding_visualization_rrd": 1120737093,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "06b2632d-c169-4e42-9683-17a1140ca559/ep8",
+        "episode_id": "ep8",
+        "top_level_session": "06b2632d-c169-4e42-9683-17a1140ca559",
+        "file_count": 7,
+        "total_bytes": 77529803,
+        "training_bytes_excluding_visualization_rrd": 77529803,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "06bcd24b-5cfd-4af8-ab70-953b9172e855/ep2",
+        "episode_id": "ep2",
+        "top_level_session": "06bcd24b-5cfd-4af8-ab70-953b9172e855",
+        "file_count": 7,
+        "total_bytes": 2275259119,
+        "training_bytes_excluding_visualization_rrd": 2275259119,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "07222681-a826-4afd-9fd9-5bf841b23aa4/ep12",
+        "episode_id": "ep12",
+        "top_level_session": "07222681-a826-4afd-9fd9-5bf841b23aa4",
+        "file_count": 7,
+        "total_bytes": 1525838403,
+        "training_bytes_excluding_visualization_rrd": 1525838403,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "07f3aeee-5d64-4fd2-8450-f8baf8c239fd/ep1",
+        "episode_id": "ep1",
+        "top_level_session": "07f3aeee-5d64-4fd2-8450-f8baf8c239fd",
+        "file_count": 7,
+        "total_bytes": 2010755314,
+        "training_bytes_excluding_visualization_rrd": 2010755314,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "08684ed6-c2b0-4356-b176-0409c4c41e3c/ep1",
+        "episode_id": "ep1",
+        "top_level_session": "08684ed6-c2b0-4356-b176-0409c4c41e3c",
+        "file_count": 7,
+        "total_bytes": 1575104381,
+        "training_bytes_excluding_visualization_rrd": 1575104381,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "08c55ca5-bb8c-42f8-b0f2-2ad7d3f8a62f/ep12",
+        "episode_id": "ep12",
+        "top_level_session": "08c55ca5-bb8c-42f8-b0f2-2ad7d3f8a62f",
+        "file_count": 7,
+        "total_bytes": 907468800,
+        "training_bytes_excluding_visualization_rrd": 907468800,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "091f6d90-ad4c-485f-8bd1-5409cf32eb25/ep11",
+        "episode_id": "ep11",
+        "top_level_session": "091f6d90-ad4c-485f-8bd1-5409cf32eb25",
+        "file_count": 7,
+        "total_bytes": 190596778,
+        "training_bytes_excluding_visualization_rrd": 190596778,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "092caa55-e2db-4df5-9725-4036240b426d/ep1",
+        "episode_id": "ep1",
+        "top_level_session": "092caa55-e2db-4df5-9725-4036240b426d",
+        "file_count": 7,
+        "total_bytes": 1757951380,
+        "training_bytes_excluding_visualization_rrd": 1757951380,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "0a5009c4-292b-40c6-b9ec-7c75cf54a112/ep12",
+        "episode_id": "ep12",
+        "top_level_session": "0a5009c4-292b-40c6-b9ec-7c75cf54a112",
+        "file_count": 7,
+        "total_bytes": 1146288978,
+        "training_bytes_excluding_visualization_rrd": 1146288978,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "0addf685-b22b-4b4a-bea5-3f37367b4847/ep12",
+        "episode_id": "ep12",
+        "top_level_session": "0addf685-b22b-4b4a-bea5-3f37367b4847",
+        "file_count": 7,
+        "total_bytes": 639414472,
+        "training_bytes_excluding_visualization_rrd": 639414472,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "0b057366-8e1b-4590-97b2-9bf863395f19/ep12",
+        "episode_id": "ep12",
+        "top_level_session": "0b057366-8e1b-4590-97b2-9bf863395f19",
+        "file_count": 7,
+        "total_bytes": 1567794273,
+        "training_bytes_excluding_visualization_rrd": 1567794273,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "0bc35fc7-95ce-4f06-8801-3a6a70be6669/ep6",
+        "episode_id": "ep6",
+        "top_level_session": "0bc35fc7-95ce-4f06-8801-3a6a70be6669",
+        "file_count": 7,
+        "total_bytes": 1218185186,
+        "training_bytes_excluding_visualization_rrd": 1218185186,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "0bf99545-e6d1-4cba-8836-0f46287b1e51/ep11",
+        "episode_id": "ep11",
+        "top_level_session": "0bf99545-e6d1-4cba-8836-0f46287b1e51",
+        "file_count": 8,
+        "total_bytes": 3994383379,
+        "training_bytes_excluding_visualization_rrd": 2016509334,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": true,
+        "missing_required_files": []
+      },
+      {
+        "episode_path": "0c2cc95e-316b-4b68-86ab-f6f75002d960/ep14",
+        "episode_id": "ep14",
+        "top_level_session": "0c2cc95e-316b-4b68-86ab-f6f75002d960",
+        "file_count": 7,
+        "total_bytes": 1028155889,
+        "training_bytes_excluding_visualization_rrd": 1028155889,
+        "has_annotation": true,
+        "has_fisheye_cam0": true,
+        "video_count": 6,
+        "has_all_six_videos": true,
+        "is_degraded_valid": true,
+        "is_complete": true,
+        "has_visualization_rrd": false,
+        "missing_required_files": []
+      }
+    ]
+  },
+  "top_level_sessions_by_file_count_top_n": [
+    [
+      "5c6418f9-f8d5-4ae3-b486-3b8b6b125fbc",
+      357
+    ],
+    [
+      "5f2c3368-2e96-497e-a2b4-a2d723b84952",
+      357
+    ],
+    [
+      "53e64c93-8947-403e-ab35-3428c431aee8",
+      343
+    ],
+    [
+      "5bb01123-0071-47e4-a260-35e94ac9f301",
+      343
+    ],
+    [
+      "6d326de4-d53e-44b0-b1f5-b40485cb6ccb",
+      343
+    ],
+    [
+      "af423ccc-1d3d-4c22-b779-085fb83d7a16",
+      343
+    ],
+    [
+      "bc53027e-d29a-473f-8f18-7b524b801e40",
+      343
+    ],
+    [
+      "d6ee04f1-bf31-42da-b785-ebd8ecafef5d",
+      343
+    ],
+    [
+      "175a573b-7294-43b4-9c14-09c29ab3fef0",
+      336
+    ],
+    [
+      "2680606b-f296-4daa-baf9-7d58c90444bc",
+      336
+    ],
+    [
+      "39ea651a-31aa-4797-bfd3-c0956269b8a7",
+      336
+    ],
+    [
+      "3c7c8e38-9a20-4979-aaf7-beb6112f59cf",
+      336
+    ],
+    [
+      "5399ef86-4df9-49bc-809f-8f4f92f9e659",
+      336
+    ],
+    [
+      "573a5795-03bf-4f16-b4ce-bc09c2c56476",
+      336
+    ],
+    [
+      "5db1c415-0528-4227-9433-31bd2bd70530",
+      336
+    ],
+    [
+      "613e0f29-cc10-4d0c-b0ce-e39fe9c8a6b9",
+      336
+    ],
+    [
+      "61e1a696-2c13-496b-8301-e9618bb01e84",
+      336
+    ],
+    [
+      "6ba9651a-9f93-4803-ac95-5371590bca17",
+      336
+    ],
+    [
+      "74c5675b-88f8-4590-95dd-dca5d00fd5d6",
+      336
+    ],
+    [
+      "8870bfee-130c-44f6-81bb-5b4f01169b97",
+      336
+    ]
+  ],
+  "top_level_sessions_by_bytes_top_n": [
+    {
+      "session": "11050402-22c1-490a-98bf-98665c38beee",
+      "bytes": 141614887800,
+      "human": "131.89 GiB"
+    },
+    {
+      "session": "147b3d65-65d6-421b-8944-7d13317f0c3d",
+      "bytes": 132236122943,
+      "human": "123.15 GiB"
+    },
+    {
+      "session": "5f2c3368-2e96-497e-a2b4-a2d723b84952",
+      "bytes": 119033330023,
+      "human": "110.86 GiB"
+    },
+    {
+      "session": "5c6418f9-f8d5-4ae3-b486-3b8b6b125fbc",
+      "bytes": 117202013704,
+      "human": "109.15 GiB"
+    },
+    {
+      "session": "d6ee04f1-bf31-42da-b785-ebd8ecafef5d",
+      "bytes": 116729046575,
+      "human": "108.71 GiB"
+    },
+    {
+      "session": "3c7c8e38-9a20-4979-aaf7-beb6112f59cf",
+      "bytes": 114016325403,
+      "human": "106.19 GiB"
+    },
+    {
+      "session": "74c5675b-88f8-4590-95dd-dca5d00fd5d6",
+      "bytes": 113998503718,
+      "human": "106.17 GiB"
+    },
+    {
+      "session": "613e0f29-cc10-4d0c-b0ce-e39fe9c8a6b9",
+      "bytes": 113957412208,
+      "human": "106.13 GiB"
+    },
+    {
+      "session": "8870bfee-130c-44f6-81bb-5b4f01169b97",
+      "bytes": 113776793174,
+      "human": "105.96 GiB"
+    },
+    {
+      "session": "5399ef86-4df9-49bc-809f-8f4f92f9e659",
+      "bytes": 113463258385,
+      "human": "105.67 GiB"
+    },
+    {
+      "session": "22b50ec8-89cd-4ccb-9b22-305ddacf1ba2",
+      "bytes": 112720510126,
+      "human": "104.98 GiB"
+    },
+    {
+      "session": "5bb01123-0071-47e4-a260-35e94ac9f301",
+      "bytes": 112606165895,
+      "human": "104.87 GiB"
+    },
+    {
+      "session": "af423ccc-1d3d-4c22-b779-085fb83d7a16",
+      "bytes": 112436036819,
+      "human": "104.71 GiB"
+    },
+    {
+      "session": "53e64c93-8947-403e-ab35-3428c431aee8",
+      "bytes": 112331528460,
+      "human": "104.62 GiB"
+    },
+    {
+      "session": "573a5795-03bf-4f16-b4ce-bc09c2c56476",
+      "bytes": 112314975216,
+      "human": "104.60 GiB"
+    },
+    {
+      "session": "6d326de4-d53e-44b0-b1f5-b40485cb6ccb",
+      "bytes": 112294693749,
+      "human": "104.58 GiB"
+    },
+    {
+      "session": "bc53027e-d29a-473f-8f18-7b524b801e40",
+      "bytes": 111770819717,
+      "human": "104.09 GiB"
+    },
+    {
+      "session": "96a437c6-56ec-4dcc-b907-e3739dd9b4d7",
+      "bytes": 110818046088,
+      "human": "103.21 GiB"
+    },
+    {
+      "session": "cd7fc4d2-fc80-448d-acf9-b9c005660a84",
+      "bytes": 110602222842,
+      "human": "103.01 GiB"
+    },
+    {
+      "session": "175a573b-7294-43b4-9c14-09c29ab3fef0",
+      "bytes": 110527695330,
+      "human": "102.94 GiB"
+    }
+  ],
+  "largest_files_top_n": [
+    {
+      "path": "1ec8b052-b6df-4eac-9647-1b6b54f304ea/ep8/visualization.rrd",
+      "bytes": 2578850872,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "1ec8b052-b6df-4eac-9647-1b6b54f304ea",
+      "human": "2.40 GiB"
+    },
+    {
+      "path": "1ec8b052-b6df-4eac-9647-1b6b54f304ea/ep4/visualization.rrd",
+      "bytes": 2543441320,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "1ec8b052-b6df-4eac-9647-1b6b54f304ea",
+      "human": "2.37 GiB"
+    },
+    {
+      "path": "11050402-22c1-490a-98bf-98665c38beee/ep30/visualization.rrd",
+      "bytes": 2506989095,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "11050402-22c1-490a-98bf-98665c38beee",
+      "human": "2.33 GiB"
+    },
+    {
+      "path": "283f6efb-f845-48de-b531-b685309a1663/ep20/visualization.rrd",
+      "bytes": 2499686282,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "283f6efb-f845-48de-b531-b685309a1663",
+      "human": "2.33 GiB"
+    },
+    {
+      "path": "2b0bafa8-c898-4c00-acdc-a84d6bd0b097/ep2/visualization.rrd",
+      "bytes": 2475927974,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "2b0bafa8-c898-4c00-acdc-a84d6bd0b097",
+      "human": "2.31 GiB"
+    },
+    {
+      "path": "283f6efb-f845-48de-b531-b685309a1663/ep18/visualization.rrd",
+      "bytes": 2474528541,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "283f6efb-f845-48de-b531-b685309a1663",
+      "human": "2.30 GiB"
+    },
+    {
+      "path": "0bf99545-e6d1-4cba-8836-0f46287b1e51/ep8/visualization.rrd",
+      "bytes": 2462161091,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "0bf99545-e6d1-4cba-8836-0f46287b1e51",
+      "human": "2.29 GiB"
+    },
+    {
+      "path": "283f6efb-f845-48de-b531-b685309a1663/ep19/visualization.rrd",
+      "bytes": 2428544153,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "283f6efb-f845-48de-b531-b685309a1663",
+      "human": "2.26 GiB"
+    },
+    {
+      "path": "1ec8b052-b6df-4eac-9647-1b6b54f304ea/ep9/visualization.rrd",
+      "bytes": 2418732590,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "1ec8b052-b6df-4eac-9647-1b6b54f304ea",
+      "human": "2.25 GiB"
+    },
+    {
+      "path": "18a0fec5-7244-4b8d-9514-19f9914de2d7/ep12/visualization.rrd",
+      "bytes": 2417694485,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "18a0fec5-7244-4b8d-9514-19f9914de2d7",
+      "human": "2.25 GiB"
+    },
+    {
+      "path": "0da1f6d2-6564-4f28-b248-1d1b71f9521b/ep10/visualization.rrd",
+      "bytes": 2399712951,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "0da1f6d2-6564-4f28-b248-1d1b71f9521b",
+      "human": "2.23 GiB"
+    },
+    {
+      "path": "1ec8b052-b6df-4eac-9647-1b6b54f304ea/ep7/visualization.rrd",
+      "bytes": 2394078531,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "1ec8b052-b6df-4eac-9647-1b6b54f304ea",
+      "human": "2.23 GiB"
+    },
+    {
+      "path": "03728b87-3d0c-4d48-b73c-3c0abc1019e4/ep1/visualization.rrd",
+      "bytes": 2392460578,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "03728b87-3d0c-4d48-b73c-3c0abc1019e4",
+      "human": "2.23 GiB"
+    },
+    {
+      "path": "283f6efb-f845-48de-b531-b685309a1663/ep21/visualization.rrd",
+      "bytes": 2378794069,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "283f6efb-f845-48de-b531-b685309a1663",
+      "human": "2.22 GiB"
+    },
+    {
+      "path": "1221bd5a-af5b-4bfe-b4ff-6278a9be7960/ep1/visualization.rrd",
+      "bytes": 2377873265,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "1221bd5a-af5b-4bfe-b4ff-6278a9be7960",
+      "human": "2.21 GiB"
+    },
+    {
+      "path": "11050402-22c1-490a-98bf-98665c38beee/ep27/visualization.rrd",
+      "bytes": 2371576075,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "11050402-22c1-490a-98bf-98665c38beee",
+      "human": "2.21 GiB"
+    },
+    {
+      "path": "13ca0290-713b-43a8-a06c-305bde0cbf6c/ep14/visualization.rrd",
+      "bytes": 2349578793,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "13ca0290-713b-43a8-a06c-305bde0cbf6c",
+      "human": "2.19 GiB"
+    },
+    {
+      "path": "0bf99545-e6d1-4cba-8836-0f46287b1e51/ep4/visualization.rrd",
+      "bytes": 2345803219,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "0bf99545-e6d1-4cba-8836-0f46287b1e51",
+      "human": "2.18 GiB"
+    },
+    {
+      "path": "22af1ff1-eaa9-4268-b96a-81a9fa35fe93/ep4/visualization.rrd",
+      "bytes": 2344707770,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "22af1ff1-eaa9-4268-b96a-81a9fa35fe93",
+      "human": "2.18 GiB"
+    },
+    {
+      "path": "2d5d1afe-5e20-41a3-bcd9-a18a01e7cf17/ep6/visualization.rrd",
+      "bytes": 2344172298,
+      "extension": ".rrd",
+      "basename": "visualization.rrd",
+      "top_level": "2d5d1afe-5e20-41a3-bcd9-a18a01e7cf17",
+      "human": "2.18 GiB"
+    }
+  ],
+  "smallest_annotation_files_top_n": [
+    {
+      "path": "9cecac72-8874-4b97-9541-18d4858f8e43/ep10/annotation.hdf5",
+      "bytes": 6687192,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "9cecac72-8874-4b97-9541-18d4858f8e43",
+      "human": "6.38 MiB"
+    },
+    {
+      "path": "cdc1ae12-a460-48ac-a892-7d314095c4b1/ep23/annotation.hdf5",
+      "bytes": 6687256,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "cdc1ae12-a460-48ac-a892-7d314095c4b1",
+      "human": "6.38 MiB"
+    },
+    {
+      "path": "10282b64-a955-461e-9ef9-a1ddf8dc619a/ep5/annotation.hdf5",
+      "bytes": 6706448,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "10282b64-a955-461e-9ef9-a1ddf8dc619a",
+      "human": "6.40 MiB"
+    },
+    {
+      "path": "160e1872-b7eb-4f59-9da7-b530c44ca623/ep12/annotation.hdf5",
+      "bytes": 9012280,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "160e1872-b7eb-4f59-9da7-b530c44ca623",
+      "human": "8.59 MiB"
+    },
+    {
+      "path": "7b4a43d3-8e76-4533-8502-60c00737f54b/ep6/annotation.hdf5",
+      "bytes": 9675640,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "7b4a43d3-8e76-4533-8502-60c00737f54b",
+      "human": "9.23 MiB"
+    },
+    {
+      "path": "9b372f31-f9a1-408e-b687-1637ecf008dc/ep46/annotation.hdf5",
+      "bytes": 10004824,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "9b372f31-f9a1-408e-b687-1637ecf008dc",
+      "human": "9.54 MiB"
+    },
+    {
+      "path": "853566a8-21a5-450c-928a-7b1975f1a43b/ep13/annotation.hdf5",
+      "bytes": 11663732,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "853566a8-21a5-450c-928a-7b1975f1a43b",
+      "human": "11.12 MiB"
+    },
+    {
+      "path": "b245fad0-4875-4f9c-9e17-f7e2db70fa66/ep1/annotation.hdf5",
+      "bytes": 12989948,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "b245fad0-4875-4f9c-9e17-f7e2db70fa66",
+      "human": "12.39 MiB"
+    },
+    {
+      "path": "1ec8b052-b6df-4eac-9647-1b6b54f304ea/ep1/annotation.hdf5",
+      "bytes": 15981784,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "1ec8b052-b6df-4eac-9647-1b6b54f304ea",
+      "human": "15.24 MiB"
+    },
+    {
+      "path": "c8cdf6b0-7cf9-412c-8506-31a78bd71ff3/ep5/annotation.hdf5",
+      "bytes": 17633400,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "c8cdf6b0-7cf9-412c-8506-31a78bd71ff3",
+      "human": "16.82 MiB"
+    },
+    {
+      "path": "1277aeda-4ad6-461f-a365-802799c16ae8/ep12/annotation.hdf5",
+      "bytes": 17971320,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "1277aeda-4ad6-461f-a365-802799c16ae8",
+      "human": "17.14 MiB"
+    },
+    {
+      "path": "d2639d6b-2c21-4e0d-8931-8cfbe8d38e88/ep1/annotation.hdf5",
+      "bytes": 22281178,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "d2639d6b-2c21-4e0d-8931-8cfbe8d38e88",
+      "human": "21.25 MiB"
+    },
+    {
+      "path": "baa3508a-0eb2-4b77-bf94-77ca528bd857/ep13/annotation.hdf5",
+      "bytes": 31599178,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "baa3508a-0eb2-4b77-bf94-77ca528bd857",
+      "human": "30.14 MiB"
+    },
+    {
+      "path": "0376a3fd-52ac-4de6-afd1-4505e235010a/ep1/annotation.hdf5",
+      "bytes": 33565174,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "0376a3fd-52ac-4de6-afd1-4505e235010a",
+      "human": "32.01 MiB"
+    },
+    {
+      "path": "55c11106-2a24-4874-a8c1-15fb7527fad7/ep10/annotation.hdf5",
+      "bytes": 39542596,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "55c11106-2a24-4874-a8c1-15fb7527fad7",
+      "human": "37.71 MiB"
+    },
+    {
+      "path": "836436e6-8bb0-49ea-919a-42ec2ce7023d/ep24/annotation.hdf5",
+      "bytes": 41243472,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "836436e6-8bb0-49ea-919a-42ec2ce7023d",
+      "human": "39.33 MiB"
+    },
+    {
+      "path": "71f1e918-8e86-460c-beda-af8d4063a238/ep8/annotation.hdf5",
+      "bytes": 41918096,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "71f1e918-8e86-460c-beda-af8d4063a238",
+      "human": "39.98 MiB"
+    },
+    {
+      "path": "00986a68-7248-492a-befc-daae0c3f9866/ep2/annotation.hdf5",
+      "bytes": 51141832,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "00986a68-7248-492a-befc-daae0c3f9866",
+      "human": "48.77 MiB"
+    },
+    {
+      "path": "3789a6ca-333d-4a5f-8a77-26d95411541a/ep3/annotation.hdf5",
+      "bytes": 52474936,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "3789a6ca-333d-4a5f-8a77-26d95411541a",
+      "human": "50.04 MiB"
+    },
+    {
+      "path": "c22ffd3f-8f92-45ad-86fd-af657d078d37/ep1/annotation.hdf5",
+      "bytes": 54132652,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "c22ffd3f-8f92-45ad-86fd-af657d078d37",
+      "human": "51.62 MiB"
+    }
+  ],
+  "median_annotation_files_top_n": [
+    {
+      "path": "999100c1-b488-4ed1-a580-c0487f308e5d/ep7/annotation.hdf5",
+      "bytes": 1969457752,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "999100c1-b488-4ed1-a580-c0487f308e5d",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "bc53027e-d29a-473f-8f18-7b524b801e40/ep38/annotation.hdf5",
+      "bytes": 1969456504,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "bc53027e-d29a-473f-8f18-7b524b801e40",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "4436aa24-8cb6-4f01-874f-866831080317/ep4/annotation.hdf5",
+      "bytes": 1969454808,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "4436aa24-8cb6-4f01-874f-866831080317",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "86b69577-0826-45bc-bffd-f847c3b03970/ep7/annotation.hdf5",
+      "bytes": 1969454416,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "86b69577-0826-45bc-bffd-f847c3b03970",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "5399ef86-4df9-49bc-809f-8f4f92f9e659/ep33/annotation.hdf5",
+      "bytes": 1969451920,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "5399ef86-4df9-49bc-809f-8f4f92f9e659",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "c115d9d9-8f1c-4c7e-ac9d-d14193a35b4e/ep22/annotation.hdf5",
+      "bytes": 1969467168,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "c115d9d9-8f1c-4c7e-ac9d-d14193a35b4e",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "092caa55-e2db-4df5-9725-4036240b426d/ep45/annotation.hdf5",
+      "bytes": 1969445184,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "092caa55-e2db-4df5-9725-4036240b426d",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "186ffc96-c19d-48a8-a157-31967ad6026e/ep10/annotation.hdf5",
+      "bytes": 1969470552,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "186ffc96-c19d-48a8-a157-31967ad6026e",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "1d682b55-298c-4aa5-8f8b-aa7d0bb7dd07/ep16/annotation.hdf5",
+      "bytes": 1969475560,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "1d682b55-298c-4aa5-8f8b-aa7d0bb7dd07",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "cf0c4437-edb2-4455-9746-2a759801116f/ep6/annotation.hdf5",
+      "bytes": 1969436940,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "cf0c4437-edb2-4455-9746-2a759801116f",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "5399ef86-4df9-49bc-809f-8f4f92f9e659/ep44/annotation.hdf5",
+      "bytes": 1969436808,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "5399ef86-4df9-49bc-809f-8f4f92f9e659",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "92853fad-7f7b-4b7d-a1b2-ad32148a3828/ep21/annotation.hdf5",
+      "bytes": 1969433896,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "92853fad-7f7b-4b7d-a1b2-ad32148a3828",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "a6d5e242-2b18-435a-b8c9-b77590dd5c0d/ep30/annotation.hdf5",
+      "bytes": 1969433416,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "a6d5e242-2b18-435a-b8c9-b77590dd5c0d",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "86b69577-0826-45bc-bffd-f847c3b03970/ep6/annotation.hdf5",
+      "bytes": 1969483872,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "86b69577-0826-45bc-bffd-f847c3b03970",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "a9dfa4f3-853f-4a8f-9041-0dfd72fdeb21/ep3/annotation.hdf5",
+      "bytes": 1969418704,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "a9dfa4f3-853f-4a8f-9041-0dfd72fdeb21",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "8533707e-d7e0-4d1a-99d1-fa154a8e62ad/ep46/annotation.hdf5",
+      "bytes": 1969417872,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "8533707e-d7e0-4d1a-99d1-fa154a8e62ad",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "2aee3428-f462-49e6-95c4-b803c1803e0b/ep17/annotation.hdf5",
+      "bytes": 1969408472,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "2aee3428-f462-49e6-95c4-b803c1803e0b",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "6ecaa9e0-26c3-44d7-bb54-05e95f71f1fb/ep8/annotation.hdf5",
+      "bytes": 1969402224,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "6ecaa9e0-26c3-44d7-bb54-05e95f71f1fb",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "a66b1d2c-51a4-4857-a7d7-3ff66677edd8/ep3/annotation.hdf5",
+      "bytes": 1969521800,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "a66b1d2c-51a4-4857-a7d7-3ff66677edd8",
+      "human": "1.83 GiB"
+    },
+    {
+      "path": "14555020-82ca-423d-bff1-d5a4305e6817/ep6/annotation.hdf5",
+      "bytes": 1969526304,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "14555020-82ca-423d-bff1-d5a4305e6817",
+      "human": "1.83 GiB"
+    }
+  ],
+  "largest_annotation_files_top_n": [
+    {
+      "path": "9507a878-279c-4002-aa01-302f012e59c3/ep48/annotation.hdf5",
+      "bytes": 2000432556,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "9507a878-279c-4002-aa01-302f012e59c3",
+      "human": "1.86 GiB"
+    },
+    {
+      "path": "3eb42d11-2121-4326-abb5-c2a534862945/ep5/annotation.hdf5",
+      "bytes": 1990526432,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "3eb42d11-2121-4326-abb5-c2a534862945",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "2db1494c-ac6f-42f6-a5c5-5f144f589a66/ep11/annotation.hdf5",
+      "bytes": 1990490088,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "2db1494c-ac6f-42f6-a5c5-5f144f589a66",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "4b0d50dc-f518-47ea-aa77-f4f889910b49/ep25/annotation.hdf5",
+      "bytes": 1990485072,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "4b0d50dc-f518-47ea-aa77-f4f889910b49",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "55561197-fa16-4ce6-8b9b-7a9a1118007f/ep17/annotation.hdf5",
+      "bytes": 1990473712,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "55561197-fa16-4ce6-8b9b-7a9a1118007f",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "4b0d50dc-f518-47ea-aa77-f4f889910b49/ep27/annotation.hdf5",
+      "bytes": 1990411632,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "4b0d50dc-f518-47ea-aa77-f4f889910b49",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "cc73ce3a-d048-4f8a-8f53-c6baa5ec3631/ep9/annotation.hdf5",
+      "bytes": 1990363824,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "cc73ce3a-d048-4f8a-8f53-c6baa5ec3631",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "78958ee3-d8a6-4933-ac2b-8708dd279418/ep1/annotation.hdf5",
+      "bytes": 1990345800,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "78958ee3-d8a6-4933-ac2b-8708dd279418",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "43340856-73ac-4f34-b924-eea5ddab1e2f/ep7/annotation.hdf5",
+      "bytes": 1990328872,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "43340856-73ac-4f34-b924-eea5ddab1e2f",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "910e4f3f-8056-47cc-98cb-75beabd6296e/ep1/annotation.hdf5",
+      "bytes": 1990324584,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "910e4f3f-8056-47cc-98cb-75beabd6296e",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "3a020e9b-9fb1-4300-b56d-263e268899ff/ep1/annotation.hdf5",
+      "bytes": 1990308216,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "3a020e9b-9fb1-4300-b56d-263e268899ff",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "a14256c7-b511-4d9d-ab7e-af6308d47f9c/ep9/annotation.hdf5",
+      "bytes": 1990302888,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "a14256c7-b511-4d9d-ab7e-af6308d47f9c",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "cd4e5522-0d6f-4e3a-bc2b-96eeb3be38f5/ep4/annotation.hdf5",
+      "bytes": 1990297752,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "cd4e5522-0d6f-4e3a-bc2b-96eeb3be38f5",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "c038defe-faa6-4fe9-aef6-84bd93726d1c/ep11/annotation.hdf5",
+      "bytes": 1990295024,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "c038defe-faa6-4fe9-aef6-84bd93726d1c",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "af8e9345-011f-4c97-a78e-ecb4a268220f/ep1/annotation.hdf5",
+      "bytes": 1990292352,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "af8e9345-011f-4c97-a78e-ecb4a268220f",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "c0283b1c-11b5-472b-ae48-b91f33966e34/ep5/annotation.hdf5",
+      "bytes": 1990291216,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "c0283b1c-11b5-472b-ae48-b91f33966e34",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "c038defe-faa6-4fe9-aef6-84bd93726d1c/ep3/annotation.hdf5",
+      "bytes": 1990291000,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "c038defe-faa6-4fe9-aef6-84bd93726d1c",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "1893387e-6d02-46c1-964b-70583ac9a329/ep3/annotation.hdf5",
+      "bytes": 1990288440,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "1893387e-6d02-46c1-964b-70583ac9a329",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "67b88040-2471-4e6f-8a9e-e7ac043d2fd5/ep2/annotation.hdf5",
+      "bytes": 1990280016,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "67b88040-2471-4e6f-8a9e-e7ac043d2fd5",
+      "human": "1.85 GiB"
+    },
+    {
+      "path": "95543d03-b296-40d0-82e9-230b6afa8805/ep2/annotation.hdf5",
+      "bytes": 1990279472,
+      "extension": ".hdf5",
+      "basename": "annotation.hdf5",
+      "top_level": "95543d03-b296-40d0-82e9-230b6afa8805",
+      "human": "1.85 GiB"
+    }
+  ],
+  "largest_episode_folders_top_n": [
+    {
+      "episode_path": "283f6efb-f845-48de-b531-b685309a1663/ep20",
+      "episode_id": "ep20",
+      "top_level_session": "283f6efb-f845-48de-b531-b685309a1663",
+      "file_count": 8,
+      "total_bytes": 5108045286,
+      "training_bytes_excluding_visualization_rrd": 2608359004,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.76 GiB",
+      "training_human": "2.43 GiB"
+    },
+    {
+      "episode_path": "283f6efb-f845-48de-b531-b685309a1663/ep18",
+      "episode_id": "ep18",
+      "top_level_session": "283f6efb-f845-48de-b531-b685309a1663",
+      "file_count": 8,
+      "total_bytes": 5025036054,
+      "training_bytes_excluding_visualization_rrd": 2550507513,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.68 GiB",
+      "training_human": "2.38 GiB"
+    },
+    {
+      "episode_path": "283f6efb-f845-48de-b531-b685309a1663/ep19",
+      "episode_id": "ep19",
+      "top_level_session": "283f6efb-f845-48de-b531-b685309a1663",
+      "file_count": 8,
+      "total_bytes": 5011084290,
+      "training_bytes_excluding_visualization_rrd": 2582540137,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.67 GiB",
+      "training_human": "2.41 GiB"
+    },
+    {
+      "episode_path": "1ec8b052-b6df-4eac-9647-1b6b54f304ea/ep4",
+      "episode_id": "ep4",
+      "top_level_session": "1ec8b052-b6df-4eac-9647-1b6b54f304ea",
+      "file_count": 8,
+      "total_bytes": 4991004889,
+      "training_bytes_excluding_visualization_rrd": 2447563569,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.65 GiB",
+      "training_human": "2.28 GiB"
+    },
+    {
+      "episode_path": "1ec8b052-b6df-4eac-9647-1b6b54f304ea/ep8",
+      "episode_id": "ep8",
+      "top_level_session": "1ec8b052-b6df-4eac-9647-1b6b54f304ea",
+      "file_count": 8,
+      "total_bytes": 4990865729,
+      "training_bytes_excluding_visualization_rrd": 2412014857,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.65 GiB",
+      "training_human": "2.25 GiB"
+    },
+    {
+      "episode_path": "0da1f6d2-6564-4f28-b248-1d1b71f9521b/ep10",
+      "episode_id": "ep10",
+      "top_level_session": "0da1f6d2-6564-4f28-b248-1d1b71f9521b",
+      "file_count": 8,
+      "total_bytes": 4943407733,
+      "training_bytes_excluding_visualization_rrd": 2543694782,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.60 GiB",
+      "training_human": "2.37 GiB"
+    },
+    {
+      "episode_path": "1ec8b052-b6df-4eac-9647-1b6b54f304ea/ep9",
+      "episode_id": "ep9",
+      "top_level_session": "1ec8b052-b6df-4eac-9647-1b6b54f304ea",
+      "file_count": 8,
+      "total_bytes": 4906021996,
+      "training_bytes_excluding_visualization_rrd": 2487289406,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.57 GiB",
+      "training_human": "2.32 GiB"
+    },
+    {
+      "episode_path": "0bf99545-e6d1-4cba-8836-0f46287b1e51/ep8",
+      "episode_id": "ep8",
+      "top_level_session": "0bf99545-e6d1-4cba-8836-0f46287b1e51",
+      "file_count": 8,
+      "total_bytes": 4884738304,
+      "training_bytes_excluding_visualization_rrd": 2422577213,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.55 GiB",
+      "training_human": "2.26 GiB"
+    },
+    {
+      "episode_path": "11050402-22c1-490a-98bf-98665c38beee/ep28",
+      "episode_id": "ep28",
+      "top_level_session": "11050402-22c1-490a-98bf-98665c38beee",
+      "file_count": 8,
+      "total_bytes": 4877726681,
+      "training_bytes_excluding_visualization_rrd": 2564143652,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.54 GiB",
+      "training_human": "2.39 GiB"
+    },
+    {
+      "episode_path": "2b0bafa8-c898-4c00-acdc-a84d6bd0b097/ep2",
+      "episode_id": "ep2",
+      "top_level_session": "2b0bafa8-c898-4c00-acdc-a84d6bd0b097",
+      "file_count": 8,
+      "total_bytes": 4877124248,
+      "training_bytes_excluding_visualization_rrd": 2401196274,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.54 GiB",
+      "training_human": "2.24 GiB"
+    },
+    {
+      "episode_path": "11050402-22c1-490a-98bf-98665c38beee/ep30",
+      "episode_id": "ep30",
+      "top_level_session": "11050402-22c1-490a-98bf-98665c38beee",
+      "file_count": 8,
+      "total_bytes": 4866220477,
+      "training_bytes_excluding_visualization_rrd": 2359231382,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.53 GiB",
+      "training_human": "2.20 GiB"
+    },
+    {
+      "episode_path": "283f6efb-f845-48de-b531-b685309a1663/ep21",
+      "episode_id": "ep21",
+      "top_level_session": "283f6efb-f845-48de-b531-b685309a1663",
+      "file_count": 8,
+      "total_bytes": 4863863187,
+      "training_bytes_excluding_visualization_rrd": 2485069118,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.53 GiB",
+      "training_human": "2.31 GiB"
+    },
+    {
+      "episode_path": "0da1f6d2-6564-4f28-b248-1d1b71f9521b/ep3",
+      "episode_id": "ep3",
+      "top_level_session": "0da1f6d2-6564-4f28-b248-1d1b71f9521b",
+      "file_count": 8,
+      "total_bytes": 4845103054,
+      "training_bytes_excluding_visualization_rrd": 2514780442,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.51 GiB",
+      "training_human": "2.34 GiB"
+    },
+    {
+      "episode_path": "03728b87-3d0c-4d48-b73c-3c0abc1019e4/ep1",
+      "episode_id": "ep1",
+      "top_level_session": "03728b87-3d0c-4d48-b73c-3c0abc1019e4",
+      "file_count": 8,
+      "total_bytes": 4814419934,
+      "training_bytes_excluding_visualization_rrd": 2421959356,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.48 GiB",
+      "training_human": "2.26 GiB"
+    },
+    {
+      "episode_path": "1221bd5a-af5b-4bfe-b4ff-6278a9be7960/ep1",
+      "episode_id": "ep1",
+      "top_level_session": "1221bd5a-af5b-4bfe-b4ff-6278a9be7960",
+      "file_count": 8,
+      "total_bytes": 4814064804,
+      "training_bytes_excluding_visualization_rrd": 2436191539,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.48 GiB",
+      "training_human": "2.27 GiB"
+    },
+    {
+      "episode_path": "11050402-22c1-490a-98bf-98665c38beee/ep23",
+      "episode_id": "ep23",
+      "top_level_session": "11050402-22c1-490a-98bf-98665c38beee",
+      "file_count": 8,
+      "total_bytes": 4810631153,
+      "training_bytes_excluding_visualization_rrd": 2523831780,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.48 GiB",
+      "training_human": "2.35 GiB"
+    },
+    {
+      "episode_path": "2d5d1afe-5e20-41a3-bcd9-a18a01e7cf17/ep6",
+      "episode_id": "ep6",
+      "top_level_session": "2d5d1afe-5e20-41a3-bcd9-a18a01e7cf17",
+      "file_count": 8,
+      "total_bytes": 4802535293,
+      "training_bytes_excluding_visualization_rrd": 2458362995,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.47 GiB",
+      "training_human": "2.29 GiB"
+    },
+    {
+      "episode_path": "335138bc-0a1f-4c49-b7b5-5fa38691b725/ep8",
+      "episode_id": "ep8",
+      "top_level_session": "335138bc-0a1f-4c49-b7b5-5fa38691b725",
+      "file_count": 8,
+      "total_bytes": 4779462563,
+      "training_bytes_excluding_visualization_rrd": 2494451823,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.45 GiB",
+      "training_human": "2.32 GiB"
+    },
+    {
+      "episode_path": "335138bc-0a1f-4c49-b7b5-5fa38691b725/ep9",
+      "episode_id": "ep9",
+      "top_level_session": "335138bc-0a1f-4c49-b7b5-5fa38691b725",
+      "file_count": 8,
+      "total_bytes": 4777244261,
+      "training_bytes_excluding_visualization_rrd": 2484215500,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.45 GiB",
+      "training_human": "2.31 GiB"
+    },
+    {
+      "episode_path": "1ec8b052-b6df-4eac-9647-1b6b54f304ea/ep7",
+      "episode_id": "ep7",
+      "top_level_session": "1ec8b052-b6df-4eac-9647-1b6b54f304ea",
+      "file_count": 8,
+      "total_bytes": 4772490529,
+      "training_bytes_excluding_visualization_rrd": 2378411998,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "4.44 GiB",
+      "training_human": "2.22 GiB"
+    }
+  ],
+  "smallest_complete_episode_training_folders_top_n": [
+    {
+      "episode_path": "9cecac72-8874-4b97-9541-18d4858f8e43/ep10",
+      "episode_id": "ep10",
+      "top_level_session": "9cecac72-8874-4b97-9541-18d4858f8e43",
+      "file_count": 7,
+      "total_bytes": 8158968,
+      "training_bytes_excluding_visualization_rrd": 8158968,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "7.78 MiB",
+      "training_human": "7.78 MiB"
+    },
+    {
+      "episode_path": "cdc1ae12-a460-48ac-a892-7d314095c4b1/ep23",
+      "episode_id": "ep23",
+      "top_level_session": "cdc1ae12-a460-48ac-a892-7d314095c4b1",
+      "file_count": 7,
+      "total_bytes": 8427985,
+      "training_bytes_excluding_visualization_rrd": 8427985,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "8.04 MiB",
+      "training_human": "8.04 MiB"
+    },
+    {
+      "episode_path": "10282b64-a955-461e-9ef9-a1ddf8dc619a/ep5",
+      "episode_id": "ep5",
+      "top_level_session": "10282b64-a955-461e-9ef9-a1ddf8dc619a",
+      "file_count": 7,
+      "total_bytes": 8702180,
+      "training_bytes_excluding_visualization_rrd": 8702180,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "8.30 MiB",
+      "training_human": "8.30 MiB"
+    },
+    {
+      "episode_path": "160e1872-b7eb-4f59-9da7-b530c44ca623/ep12",
+      "episode_id": "ep12",
+      "top_level_session": "160e1872-b7eb-4f59-9da7-b530c44ca623",
+      "file_count": 7,
+      "total_bytes": 10991484,
+      "training_bytes_excluding_visualization_rrd": 10991484,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "10.48 MiB",
+      "training_human": "10.48 MiB"
+    },
+    {
+      "episode_path": "7b4a43d3-8e76-4533-8502-60c00737f54b/ep6",
+      "episode_id": "ep6",
+      "top_level_session": "7b4a43d3-8e76-4533-8502-60c00737f54b",
+      "file_count": 7,
+      "total_bytes": 11851192,
+      "training_bytes_excluding_visualization_rrd": 11851192,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "11.30 MiB",
+      "training_human": "11.30 MiB"
+    },
+    {
+      "episode_path": "9b372f31-f9a1-408e-b687-1637ecf008dc/ep46",
+      "episode_id": "ep46",
+      "top_level_session": "9b372f31-f9a1-408e-b687-1637ecf008dc",
+      "file_count": 7,
+      "total_bytes": 12695610,
+      "training_bytes_excluding_visualization_rrd": 12695610,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "12.11 MiB",
+      "training_human": "12.11 MiB"
+    },
+    {
+      "episode_path": "853566a8-21a5-450c-928a-7b1975f1a43b/ep13",
+      "episode_id": "ep13",
+      "top_level_session": "853566a8-21a5-450c-928a-7b1975f1a43b",
+      "file_count": 7,
+      "total_bytes": 14065375,
+      "training_bytes_excluding_visualization_rrd": 14065375,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "13.41 MiB",
+      "training_human": "13.41 MiB"
+    },
+    {
+      "episode_path": "b245fad0-4875-4f9c-9e17-f7e2db70fa66/ep1",
+      "episode_id": "ep1",
+      "top_level_session": "b245fad0-4875-4f9c-9e17-f7e2db70fa66",
+      "file_count": 7,
+      "total_bytes": 15186869,
+      "training_bytes_excluding_visualization_rrd": 15186869,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "14.48 MiB",
+      "training_human": "14.48 MiB"
+    },
+    {
+      "episode_path": "1ec8b052-b6df-4eac-9647-1b6b54f304ea/ep1",
+      "episode_id": "ep1",
+      "top_level_session": "1ec8b052-b6df-4eac-9647-1b6b54f304ea",
+      "file_count": 8,
+      "total_bytes": 40792860,
+      "training_bytes_excluding_visualization_rrd": 19010145,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "38.90 MiB",
+      "training_human": "18.13 MiB"
+    },
+    {
+      "episode_path": "c8cdf6b0-7cf9-412c-8506-31a78bd71ff3/ep5",
+      "episode_id": "ep5",
+      "top_level_session": "c8cdf6b0-7cf9-412c-8506-31a78bd71ff3",
+      "file_count": 7,
+      "total_bytes": 20891056,
+      "training_bytes_excluding_visualization_rrd": 20891056,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "19.92 MiB",
+      "training_human": "19.92 MiB"
+    },
+    {
+      "episode_path": "1277aeda-4ad6-461f-a365-802799c16ae8/ep12",
+      "episode_id": "ep12",
+      "top_level_session": "1277aeda-4ad6-461f-a365-802799c16ae8",
+      "file_count": 7,
+      "total_bytes": 21280280,
+      "training_bytes_excluding_visualization_rrd": 21280280,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "20.29 MiB",
+      "training_human": "20.29 MiB"
+    },
+    {
+      "episode_path": "d2639d6b-2c21-4e0d-8931-8cfbe8d38e88/ep1",
+      "episode_id": "ep1",
+      "top_level_session": "d2639d6b-2c21-4e0d-8931-8cfbe8d38e88",
+      "file_count": 7,
+      "total_bytes": 25758113,
+      "training_bytes_excluding_visualization_rrd": 25758113,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "24.56 MiB",
+      "training_human": "24.56 MiB"
+    },
+    {
+      "episode_path": "baa3508a-0eb2-4b77-bf94-77ca528bd857/ep13",
+      "episode_id": "ep13",
+      "top_level_session": "baa3508a-0eb2-4b77-bf94-77ca528bd857",
+      "file_count": 7,
+      "total_bytes": 38362634,
+      "training_bytes_excluding_visualization_rrd": 38362634,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "36.59 MiB",
+      "training_human": "36.59 MiB"
+    },
+    {
+      "episode_path": "0376a3fd-52ac-4de6-afd1-4505e235010a/ep1",
+      "episode_id": "ep1",
+      "top_level_session": "0376a3fd-52ac-4de6-afd1-4505e235010a",
+      "file_count": 8,
+      "total_bytes": 85280549,
+      "training_bytes_excluding_visualization_rrd": 38576910,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "81.33 MiB",
+      "training_human": "36.79 MiB"
+    },
+    {
+      "episode_path": "55c11106-2a24-4874-a8c1-15fb7527fad7/ep10",
+      "episode_id": "ep10",
+      "top_level_session": "55c11106-2a24-4874-a8c1-15fb7527fad7",
+      "file_count": 7,
+      "total_bytes": 47390032,
+      "training_bytes_excluding_visualization_rrd": 47390032,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "45.19 MiB",
+      "training_human": "45.19 MiB"
+    },
+    {
+      "episode_path": "836436e6-8bb0-49ea-919a-42ec2ce7023d/ep24",
+      "episode_id": "ep24",
+      "top_level_session": "836436e6-8bb0-49ea-919a-42ec2ce7023d",
+      "file_count": 7,
+      "total_bytes": 50740055,
+      "training_bytes_excluding_visualization_rrd": 50740055,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "48.39 MiB",
+      "training_human": "48.39 MiB"
+    },
+    {
+      "episode_path": "71f1e918-8e86-460c-beda-af8d4063a238/ep8",
+      "episode_id": "ep8",
+      "top_level_session": "71f1e918-8e86-460c-beda-af8d4063a238",
+      "file_count": 7,
+      "total_bytes": 52666111,
+      "training_bytes_excluding_visualization_rrd": 52666111,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "50.23 MiB",
+      "training_human": "50.23 MiB"
+    },
+    {
+      "episode_path": "00986a68-7248-492a-befc-daae0c3f9866/ep2",
+      "episode_id": "ep2",
+      "top_level_session": "00986a68-7248-492a-befc-daae0c3f9866",
+      "file_count": 8,
+      "total_bytes": 107130526,
+      "training_bytes_excluding_visualization_rrd": 58620463,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": true,
+      "missing_required_files": [],
+      "total_human": "102.17 MiB",
+      "training_human": "55.90 MiB"
+    },
+    {
+      "episode_path": "c22ffd3f-8f92-45ad-86fd-af657d078d37/ep1",
+      "episode_id": "ep1",
+      "top_level_session": "c22ffd3f-8f92-45ad-86fd-af657d078d37",
+      "file_count": 7,
+      "total_bytes": 63069086,
+      "training_bytes_excluding_visualization_rrd": 63069086,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "60.15 MiB",
+      "training_human": "60.15 MiB"
+    },
+    {
+      "episode_path": "3789a6ca-333d-4a5f-8a77-26d95411541a/ep3",
+      "episode_id": "ep3",
+      "top_level_session": "3789a6ca-333d-4a5f-8a77-26d95411541a",
+      "file_count": 7,
+      "total_bytes": 63954531,
+      "training_bytes_excluding_visualization_rrd": 63954531,
+      "has_annotation": true,
+      "has_fisheye_cam0": true,
+      "video_count": 6,
+      "has_all_six_videos": true,
+      "is_degraded_valid": true,
+      "is_complete": true,
+      "has_visualization_rrd": false,
+      "missing_required_files": [],
+      "total_human": "60.99 MiB",
+      "training_human": "60.99 MiB"
+    }
+  ],
+  "download_recommendation": {
+    "metadata_only_audit_requires_training_host": false,
+    "recommended_download_host": "Any HF-reachable relay host with enough scratch storage; transfer staged episodes to the training host if that host cannot access Hugging Face.",
+    "training_host_role": "training and local manifest validation after data is staged",
+    "exclude_files": [
+      "visualization.rrd"
+    ],
+    "minimum_pilot": "32 complete episodes from different top-level sessions if storage permits; degraded-valid episodes only for loader smoke tests."
+  }
+}

results/omni_finetune/xperience10m_128_episode_download_files.txt

@@ -0,0 +1,896 @@
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c/ep1/annotation.hdf5
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c/ep1/fisheye_cam0.mp4
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c/ep1/fisheye_cam1.mp4
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c/ep1/fisheye_cam2.mp4
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c/ep1/fisheye_cam3.mp4
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c/ep1/stereo_left.mp4
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c/ep1/stereo_right.mp4
+27c9fc42-2bb4-4737-b09c-08d2dd88aed4/ep4/annotation.hdf5
+27c9fc42-2bb4-4737-b09c-08d2dd88aed4/ep4/fisheye_cam0.mp4
+27c9fc42-2bb4-4737-b09c-08d2dd88aed4/ep4/fisheye_cam1.mp4
+27c9fc42-2bb4-4737-b09c-08d2dd88aed4/ep4/fisheye_cam2.mp4
+27c9fc42-2bb4-4737-b09c-08d2dd88aed4/ep4/fisheye_cam3.mp4
+27c9fc42-2bb4-4737-b09c-08d2dd88aed4/ep4/stereo_left.mp4
+27c9fc42-2bb4-4737-b09c-08d2dd88aed4/ep4/stereo_right.mp4
+363abff7-e5ce-425e-a85c-5397186aefd2/ep3/annotation.hdf5
+363abff7-e5ce-425e-a85c-5397186aefd2/ep3/fisheye_cam0.mp4
+363abff7-e5ce-425e-a85c-5397186aefd2/ep3/fisheye_cam1.mp4
+363abff7-e5ce-425e-a85c-5397186aefd2/ep3/fisheye_cam2.mp4
+363abff7-e5ce-425e-a85c-5397186aefd2/ep3/fisheye_cam3.mp4
+363abff7-e5ce-425e-a85c-5397186aefd2/ep3/stereo_left.mp4
+363abff7-e5ce-425e-a85c-5397186aefd2/ep3/stereo_right.mp4
+0e4f59cc-c232-4ad0-9d0c-fed6e026422a/ep3/annotation.hdf5
+0e4f59cc-c232-4ad0-9d0c-fed6e026422a/ep3/fisheye_cam0.mp4
+0e4f59cc-c232-4ad0-9d0c-fed6e026422a/ep3/fisheye_cam1.mp4
+0e4f59cc-c232-4ad0-9d0c-fed6e026422a/ep3/fisheye_cam2.mp4
+0e4f59cc-c232-4ad0-9d0c-fed6e026422a/ep3/fisheye_cam3.mp4
+0e4f59cc-c232-4ad0-9d0c-fed6e026422a/ep3/stereo_left.mp4
+0e4f59cc-c232-4ad0-9d0c-fed6e026422a/ep3/stereo_right.mp4
+705435da-879d-456e-a28e-f15f86e75027/ep3/annotation.hdf5
+705435da-879d-456e-a28e-f15f86e75027/ep3/fisheye_cam0.mp4
+705435da-879d-456e-a28e-f15f86e75027/ep3/fisheye_cam1.mp4
+705435da-879d-456e-a28e-f15f86e75027/ep3/fisheye_cam2.mp4
+705435da-879d-456e-a28e-f15f86e75027/ep3/fisheye_cam3.mp4
+705435da-879d-456e-a28e-f15f86e75027/ep3/stereo_left.mp4
+705435da-879d-456e-a28e-f15f86e75027/ep3/stereo_right.mp4
+65adf646-cb23-430e-9583-12482dd451a1/ep8/annotation.hdf5
+65adf646-cb23-430e-9583-12482dd451a1/ep8/fisheye_cam0.mp4
+65adf646-cb23-430e-9583-12482dd451a1/ep8/fisheye_cam1.mp4
+65adf646-cb23-430e-9583-12482dd451a1/ep8/fisheye_cam2.mp4
+65adf646-cb23-430e-9583-12482dd451a1/ep8/fisheye_cam3.mp4
+65adf646-cb23-430e-9583-12482dd451a1/ep8/stereo_left.mp4
+65adf646-cb23-430e-9583-12482dd451a1/ep8/stereo_right.mp4
+0474d134-f983-4590-9cf9-29bd1c6e33cb/ep3/annotation.hdf5
+0474d134-f983-4590-9cf9-29bd1c6e33cb/ep3/fisheye_cam0.mp4
+0474d134-f983-4590-9cf9-29bd1c6e33cb/ep3/fisheye_cam1.mp4
+0474d134-f983-4590-9cf9-29bd1c6e33cb/ep3/fisheye_cam2.mp4
+0474d134-f983-4590-9cf9-29bd1c6e33cb/ep3/fisheye_cam3.mp4
+0474d134-f983-4590-9cf9-29bd1c6e33cb/ep3/stereo_left.mp4
+0474d134-f983-4590-9cf9-29bd1c6e33cb/ep3/stereo_right.mp4
+33f7ae08-ac1d-4321-9cb9-eca79016b359/ep1/annotation.hdf5
+33f7ae08-ac1d-4321-9cb9-eca79016b359/ep1/fisheye_cam0.mp4
+33f7ae08-ac1d-4321-9cb9-eca79016b359/ep1/fisheye_cam1.mp4
+33f7ae08-ac1d-4321-9cb9-eca79016b359/ep1/fisheye_cam2.mp4
+33f7ae08-ac1d-4321-9cb9-eca79016b359/ep1/fisheye_cam3.mp4
+33f7ae08-ac1d-4321-9cb9-eca79016b359/ep1/stereo_left.mp4
+33f7ae08-ac1d-4321-9cb9-eca79016b359/ep1/stereo_right.mp4
+34f07a04-eb37-45a3-95ec-189ed5f4a85b/ep5/annotation.hdf5
+34f07a04-eb37-45a3-95ec-189ed5f4a85b/ep5/fisheye_cam0.mp4
+34f07a04-eb37-45a3-95ec-189ed5f4a85b/ep5/fisheye_cam1.mp4
+34f07a04-eb37-45a3-95ec-189ed5f4a85b/ep5/fisheye_cam2.mp4
+34f07a04-eb37-45a3-95ec-189ed5f4a85b/ep5/fisheye_cam3.mp4
+34f07a04-eb37-45a3-95ec-189ed5f4a85b/ep5/stereo_left.mp4
+34f07a04-eb37-45a3-95ec-189ed5f4a85b/ep5/stereo_right.mp4
+2f579b88-12fb-41c2-aa2a-9b8f5ae8f918/ep1/annotation.hdf5
+2f579b88-12fb-41c2-aa2a-9b8f5ae8f918/ep1/fisheye_cam0.mp4
+2f579b88-12fb-41c2-aa2a-9b8f5ae8f918/ep1/fisheye_cam1.mp4
+2f579b88-12fb-41c2-aa2a-9b8f5ae8f918/ep1/fisheye_cam2.mp4
+2f579b88-12fb-41c2-aa2a-9b8f5ae8f918/ep1/fisheye_cam3.mp4
+2f579b88-12fb-41c2-aa2a-9b8f5ae8f918/ep1/stereo_left.mp4
+2f579b88-12fb-41c2-aa2a-9b8f5ae8f918/ep1/stereo_right.mp4
+0ffa6540-b99f-4909-9f80-c1b3e4c14a85/ep6/annotation.hdf5
+0ffa6540-b99f-4909-9f80-c1b3e4c14a85/ep6/fisheye_cam0.mp4
+0ffa6540-b99f-4909-9f80-c1b3e4c14a85/ep6/fisheye_cam1.mp4
+0ffa6540-b99f-4909-9f80-c1b3e4c14a85/ep6/fisheye_cam2.mp4
+0ffa6540-b99f-4909-9f80-c1b3e4c14a85/ep6/fisheye_cam3.mp4
+0ffa6540-b99f-4909-9f80-c1b3e4c14a85/ep6/stereo_left.mp4
+0ffa6540-b99f-4909-9f80-c1b3e4c14a85/ep6/stereo_right.mp4
+c5e31792-0961-463c-bfb4-ff89030c2ac8/ep1/annotation.hdf5
+c5e31792-0961-463c-bfb4-ff89030c2ac8/ep1/fisheye_cam0.mp4
+c5e31792-0961-463c-bfb4-ff89030c2ac8/ep1/fisheye_cam1.mp4
+c5e31792-0961-463c-bfb4-ff89030c2ac8/ep1/fisheye_cam2.mp4
+c5e31792-0961-463c-bfb4-ff89030c2ac8/ep1/fisheye_cam3.mp4
+c5e31792-0961-463c-bfb4-ff89030c2ac8/ep1/stereo_left.mp4
+c5e31792-0961-463c-bfb4-ff89030c2ac8/ep1/stereo_right.mp4
+6dc018a4-0614-478c-9332-86bd6c4af9e8/ep2/annotation.hdf5
+6dc018a4-0614-478c-9332-86bd6c4af9e8/ep2/fisheye_cam0.mp4
+6dc018a4-0614-478c-9332-86bd6c4af9e8/ep2/fisheye_cam1.mp4
+6dc018a4-0614-478c-9332-86bd6c4af9e8/ep2/fisheye_cam2.mp4
+6dc018a4-0614-478c-9332-86bd6c4af9e8/ep2/fisheye_cam3.mp4
+6dc018a4-0614-478c-9332-86bd6c4af9e8/ep2/stereo_left.mp4
+6dc018a4-0614-478c-9332-86bd6c4af9e8/ep2/stereo_right.mp4
+c09b25c6-218c-4e06-95ff-d63eccf20f2b/ep8/annotation.hdf5
+c09b25c6-218c-4e06-95ff-d63eccf20f2b/ep8/fisheye_cam0.mp4
+c09b25c6-218c-4e06-95ff-d63eccf20f2b/ep8/fisheye_cam1.mp4
+c09b25c6-218c-4e06-95ff-d63eccf20f2b/ep8/fisheye_cam2.mp4
+c09b25c6-218c-4e06-95ff-d63eccf20f2b/ep8/fisheye_cam3.mp4
+c09b25c6-218c-4e06-95ff-d63eccf20f2b/ep8/stereo_left.mp4
+c09b25c6-218c-4e06-95ff-d63eccf20f2b/ep8/stereo_right.mp4
+0d330006-684f-4e93-bf38-06d53be3fe5f/ep3/annotation.hdf5
+0d330006-684f-4e93-bf38-06d53be3fe5f/ep3/fisheye_cam0.mp4
+0d330006-684f-4e93-bf38-06d53be3fe5f/ep3/fisheye_cam1.mp4
+0d330006-684f-4e93-bf38-06d53be3fe5f/ep3/fisheye_cam2.mp4
+0d330006-684f-4e93-bf38-06d53be3fe5f/ep3/fisheye_cam3.mp4
+0d330006-684f-4e93-bf38-06d53be3fe5f/ep3/stereo_left.mp4
+0d330006-684f-4e93-bf38-06d53be3fe5f/ep3/stereo_right.mp4
+cba9c19e-a55f-46e8-bda8-423cdcc2de5f/ep1/annotation.hdf5
+cba9c19e-a55f-46e8-bda8-423cdcc2de5f/ep1/fisheye_cam0.mp4
+cba9c19e-a55f-46e8-bda8-423cdcc2de5f/ep1/fisheye_cam1.mp4
+cba9c19e-a55f-46e8-bda8-423cdcc2de5f/ep1/fisheye_cam2.mp4
+cba9c19e-a55f-46e8-bda8-423cdcc2de5f/ep1/fisheye_cam3.mp4
+cba9c19e-a55f-46e8-bda8-423cdcc2de5f/ep1/stereo_left.mp4
+cba9c19e-a55f-46e8-bda8-423cdcc2de5f/ep1/stereo_right.mp4
+322958e6-2d26-40d7-af3d-cb16362c78f7/ep2/annotation.hdf5
+322958e6-2d26-40d7-af3d-cb16362c78f7/ep2/fisheye_cam0.mp4
+322958e6-2d26-40d7-af3d-cb16362c78f7/ep2/fisheye_cam1.mp4
+322958e6-2d26-40d7-af3d-cb16362c78f7/ep2/fisheye_cam2.mp4
+322958e6-2d26-40d7-af3d-cb16362c78f7/ep2/fisheye_cam3.mp4
+322958e6-2d26-40d7-af3d-cb16362c78f7/ep2/stereo_left.mp4
+322958e6-2d26-40d7-af3d-cb16362c78f7/ep2/stereo_right.mp4
+7f443115-54a4-4d50-acfb-e05fe506c270/ep5/annotation.hdf5
+7f443115-54a4-4d50-acfb-e05fe506c270/ep5/fisheye_cam0.mp4
+7f443115-54a4-4d50-acfb-e05fe506c270/ep5/fisheye_cam1.mp4
+7f443115-54a4-4d50-acfb-e05fe506c270/ep5/fisheye_cam2.mp4
+7f443115-54a4-4d50-acfb-e05fe506c270/ep5/fisheye_cam3.mp4
+7f443115-54a4-4d50-acfb-e05fe506c270/ep5/stereo_left.mp4
+7f443115-54a4-4d50-acfb-e05fe506c270/ep5/stereo_right.mp4
+af8e9345-011f-4c97-a78e-ecb4a268220f/ep4/annotation.hdf5
+af8e9345-011f-4c97-a78e-ecb4a268220f/ep4/fisheye_cam0.mp4
+af8e9345-011f-4c97-a78e-ecb4a268220f/ep4/fisheye_cam1.mp4
+af8e9345-011f-4c97-a78e-ecb4a268220f/ep4/fisheye_cam2.mp4
+af8e9345-011f-4c97-a78e-ecb4a268220f/ep4/fisheye_cam3.mp4
+af8e9345-011f-4c97-a78e-ecb4a268220f/ep4/stereo_left.mp4
+af8e9345-011f-4c97-a78e-ecb4a268220f/ep4/stereo_right.mp4
+7f723cf6-13a4-451d-96bf-d37cbb3f84f3/ep9/annotation.hdf5
+7f723cf6-13a4-451d-96bf-d37cbb3f84f3/ep9/fisheye_cam0.mp4
+7f723cf6-13a4-451d-96bf-d37cbb3f84f3/ep9/fisheye_cam1.mp4
+7f723cf6-13a4-451d-96bf-d37cbb3f84f3/ep9/fisheye_cam2.mp4
+7f723cf6-13a4-451d-96bf-d37cbb3f84f3/ep9/fisheye_cam3.mp4
+7f723cf6-13a4-451d-96bf-d37cbb3f84f3/ep9/stereo_left.mp4
+7f723cf6-13a4-451d-96bf-d37cbb3f84f3/ep9/stereo_right.mp4
+3ebbfff2-3a60-4cf4-af98-9db074a46e73/ep11/annotation.hdf5
+3ebbfff2-3a60-4cf4-af98-9db074a46e73/ep11/fisheye_cam0.mp4
+3ebbfff2-3a60-4cf4-af98-9db074a46e73/ep11/fisheye_cam1.mp4
+3ebbfff2-3a60-4cf4-af98-9db074a46e73/ep11/fisheye_cam2.mp4
+3ebbfff2-3a60-4cf4-af98-9db074a46e73/ep11/fisheye_cam3.mp4
+3ebbfff2-3a60-4cf4-af98-9db074a46e73/ep11/stereo_left.mp4
+3ebbfff2-3a60-4cf4-af98-9db074a46e73/ep11/stereo_right.mp4
+90fe845b-df65-4297-838c-7924d4c3b6ad/ep9/annotation.hdf5
+90fe845b-df65-4297-838c-7924d4c3b6ad/ep9/fisheye_cam0.mp4
+90fe845b-df65-4297-838c-7924d4c3b6ad/ep9/fisheye_cam1.mp4
+90fe845b-df65-4297-838c-7924d4c3b6ad/ep9/fisheye_cam2.mp4
+90fe845b-df65-4297-838c-7924d4c3b6ad/ep9/fisheye_cam3.mp4
+90fe845b-df65-4297-838c-7924d4c3b6ad/ep9/stereo_left.mp4
+90fe845b-df65-4297-838c-7924d4c3b6ad/ep9/stereo_right.mp4
+638630bb-caf0-4367-a990-7a3d02dd395a/ep2/annotation.hdf5
+638630bb-caf0-4367-a990-7a3d02dd395a/ep2/fisheye_cam0.mp4
+638630bb-caf0-4367-a990-7a3d02dd395a/ep2/fisheye_cam1.mp4
+638630bb-caf0-4367-a990-7a3d02dd395a/ep2/fisheye_cam2.mp4
+638630bb-caf0-4367-a990-7a3d02dd395a/ep2/fisheye_cam3.mp4
+638630bb-caf0-4367-a990-7a3d02dd395a/ep2/stereo_left.mp4
+638630bb-caf0-4367-a990-7a3d02dd395a/ep2/stereo_right.mp4
+17eea990-0d39-4cc8-a7c8-03fd56b4bf04/ep10/annotation.hdf5
+17eea990-0d39-4cc8-a7c8-03fd56b4bf04/ep10/fisheye_cam0.mp4
+17eea990-0d39-4cc8-a7c8-03fd56b4bf04/ep10/fisheye_cam1.mp4
+17eea990-0d39-4cc8-a7c8-03fd56b4bf04/ep10/fisheye_cam2.mp4
+17eea990-0d39-4cc8-a7c8-03fd56b4bf04/ep10/fisheye_cam3.mp4
+17eea990-0d39-4cc8-a7c8-03fd56b4bf04/ep10/stereo_left.mp4
+17eea990-0d39-4cc8-a7c8-03fd56b4bf04/ep10/stereo_right.mp4
+81d4c5c5-b164-419c-86e0-c26347284f83/ep9/annotation.hdf5
+81d4c5c5-b164-419c-86e0-c26347284f83/ep9/fisheye_cam0.mp4
+81d4c5c5-b164-419c-86e0-c26347284f83/ep9/fisheye_cam1.mp4
+81d4c5c5-b164-419c-86e0-c26347284f83/ep9/fisheye_cam2.mp4
+81d4c5c5-b164-419c-86e0-c26347284f83/ep9/fisheye_cam3.mp4
+81d4c5c5-b164-419c-86e0-c26347284f83/ep9/stereo_left.mp4
+81d4c5c5-b164-419c-86e0-c26347284f83/ep9/stereo_right.mp4
+85a956ae-65c6-4d11-9e7f-720d6d71ae5b/ep5/annotation.hdf5
+85a956ae-65c6-4d11-9e7f-720d6d71ae5b/ep5/fisheye_cam0.mp4
+85a956ae-65c6-4d11-9e7f-720d6d71ae5b/ep5/fisheye_cam1.mp4
+85a956ae-65c6-4d11-9e7f-720d6d71ae5b/ep5/fisheye_cam2.mp4
+85a956ae-65c6-4d11-9e7f-720d6d71ae5b/ep5/fisheye_cam3.mp4
+85a956ae-65c6-4d11-9e7f-720d6d71ae5b/ep5/stereo_left.mp4
+85a956ae-65c6-4d11-9e7f-720d6d71ae5b/ep5/stereo_right.mp4
+4b7cc33d-56cf-4029-bda8-f154bd146447/ep5/annotation.hdf5
+4b7cc33d-56cf-4029-bda8-f154bd146447/ep5/fisheye_cam0.mp4
+4b7cc33d-56cf-4029-bda8-f154bd146447/ep5/fisheye_cam1.mp4
+4b7cc33d-56cf-4029-bda8-f154bd146447/ep5/fisheye_cam2.mp4
+4b7cc33d-56cf-4029-bda8-f154bd146447/ep5/fisheye_cam3.mp4
+4b7cc33d-56cf-4029-bda8-f154bd146447/ep5/stereo_left.mp4
+4b7cc33d-56cf-4029-bda8-f154bd146447/ep5/stereo_right.mp4
+5d994eb3-bc12-46dd-8984-da2d5f1a7055/ep9/annotation.hdf5
+5d994eb3-bc12-46dd-8984-da2d5f1a7055/ep9/fisheye_cam0.mp4
+5d994eb3-bc12-46dd-8984-da2d5f1a7055/ep9/fisheye_cam1.mp4
+5d994eb3-bc12-46dd-8984-da2d5f1a7055/ep9/fisheye_cam2.mp4
+5d994eb3-bc12-46dd-8984-da2d5f1a7055/ep9/fisheye_cam3.mp4
+5d994eb3-bc12-46dd-8984-da2d5f1a7055/ep9/stereo_left.mp4
+5d994eb3-bc12-46dd-8984-da2d5f1a7055/ep9/stereo_right.mp4
+9dc8fc7c-977f-444a-9331-06d2dd7bf120/ep2/annotation.hdf5
+9dc8fc7c-977f-444a-9331-06d2dd7bf120/ep2/fisheye_cam0.mp4
+9dc8fc7c-977f-444a-9331-06d2dd7bf120/ep2/fisheye_cam1.mp4
+9dc8fc7c-977f-444a-9331-06d2dd7bf120/ep2/fisheye_cam2.mp4
+9dc8fc7c-977f-444a-9331-06d2dd7bf120/ep2/fisheye_cam3.mp4
+9dc8fc7c-977f-444a-9331-06d2dd7bf120/ep2/stereo_left.mp4
+9dc8fc7c-977f-444a-9331-06d2dd7bf120/ep2/stereo_right.mp4
+59d0d93f-55c9-4a55-8cd4-e570cc0499d6/ep4/annotation.hdf5
+59d0d93f-55c9-4a55-8cd4-e570cc0499d6/ep4/fisheye_cam0.mp4
+59d0d93f-55c9-4a55-8cd4-e570cc0499d6/ep4/fisheye_cam1.mp4
+59d0d93f-55c9-4a55-8cd4-e570cc0499d6/ep4/fisheye_cam2.mp4
+59d0d93f-55c9-4a55-8cd4-e570cc0499d6/ep4/fisheye_cam3.mp4
+59d0d93f-55c9-4a55-8cd4-e570cc0499d6/ep4/stereo_left.mp4
+59d0d93f-55c9-4a55-8cd4-e570cc0499d6/ep4/stereo_right.mp4
+acb1b145-804e-41fe-915e-16ff7f59433a/ep7/annotation.hdf5
+acb1b145-804e-41fe-915e-16ff7f59433a/ep7/fisheye_cam0.mp4
+acb1b145-804e-41fe-915e-16ff7f59433a/ep7/fisheye_cam1.mp4
+acb1b145-804e-41fe-915e-16ff7f59433a/ep7/fisheye_cam2.mp4
+acb1b145-804e-41fe-915e-16ff7f59433a/ep7/fisheye_cam3.mp4
+acb1b145-804e-41fe-915e-16ff7f59433a/ep7/stereo_left.mp4
+acb1b145-804e-41fe-915e-16ff7f59433a/ep7/stereo_right.mp4
+0da1f6d2-6564-4f28-b248-1d1b71f9521b/ep1/annotation.hdf5
+0da1f6d2-6564-4f28-b248-1d1b71f9521b/ep1/fisheye_cam0.mp4
+0da1f6d2-6564-4f28-b248-1d1b71f9521b/ep1/fisheye_cam1.mp4
+0da1f6d2-6564-4f28-b248-1d1b71f9521b/ep1/fisheye_cam2.mp4
+0da1f6d2-6564-4f28-b248-1d1b71f9521b/ep1/fisheye_cam3.mp4
+0da1f6d2-6564-4f28-b248-1d1b71f9521b/ep1/stereo_left.mp4
+0da1f6d2-6564-4f28-b248-1d1b71f9521b/ep1/stereo_right.mp4
+994e4c6c-660a-49f2-90a7-3ebbbb55eb9b/ep2/annotation.hdf5
+994e4c6c-660a-49f2-90a7-3ebbbb55eb9b/ep2/fisheye_cam0.mp4
+994e4c6c-660a-49f2-90a7-3ebbbb55eb9b/ep2/fisheye_cam1.mp4
+994e4c6c-660a-49f2-90a7-3ebbbb55eb9b/ep2/fisheye_cam2.mp4
+994e4c6c-660a-49f2-90a7-3ebbbb55eb9b/ep2/fisheye_cam3.mp4
+994e4c6c-660a-49f2-90a7-3ebbbb55eb9b/ep2/stereo_left.mp4
+994e4c6c-660a-49f2-90a7-3ebbbb55eb9b/ep2/stereo_right.mp4
+53ccaadb-44df-4ddf-bc40-c1bcedb92faa/ep7/annotation.hdf5
+53ccaadb-44df-4ddf-bc40-c1bcedb92faa/ep7/fisheye_cam0.mp4
+53ccaadb-44df-4ddf-bc40-c1bcedb92faa/ep7/fisheye_cam1.mp4
+53ccaadb-44df-4ddf-bc40-c1bcedb92faa/ep7/fisheye_cam2.mp4
+53ccaadb-44df-4ddf-bc40-c1bcedb92faa/ep7/fisheye_cam3.mp4
+53ccaadb-44df-4ddf-bc40-c1bcedb92faa/ep7/stereo_left.mp4
+53ccaadb-44df-4ddf-bc40-c1bcedb92faa/ep7/stereo_right.mp4
+d5520c73-fa72-4e92-ad2a-4c3e48d8ad7c/ep6/annotation.hdf5
+d5520c73-fa72-4e92-ad2a-4c3e48d8ad7c/ep6/fisheye_cam0.mp4
+d5520c73-fa72-4e92-ad2a-4c3e48d8ad7c/ep6/fisheye_cam1.mp4
+d5520c73-fa72-4e92-ad2a-4c3e48d8ad7c/ep6/fisheye_cam2.mp4
+d5520c73-fa72-4e92-ad2a-4c3e48d8ad7c/ep6/fisheye_cam3.mp4
+d5520c73-fa72-4e92-ad2a-4c3e48d8ad7c/ep6/stereo_left.mp4
+d5520c73-fa72-4e92-ad2a-4c3e48d8ad7c/ep6/stereo_right.mp4
+53f2c164-09dd-48ca-ae69-4f5955a902ed/ep4/annotation.hdf5
+53f2c164-09dd-48ca-ae69-4f5955a902ed/ep4/fisheye_cam0.mp4
+53f2c164-09dd-48ca-ae69-4f5955a902ed/ep4/fisheye_cam1.mp4
+53f2c164-09dd-48ca-ae69-4f5955a902ed/ep4/fisheye_cam2.mp4
+53f2c164-09dd-48ca-ae69-4f5955a902ed/ep4/fisheye_cam3.mp4
+53f2c164-09dd-48ca-ae69-4f5955a902ed/ep4/stereo_left.mp4
+53f2c164-09dd-48ca-ae69-4f5955a902ed/ep4/stereo_right.mp4
+a00c1d99-de08-42bc-8623-712b9dde7fde/ep3/annotation.hdf5
+a00c1d99-de08-42bc-8623-712b9dde7fde/ep3/fisheye_cam0.mp4
+a00c1d99-de08-42bc-8623-712b9dde7fde/ep3/fisheye_cam1.mp4
+a00c1d99-de08-42bc-8623-712b9dde7fde/ep3/fisheye_cam2.mp4
+a00c1d99-de08-42bc-8623-712b9dde7fde/ep3/fisheye_cam3.mp4
+a00c1d99-de08-42bc-8623-712b9dde7fde/ep3/stereo_left.mp4
+a00c1d99-de08-42bc-8623-712b9dde7fde/ep3/stereo_right.mp4
+d342e6e2-c381-4a1c-b0ac-e62a5c80624f/ep3/annotation.hdf5
+d342e6e2-c381-4a1c-b0ac-e62a5c80624f/ep3/fisheye_cam0.mp4
+d342e6e2-c381-4a1c-b0ac-e62a5c80624f/ep3/fisheye_cam1.mp4
+d342e6e2-c381-4a1c-b0ac-e62a5c80624f/ep3/fisheye_cam2.mp4
+d342e6e2-c381-4a1c-b0ac-e62a5c80624f/ep3/fisheye_cam3.mp4
+d342e6e2-c381-4a1c-b0ac-e62a5c80624f/ep3/stereo_left.mp4
+d342e6e2-c381-4a1c-b0ac-e62a5c80624f/ep3/stereo_right.mp4
+46f7dea3-76bd-4de4-a0fe-56c0c1e1a276/ep3/annotation.hdf5
+46f7dea3-76bd-4de4-a0fe-56c0c1e1a276/ep3/fisheye_cam0.mp4
+46f7dea3-76bd-4de4-a0fe-56c0c1e1a276/ep3/fisheye_cam1.mp4
+46f7dea3-76bd-4de4-a0fe-56c0c1e1a276/ep3/fisheye_cam2.mp4
+46f7dea3-76bd-4de4-a0fe-56c0c1e1a276/ep3/fisheye_cam3.mp4
+46f7dea3-76bd-4de4-a0fe-56c0c1e1a276/ep3/stereo_left.mp4
+46f7dea3-76bd-4de4-a0fe-56c0c1e1a276/ep3/stereo_right.mp4
+adb1bb6f-6df3-4f46-b895-f045ac3e0772/ep2/annotation.hdf5
+adb1bb6f-6df3-4f46-b895-f045ac3e0772/ep2/fisheye_cam0.mp4
+adb1bb6f-6df3-4f46-b895-f045ac3e0772/ep2/fisheye_cam1.mp4
+adb1bb6f-6df3-4f46-b895-f045ac3e0772/ep2/fisheye_cam2.mp4
+adb1bb6f-6df3-4f46-b895-f045ac3e0772/ep2/fisheye_cam3.mp4
+adb1bb6f-6df3-4f46-b895-f045ac3e0772/ep2/stereo_left.mp4
+adb1bb6f-6df3-4f46-b895-f045ac3e0772/ep2/stereo_right.mp4
+5399ef86-4df9-49bc-809f-8f4f92f9e659/ep6/annotation.hdf5
+5399ef86-4df9-49bc-809f-8f4f92f9e659/ep6/fisheye_cam0.mp4
+5399ef86-4df9-49bc-809f-8f4f92f9e659/ep6/fisheye_cam1.mp4
+5399ef86-4df9-49bc-809f-8f4f92f9e659/ep6/fisheye_cam2.mp4
+5399ef86-4df9-49bc-809f-8f4f92f9e659/ep6/fisheye_cam3.mp4
+5399ef86-4df9-49bc-809f-8f4f92f9e659/ep6/stereo_left.mp4
+5399ef86-4df9-49bc-809f-8f4f92f9e659/ep6/stereo_right.mp4
+5aeb0920-ab9f-4dc2-a261-747a678bf9cb/ep2/annotation.hdf5
+5aeb0920-ab9f-4dc2-a261-747a678bf9cb/ep2/fisheye_cam0.mp4
+5aeb0920-ab9f-4dc2-a261-747a678bf9cb/ep2/fisheye_cam1.mp4
+5aeb0920-ab9f-4dc2-a261-747a678bf9cb/ep2/fisheye_cam2.mp4
+5aeb0920-ab9f-4dc2-a261-747a678bf9cb/ep2/fisheye_cam3.mp4
+5aeb0920-ab9f-4dc2-a261-747a678bf9cb/ep2/stereo_left.mp4
+5aeb0920-ab9f-4dc2-a261-747a678bf9cb/ep2/stereo_right.mp4
+877779cd-25f3-4293-a3c4-39067dd9558c/ep4/annotation.hdf5
+877779cd-25f3-4293-a3c4-39067dd9558c/ep4/fisheye_cam0.mp4
+877779cd-25f3-4293-a3c4-39067dd9558c/ep4/fisheye_cam1.mp4
+877779cd-25f3-4293-a3c4-39067dd9558c/ep4/fisheye_cam2.mp4
+877779cd-25f3-4293-a3c4-39067dd9558c/ep4/fisheye_cam3.mp4
+877779cd-25f3-4293-a3c4-39067dd9558c/ep4/stereo_left.mp4
+877779cd-25f3-4293-a3c4-39067dd9558c/ep4/stereo_right.mp4
+9f84a0a7-8964-48c5-b7e1-8a1d28094c86/ep2/annotation.hdf5
+9f84a0a7-8964-48c5-b7e1-8a1d28094c86/ep2/fisheye_cam0.mp4
+9f84a0a7-8964-48c5-b7e1-8a1d28094c86/ep2/fisheye_cam1.mp4
+9f84a0a7-8964-48c5-b7e1-8a1d28094c86/ep2/fisheye_cam2.mp4
+9f84a0a7-8964-48c5-b7e1-8a1d28094c86/ep2/fisheye_cam3.mp4
+9f84a0a7-8964-48c5-b7e1-8a1d28094c86/ep2/stereo_left.mp4
+9f84a0a7-8964-48c5-b7e1-8a1d28094c86/ep2/stereo_right.mp4
+84106bb9-1cf8-47b6-8d11-fddbfe381e4f/ep5/annotation.hdf5
+84106bb9-1cf8-47b6-8d11-fddbfe381e4f/ep5/fisheye_cam0.mp4
+84106bb9-1cf8-47b6-8d11-fddbfe381e4f/ep5/fisheye_cam1.mp4
+84106bb9-1cf8-47b6-8d11-fddbfe381e4f/ep5/fisheye_cam2.mp4
+84106bb9-1cf8-47b6-8d11-fddbfe381e4f/ep5/fisheye_cam3.mp4
+84106bb9-1cf8-47b6-8d11-fddbfe381e4f/ep5/stereo_left.mp4
+84106bb9-1cf8-47b6-8d11-fddbfe381e4f/ep5/stereo_right.mp4
+ba18b7c1-21ff-45da-8452-41acce7fc8de/ep2/annotation.hdf5
+ba18b7c1-21ff-45da-8452-41acce7fc8de/ep2/fisheye_cam0.mp4
+ba18b7c1-21ff-45da-8452-41acce7fc8de/ep2/fisheye_cam1.mp4
+ba18b7c1-21ff-45da-8452-41acce7fc8de/ep2/fisheye_cam2.mp4
+ba18b7c1-21ff-45da-8452-41acce7fc8de/ep2/fisheye_cam3.mp4
+ba18b7c1-21ff-45da-8452-41acce7fc8de/ep2/stereo_left.mp4
+ba18b7c1-21ff-45da-8452-41acce7fc8de/ep2/stereo_right.mp4
+6fbdecec-d4d5-477a-b94b-2486adb0dbd8/ep6/annotation.hdf5
+6fbdecec-d4d5-477a-b94b-2486adb0dbd8/ep6/fisheye_cam0.mp4
+6fbdecec-d4d5-477a-b94b-2486adb0dbd8/ep6/fisheye_cam1.mp4
+6fbdecec-d4d5-477a-b94b-2486adb0dbd8/ep6/fisheye_cam2.mp4
+6fbdecec-d4d5-477a-b94b-2486adb0dbd8/ep6/fisheye_cam3.mp4
+6fbdecec-d4d5-477a-b94b-2486adb0dbd8/ep6/stereo_left.mp4
+6fbdecec-d4d5-477a-b94b-2486adb0dbd8/ep6/stereo_right.mp4
+c5e5183a-9eca-448a-9ec1-a0a40682b81a/ep4/annotation.hdf5
+c5e5183a-9eca-448a-9ec1-a0a40682b81a/ep4/fisheye_cam0.mp4
+c5e5183a-9eca-448a-9ec1-a0a40682b81a/ep4/fisheye_cam1.mp4
+c5e5183a-9eca-448a-9ec1-a0a40682b81a/ep4/fisheye_cam2.mp4
+c5e5183a-9eca-448a-9ec1-a0a40682b81a/ep4/fisheye_cam3.mp4
+c5e5183a-9eca-448a-9ec1-a0a40682b81a/ep4/stereo_left.mp4
+c5e5183a-9eca-448a-9ec1-a0a40682b81a/ep4/stereo_right.mp4
+003dcaf0-edba-4787-ada0-187d2748f684/ep1/annotation.hdf5
+003dcaf0-edba-4787-ada0-187d2748f684/ep1/fisheye_cam0.mp4
+003dcaf0-edba-4787-ada0-187d2748f684/ep1/fisheye_cam1.mp4
+003dcaf0-edba-4787-ada0-187d2748f684/ep1/fisheye_cam2.mp4
+003dcaf0-edba-4787-ada0-187d2748f684/ep1/fisheye_cam3.mp4
+003dcaf0-edba-4787-ada0-187d2748f684/ep1/stereo_left.mp4
+003dcaf0-edba-4787-ada0-187d2748f684/ep1/stereo_right.mp4
+55f345d5-aa45-4d75-829b-a6ec41dca2ed/ep7/annotation.hdf5
+55f345d5-aa45-4d75-829b-a6ec41dca2ed/ep7/fisheye_cam0.mp4
+55f345d5-aa45-4d75-829b-a6ec41dca2ed/ep7/fisheye_cam1.mp4
+55f345d5-aa45-4d75-829b-a6ec41dca2ed/ep7/fisheye_cam2.mp4
+55f345d5-aa45-4d75-829b-a6ec41dca2ed/ep7/fisheye_cam3.mp4
+55f345d5-aa45-4d75-829b-a6ec41dca2ed/ep7/stereo_left.mp4
+55f345d5-aa45-4d75-829b-a6ec41dca2ed/ep7/stereo_right.mp4
+5e221284-f4af-4408-9177-25200282b6de/ep4/annotation.hdf5
+5e221284-f4af-4408-9177-25200282b6de/ep4/fisheye_cam0.mp4
+5e221284-f4af-4408-9177-25200282b6de/ep4/fisheye_cam1.mp4
+5e221284-f4af-4408-9177-25200282b6de/ep4/fisheye_cam2.mp4
+5e221284-f4af-4408-9177-25200282b6de/ep4/fisheye_cam3.mp4
+5e221284-f4af-4408-9177-25200282b6de/ep4/stereo_left.mp4
+5e221284-f4af-4408-9177-25200282b6de/ep4/stereo_right.mp4
+676c3db3-464e-406a-be9f-8283518e8181/ep4/annotation.hdf5
+676c3db3-464e-406a-be9f-8283518e8181/ep4/fisheye_cam0.mp4
+676c3db3-464e-406a-be9f-8283518e8181/ep4/fisheye_cam1.mp4
+676c3db3-464e-406a-be9f-8283518e8181/ep4/fisheye_cam2.mp4
+676c3db3-464e-406a-be9f-8283518e8181/ep4/fisheye_cam3.mp4
+676c3db3-464e-406a-be9f-8283518e8181/ep4/stereo_left.mp4
+676c3db3-464e-406a-be9f-8283518e8181/ep4/stereo_right.mp4
+46b507a1-b1cb-4851-8558-9435de4abf68/ep2/annotation.hdf5
+46b507a1-b1cb-4851-8558-9435de4abf68/ep2/fisheye_cam0.mp4
+46b507a1-b1cb-4851-8558-9435de4abf68/ep2/fisheye_cam1.mp4
+46b507a1-b1cb-4851-8558-9435de4abf68/ep2/fisheye_cam2.mp4
+46b507a1-b1cb-4851-8558-9435de4abf68/ep2/fisheye_cam3.mp4
+46b507a1-b1cb-4851-8558-9435de4abf68/ep2/stereo_left.mp4
+46b507a1-b1cb-4851-8558-9435de4abf68/ep2/stereo_right.mp4
+3cd70c8e-4ef5-481c-aa61-88b0dd4dec1f/ep4/annotation.hdf5
+3cd70c8e-4ef5-481c-aa61-88b0dd4dec1f/ep4/fisheye_cam0.mp4
+3cd70c8e-4ef5-481c-aa61-88b0dd4dec1f/ep4/fisheye_cam1.mp4
+3cd70c8e-4ef5-481c-aa61-88b0dd4dec1f/ep4/fisheye_cam2.mp4
+3cd70c8e-4ef5-481c-aa61-88b0dd4dec1f/ep4/fisheye_cam3.mp4
+3cd70c8e-4ef5-481c-aa61-88b0dd4dec1f/ep4/stereo_left.mp4
+3cd70c8e-4ef5-481c-aa61-88b0dd4dec1f/ep4/stereo_right.mp4
+aa3ca3ec-56fd-4bbd-ad15-055eabc1838a/ep7/annotation.hdf5
+aa3ca3ec-56fd-4bbd-ad15-055eabc1838a/ep7/fisheye_cam0.mp4
+aa3ca3ec-56fd-4bbd-ad15-055eabc1838a/ep7/fisheye_cam1.mp4
+aa3ca3ec-56fd-4bbd-ad15-055eabc1838a/ep7/fisheye_cam2.mp4
+aa3ca3ec-56fd-4bbd-ad15-055eabc1838a/ep7/fisheye_cam3.mp4
+aa3ca3ec-56fd-4bbd-ad15-055eabc1838a/ep7/stereo_left.mp4
+aa3ca3ec-56fd-4bbd-ad15-055eabc1838a/ep7/stereo_right.mp4
+82c95b73-ba14-4083-8929-ddeb85d22e9f/ep3/annotation.hdf5
+82c95b73-ba14-4083-8929-ddeb85d22e9f/ep3/fisheye_cam0.mp4
+82c95b73-ba14-4083-8929-ddeb85d22e9f/ep3/fisheye_cam1.mp4
+82c95b73-ba14-4083-8929-ddeb85d22e9f/ep3/fisheye_cam2.mp4
+82c95b73-ba14-4083-8929-ddeb85d22e9f/ep3/fisheye_cam3.mp4
+82c95b73-ba14-4083-8929-ddeb85d22e9f/ep3/stereo_left.mp4
+82c95b73-ba14-4083-8929-ddeb85d22e9f/ep3/stereo_right.mp4
+8533707e-d7e0-4d1a-99d1-fa154a8e62ad/ep5/annotation.hdf5
+8533707e-d7e0-4d1a-99d1-fa154a8e62ad/ep5/fisheye_cam0.mp4
+8533707e-d7e0-4d1a-99d1-fa154a8e62ad/ep5/fisheye_cam1.mp4
+8533707e-d7e0-4d1a-99d1-fa154a8e62ad/ep5/fisheye_cam2.mp4
+8533707e-d7e0-4d1a-99d1-fa154a8e62ad/ep5/fisheye_cam3.mp4
+8533707e-d7e0-4d1a-99d1-fa154a8e62ad/ep5/stereo_left.mp4
+8533707e-d7e0-4d1a-99d1-fa154a8e62ad/ep5/stereo_right.mp4
+81e9362f-fd63-4aec-92fa-1f489d6ac025/ep2/annotation.hdf5
+81e9362f-fd63-4aec-92fa-1f489d6ac025/ep2/fisheye_cam0.mp4
+81e9362f-fd63-4aec-92fa-1f489d6ac025/ep2/fisheye_cam1.mp4
+81e9362f-fd63-4aec-92fa-1f489d6ac025/ep2/fisheye_cam2.mp4
+81e9362f-fd63-4aec-92fa-1f489d6ac025/ep2/fisheye_cam3.mp4
+81e9362f-fd63-4aec-92fa-1f489d6ac025/ep2/stereo_left.mp4
+81e9362f-fd63-4aec-92fa-1f489d6ac025/ep2/stereo_right.mp4
+cd364707-88fc-43bd-a95b-61aee9f8bd20/ep5/annotation.hdf5
+cd364707-88fc-43bd-a95b-61aee9f8bd20/ep5/fisheye_cam0.mp4
+cd364707-88fc-43bd-a95b-61aee9f8bd20/ep5/fisheye_cam1.mp4
+cd364707-88fc-43bd-a95b-61aee9f8bd20/ep5/fisheye_cam2.mp4
+cd364707-88fc-43bd-a95b-61aee9f8bd20/ep5/fisheye_cam3.mp4
+cd364707-88fc-43bd-a95b-61aee9f8bd20/ep5/stereo_left.mp4
+cd364707-88fc-43bd-a95b-61aee9f8bd20/ep5/stereo_right.mp4
+c038defe-faa6-4fe9-aef6-84bd93726d1c/ep2/annotation.hdf5
+c038defe-faa6-4fe9-aef6-84bd93726d1c/ep2/fisheye_cam0.mp4
+c038defe-faa6-4fe9-aef6-84bd93726d1c/ep2/fisheye_cam1.mp4
+c038defe-faa6-4fe9-aef6-84bd93726d1c/ep2/fisheye_cam2.mp4
+c038defe-faa6-4fe9-aef6-84bd93726d1c/ep2/fisheye_cam3.mp4
+c038defe-faa6-4fe9-aef6-84bd93726d1c/ep2/stereo_left.mp4
+c038defe-faa6-4fe9-aef6-84bd93726d1c/ep2/stereo_right.mp4
+999100c1-b488-4ed1-a580-c0487f308e5d/ep5/annotation.hdf5
+999100c1-b488-4ed1-a580-c0487f308e5d/ep5/fisheye_cam0.mp4
+999100c1-b488-4ed1-a580-c0487f308e5d/ep5/fisheye_cam1.mp4
+999100c1-b488-4ed1-a580-c0487f308e5d/ep5/fisheye_cam2.mp4
+999100c1-b488-4ed1-a580-c0487f308e5d/ep5/fisheye_cam3.mp4
+999100c1-b488-4ed1-a580-c0487f308e5d/ep5/stereo_left.mp4
+999100c1-b488-4ed1-a580-c0487f308e5d/ep5/stereo_right.mp4
+bf622eaf-363e-40c4-8a04-1dfedd85a718/ep10/annotation.hdf5
+bf622eaf-363e-40c4-8a04-1dfedd85a718/ep10/fisheye_cam0.mp4
+bf622eaf-363e-40c4-8a04-1dfedd85a718/ep10/fisheye_cam1.mp4
+bf622eaf-363e-40c4-8a04-1dfedd85a718/ep10/fisheye_cam2.mp4
+bf622eaf-363e-40c4-8a04-1dfedd85a718/ep10/fisheye_cam3.mp4
+bf622eaf-363e-40c4-8a04-1dfedd85a718/ep10/stereo_left.mp4
+bf622eaf-363e-40c4-8a04-1dfedd85a718/ep10/stereo_right.mp4
+a3e06450-50f8-4894-a22f-1f891fad3809/ep7/annotation.hdf5
+a3e06450-50f8-4894-a22f-1f891fad3809/ep7/fisheye_cam0.mp4
+a3e06450-50f8-4894-a22f-1f891fad3809/ep7/fisheye_cam1.mp4
+a3e06450-50f8-4894-a22f-1f891fad3809/ep7/fisheye_cam2.mp4
+a3e06450-50f8-4894-a22f-1f891fad3809/ep7/fisheye_cam3.mp4
+a3e06450-50f8-4894-a22f-1f891fad3809/ep7/stereo_left.mp4
+a3e06450-50f8-4894-a22f-1f891fad3809/ep7/stereo_right.mp4
+04d9f378-2812-4c7f-a269-1ad0bc43445f/ep4/annotation.hdf5
+04d9f378-2812-4c7f-a269-1ad0bc43445f/ep4/fisheye_cam0.mp4
+04d9f378-2812-4c7f-a269-1ad0bc43445f/ep4/fisheye_cam1.mp4
+04d9f378-2812-4c7f-a269-1ad0bc43445f/ep4/fisheye_cam2.mp4
+04d9f378-2812-4c7f-a269-1ad0bc43445f/ep4/fisheye_cam3.mp4
+04d9f378-2812-4c7f-a269-1ad0bc43445f/ep4/stereo_left.mp4
+04d9f378-2812-4c7f-a269-1ad0bc43445f/ep4/stereo_right.mp4
+a1012a57-385e-45a9-8a59-694a26fe92a5/ep1/annotation.hdf5
+a1012a57-385e-45a9-8a59-694a26fe92a5/ep1/fisheye_cam0.mp4
+a1012a57-385e-45a9-8a59-694a26fe92a5/ep1/fisheye_cam1.mp4
+a1012a57-385e-45a9-8a59-694a26fe92a5/ep1/fisheye_cam2.mp4
+a1012a57-385e-45a9-8a59-694a26fe92a5/ep1/fisheye_cam3.mp4
+a1012a57-385e-45a9-8a59-694a26fe92a5/ep1/stereo_left.mp4
+a1012a57-385e-45a9-8a59-694a26fe92a5/ep1/stereo_right.mp4
+41e90813-8fff-4053-94f0-9537b1393bc2/ep1/annotation.hdf5
+41e90813-8fff-4053-94f0-9537b1393bc2/ep1/fisheye_cam0.mp4
+41e90813-8fff-4053-94f0-9537b1393bc2/ep1/fisheye_cam1.mp4
+41e90813-8fff-4053-94f0-9537b1393bc2/ep1/fisheye_cam2.mp4
+41e90813-8fff-4053-94f0-9537b1393bc2/ep1/fisheye_cam3.mp4
+41e90813-8fff-4053-94f0-9537b1393bc2/ep1/stereo_left.mp4
+41e90813-8fff-4053-94f0-9537b1393bc2/ep1/stereo_right.mp4
+bc53027e-d29a-473f-8f18-7b524b801e40/ep2/annotation.hdf5
+bc53027e-d29a-473f-8f18-7b524b801e40/ep2/fisheye_cam0.mp4
+bc53027e-d29a-473f-8f18-7b524b801e40/ep2/fisheye_cam1.mp4
+bc53027e-d29a-473f-8f18-7b524b801e40/ep2/fisheye_cam2.mp4
+bc53027e-d29a-473f-8f18-7b524b801e40/ep2/fisheye_cam3.mp4
+bc53027e-d29a-473f-8f18-7b524b801e40/ep2/stereo_left.mp4
+bc53027e-d29a-473f-8f18-7b524b801e40/ep2/stereo_right.mp4
+727f582a-fe2a-45c3-b506-216133a28c35/ep1/annotation.hdf5
+727f582a-fe2a-45c3-b506-216133a28c35/ep1/fisheye_cam0.mp4
+727f582a-fe2a-45c3-b506-216133a28c35/ep1/fisheye_cam1.mp4
+727f582a-fe2a-45c3-b506-216133a28c35/ep1/fisheye_cam2.mp4
+727f582a-fe2a-45c3-b506-216133a28c35/ep1/fisheye_cam3.mp4
+727f582a-fe2a-45c3-b506-216133a28c35/ep1/stereo_left.mp4
+727f582a-fe2a-45c3-b506-216133a28c35/ep1/stereo_right.mp4
+9c553886-83c5-4dc4-be5c-dcb269b3a771/ep2/annotation.hdf5
+9c553886-83c5-4dc4-be5c-dcb269b3a771/ep2/fisheye_cam0.mp4
+9c553886-83c5-4dc4-be5c-dcb269b3a771/ep2/fisheye_cam1.mp4
+9c553886-83c5-4dc4-be5c-dcb269b3a771/ep2/fisheye_cam2.mp4
+9c553886-83c5-4dc4-be5c-dcb269b3a771/ep2/fisheye_cam3.mp4
+9c553886-83c5-4dc4-be5c-dcb269b3a771/ep2/stereo_left.mp4
+9c553886-83c5-4dc4-be5c-dcb269b3a771/ep2/stereo_right.mp4
+b1292f20-bfca-497b-a41a-e7fd62bbf913/ep2/annotation.hdf5
+b1292f20-bfca-497b-a41a-e7fd62bbf913/ep2/fisheye_cam0.mp4
+b1292f20-bfca-497b-a41a-e7fd62bbf913/ep2/fisheye_cam1.mp4
+b1292f20-bfca-497b-a41a-e7fd62bbf913/ep2/fisheye_cam2.mp4
+b1292f20-bfca-497b-a41a-e7fd62bbf913/ep2/fisheye_cam3.mp4
+b1292f20-bfca-497b-a41a-e7fd62bbf913/ep2/stereo_left.mp4
+b1292f20-bfca-497b-a41a-e7fd62bbf913/ep2/stereo_right.mp4
+b49d8a3c-7a1e-4ca3-8ce8-ae1cce0df001/ep2/annotation.hdf5
+b49d8a3c-7a1e-4ca3-8ce8-ae1cce0df001/ep2/fisheye_cam0.mp4
+b49d8a3c-7a1e-4ca3-8ce8-ae1cce0df001/ep2/fisheye_cam1.mp4
+b49d8a3c-7a1e-4ca3-8ce8-ae1cce0df001/ep2/fisheye_cam2.mp4
+b49d8a3c-7a1e-4ca3-8ce8-ae1cce0df001/ep2/fisheye_cam3.mp4
+b49d8a3c-7a1e-4ca3-8ce8-ae1cce0df001/ep2/stereo_left.mp4
+b49d8a3c-7a1e-4ca3-8ce8-ae1cce0df001/ep2/stereo_right.mp4
+8692d5eb-19b2-4c49-9f88-da2ec373796e/ep3/annotation.hdf5
+8692d5eb-19b2-4c49-9f88-da2ec373796e/ep3/fisheye_cam0.mp4
+8692d5eb-19b2-4c49-9f88-da2ec373796e/ep3/fisheye_cam1.mp4
+8692d5eb-19b2-4c49-9f88-da2ec373796e/ep3/fisheye_cam2.mp4
+8692d5eb-19b2-4c49-9f88-da2ec373796e/ep3/fisheye_cam3.mp4
+8692d5eb-19b2-4c49-9f88-da2ec373796e/ep3/stereo_left.mp4
+8692d5eb-19b2-4c49-9f88-da2ec373796e/ep3/stereo_right.mp4
+54bf2b92-a5f8-440c-81fe-8f820bb90aab/ep3/annotation.hdf5
+54bf2b92-a5f8-440c-81fe-8f820bb90aab/ep3/fisheye_cam0.mp4
+54bf2b92-a5f8-440c-81fe-8f820bb90aab/ep3/fisheye_cam1.mp4
+54bf2b92-a5f8-440c-81fe-8f820bb90aab/ep3/fisheye_cam2.mp4
+54bf2b92-a5f8-440c-81fe-8f820bb90aab/ep3/fisheye_cam3.mp4
+54bf2b92-a5f8-440c-81fe-8f820bb90aab/ep3/stereo_left.mp4
+54bf2b92-a5f8-440c-81fe-8f820bb90aab/ep3/stereo_right.mp4
+3eb8ef88-902d-4e84-8a72-a361c9d18647/ep3/annotation.hdf5
+3eb8ef88-902d-4e84-8a72-a361c9d18647/ep3/fisheye_cam0.mp4
+3eb8ef88-902d-4e84-8a72-a361c9d18647/ep3/fisheye_cam1.mp4
+3eb8ef88-902d-4e84-8a72-a361c9d18647/ep3/fisheye_cam2.mp4
+3eb8ef88-902d-4e84-8a72-a361c9d18647/ep3/fisheye_cam3.mp4
+3eb8ef88-902d-4e84-8a72-a361c9d18647/ep3/stereo_left.mp4
+3eb8ef88-902d-4e84-8a72-a361c9d18647/ep3/stereo_right.mp4
+2aee3428-f462-49e6-95c4-b803c1803e0b/ep3/annotation.hdf5
+2aee3428-f462-49e6-95c4-b803c1803e0b/ep3/fisheye_cam0.mp4
+2aee3428-f462-49e6-95c4-b803c1803e0b/ep3/fisheye_cam1.mp4
+2aee3428-f462-49e6-95c4-b803c1803e0b/ep3/fisheye_cam2.mp4
+2aee3428-f462-49e6-95c4-b803c1803e0b/ep3/fisheye_cam3.mp4
+2aee3428-f462-49e6-95c4-b803c1803e0b/ep3/stereo_left.mp4
+2aee3428-f462-49e6-95c4-b803c1803e0b/ep3/stereo_right.mp4
+40611d4c-a36d-4ae3-8a71-da94b81cdce4/ep2/annotation.hdf5
+40611d4c-a36d-4ae3-8a71-da94b81cdce4/ep2/fisheye_cam0.mp4
+40611d4c-a36d-4ae3-8a71-da94b81cdce4/ep2/fisheye_cam1.mp4
+40611d4c-a36d-4ae3-8a71-da94b81cdce4/ep2/fisheye_cam2.mp4
+40611d4c-a36d-4ae3-8a71-da94b81cdce4/ep2/fisheye_cam3.mp4
+40611d4c-a36d-4ae3-8a71-da94b81cdce4/ep2/stereo_left.mp4
+40611d4c-a36d-4ae3-8a71-da94b81cdce4/ep2/stereo_right.mp4
+832369c5-8fca-4591-940a-7a7a28206de2/ep4/annotation.hdf5
+832369c5-8fca-4591-940a-7a7a28206de2/ep4/fisheye_cam0.mp4
+832369c5-8fca-4591-940a-7a7a28206de2/ep4/fisheye_cam1.mp4
+832369c5-8fca-4591-940a-7a7a28206de2/ep4/fisheye_cam2.mp4
+832369c5-8fca-4591-940a-7a7a28206de2/ep4/fisheye_cam3.mp4
+832369c5-8fca-4591-940a-7a7a28206de2/ep4/stereo_left.mp4
+832369c5-8fca-4591-940a-7a7a28206de2/ep4/stereo_right.mp4
+34353740-0ce5-4b2f-9394-f9fd9a19b8fe/ep3/annotation.hdf5
+34353740-0ce5-4b2f-9394-f9fd9a19b8fe/ep3/fisheye_cam0.mp4
+34353740-0ce5-4b2f-9394-f9fd9a19b8fe/ep3/fisheye_cam1.mp4
+34353740-0ce5-4b2f-9394-f9fd9a19b8fe/ep3/fisheye_cam2.mp4
+34353740-0ce5-4b2f-9394-f9fd9a19b8fe/ep3/fisheye_cam3.mp4
+34353740-0ce5-4b2f-9394-f9fd9a19b8fe/ep3/stereo_left.mp4
+34353740-0ce5-4b2f-9394-f9fd9a19b8fe/ep3/stereo_right.mp4
+4e3da960-c06f-4b2c-996d-fe6f04fa4e36/ep2/annotation.hdf5
+4e3da960-c06f-4b2c-996d-fe6f04fa4e36/ep2/fisheye_cam0.mp4
+4e3da960-c06f-4b2c-996d-fe6f04fa4e36/ep2/fisheye_cam1.mp4
+4e3da960-c06f-4b2c-996d-fe6f04fa4e36/ep2/fisheye_cam2.mp4
+4e3da960-c06f-4b2c-996d-fe6f04fa4e36/ep2/fisheye_cam3.mp4
+4e3da960-c06f-4b2c-996d-fe6f04fa4e36/ep2/stereo_left.mp4
+4e3da960-c06f-4b2c-996d-fe6f04fa4e36/ep2/stereo_right.mp4
+2a433534-63d5-48ad-b75e-a5607087b5df/ep2/annotation.hdf5
+2a433534-63d5-48ad-b75e-a5607087b5df/ep2/fisheye_cam0.mp4
+2a433534-63d5-48ad-b75e-a5607087b5df/ep2/fisheye_cam1.mp4
+2a433534-63d5-48ad-b75e-a5607087b5df/ep2/fisheye_cam2.mp4
+2a433534-63d5-48ad-b75e-a5607087b5df/ep2/fisheye_cam3.mp4
+2a433534-63d5-48ad-b75e-a5607087b5df/ep2/stereo_left.mp4
+2a433534-63d5-48ad-b75e-a5607087b5df/ep2/stereo_right.mp4
+480cb308-6ad0-4791-8bb6-f029f439548a/ep5/annotation.hdf5
+480cb308-6ad0-4791-8bb6-f029f439548a/ep5/fisheye_cam0.mp4
+480cb308-6ad0-4791-8bb6-f029f439548a/ep5/fisheye_cam1.mp4
+480cb308-6ad0-4791-8bb6-f029f439548a/ep5/fisheye_cam2.mp4
+480cb308-6ad0-4791-8bb6-f029f439548a/ep5/fisheye_cam3.mp4
+480cb308-6ad0-4791-8bb6-f029f439548a/ep5/stereo_left.mp4
+480cb308-6ad0-4791-8bb6-f029f439548a/ep5/stereo_right.mp4
+613e0f29-cc10-4d0c-b0ce-e39fe9c8a6b9/ep2/annotation.hdf5
+613e0f29-cc10-4d0c-b0ce-e39fe9c8a6b9/ep2/fisheye_cam0.mp4
+613e0f29-cc10-4d0c-b0ce-e39fe9c8a6b9/ep2/fisheye_cam1.mp4
+613e0f29-cc10-4d0c-b0ce-e39fe9c8a6b9/ep2/fisheye_cam2.mp4
+613e0f29-cc10-4d0c-b0ce-e39fe9c8a6b9/ep2/fisheye_cam3.mp4
+613e0f29-cc10-4d0c-b0ce-e39fe9c8a6b9/ep2/stereo_left.mp4
+613e0f29-cc10-4d0c-b0ce-e39fe9c8a6b9/ep2/stereo_right.mp4
+3efd6a52-1457-4437-a3bf-9d5894614fe6/ep4/annotation.hdf5
+3efd6a52-1457-4437-a3bf-9d5894614fe6/ep4/fisheye_cam0.mp4
+3efd6a52-1457-4437-a3bf-9d5894614fe6/ep4/fisheye_cam1.mp4
+3efd6a52-1457-4437-a3bf-9d5894614fe6/ep4/fisheye_cam2.mp4
+3efd6a52-1457-4437-a3bf-9d5894614fe6/ep4/fisheye_cam3.mp4
+3efd6a52-1457-4437-a3bf-9d5894614fe6/ep4/stereo_left.mp4
+3efd6a52-1457-4437-a3bf-9d5894614fe6/ep4/stereo_right.mp4
+17901b9e-7d5d-475a-86f8-b4a9bc9a5709/ep2/annotation.hdf5
+17901b9e-7d5d-475a-86f8-b4a9bc9a5709/ep2/fisheye_cam0.mp4
+17901b9e-7d5d-475a-86f8-b4a9bc9a5709/ep2/fisheye_cam1.mp4
+17901b9e-7d5d-475a-86f8-b4a9bc9a5709/ep2/fisheye_cam2.mp4
+17901b9e-7d5d-475a-86f8-b4a9bc9a5709/ep2/fisheye_cam3.mp4
+17901b9e-7d5d-475a-86f8-b4a9bc9a5709/ep2/stereo_left.mp4
+17901b9e-7d5d-475a-86f8-b4a9bc9a5709/ep2/stereo_right.mp4
+628d6ea4-0a63-4d8f-9725-1e41d2a56b8c/ep5/annotation.hdf5
+628d6ea4-0a63-4d8f-9725-1e41d2a56b8c/ep5/fisheye_cam0.mp4
+628d6ea4-0a63-4d8f-9725-1e41d2a56b8c/ep5/fisheye_cam1.mp4
+628d6ea4-0a63-4d8f-9725-1e41d2a56b8c/ep5/fisheye_cam2.mp4
+628d6ea4-0a63-4d8f-9725-1e41d2a56b8c/ep5/fisheye_cam3.mp4
+628d6ea4-0a63-4d8f-9725-1e41d2a56b8c/ep5/stereo_left.mp4
+628d6ea4-0a63-4d8f-9725-1e41d2a56b8c/ep5/stereo_right.mp4
+960e7265-e8f2-4bca-9c21-501e2c250ce8/ep2/annotation.hdf5
+960e7265-e8f2-4bca-9c21-501e2c250ce8/ep2/fisheye_cam0.mp4
+960e7265-e8f2-4bca-9c21-501e2c250ce8/ep2/fisheye_cam1.mp4
+960e7265-e8f2-4bca-9c21-501e2c250ce8/ep2/fisheye_cam2.mp4
+960e7265-e8f2-4bca-9c21-501e2c250ce8/ep2/fisheye_cam3.mp4
+960e7265-e8f2-4bca-9c21-501e2c250ce8/ep2/stereo_left.mp4
+960e7265-e8f2-4bca-9c21-501e2c250ce8/ep2/stereo_right.mp4
+70a56200-9f80-4f13-8adb-33c0f931e1de/ep2/annotation.hdf5
+70a56200-9f80-4f13-8adb-33c0f931e1de/ep2/fisheye_cam0.mp4
+70a56200-9f80-4f13-8adb-33c0f931e1de/ep2/fisheye_cam1.mp4
+70a56200-9f80-4f13-8adb-33c0f931e1de/ep2/fisheye_cam2.mp4
+70a56200-9f80-4f13-8adb-33c0f931e1de/ep2/fisheye_cam3.mp4
+70a56200-9f80-4f13-8adb-33c0f931e1de/ep2/stereo_left.mp4
+70a56200-9f80-4f13-8adb-33c0f931e1de/ep2/stereo_right.mp4
+833144f1-bd03-4b93-bdc7-3d2e46aedcf1/ep2/annotation.hdf5
+833144f1-bd03-4b93-bdc7-3d2e46aedcf1/ep2/fisheye_cam0.mp4
+833144f1-bd03-4b93-bdc7-3d2e46aedcf1/ep2/fisheye_cam1.mp4
+833144f1-bd03-4b93-bdc7-3d2e46aedcf1/ep2/fisheye_cam2.mp4
+833144f1-bd03-4b93-bdc7-3d2e46aedcf1/ep2/fisheye_cam3.mp4
+833144f1-bd03-4b93-bdc7-3d2e46aedcf1/ep2/stereo_left.mp4
+833144f1-bd03-4b93-bdc7-3d2e46aedcf1/ep2/stereo_right.mp4
+573a5795-03bf-4f16-b4ce-bc09c2c56476/ep3/annotation.hdf5
+573a5795-03bf-4f16-b4ce-bc09c2c56476/ep3/fisheye_cam0.mp4
+573a5795-03bf-4f16-b4ce-bc09c2c56476/ep3/fisheye_cam1.mp4
+573a5795-03bf-4f16-b4ce-bc09c2c56476/ep3/fisheye_cam2.mp4
+573a5795-03bf-4f16-b4ce-bc09c2c56476/ep3/fisheye_cam3.mp4
+573a5795-03bf-4f16-b4ce-bc09c2c56476/ep3/stereo_left.mp4
+573a5795-03bf-4f16-b4ce-bc09c2c56476/ep3/stereo_right.mp4
+b9dd769b-e31a-4fdb-945e-5a60db6487b0/ep2/annotation.hdf5
+b9dd769b-e31a-4fdb-945e-5a60db6487b0/ep2/fisheye_cam0.mp4
+b9dd769b-e31a-4fdb-945e-5a60db6487b0/ep2/fisheye_cam1.mp4
+b9dd769b-e31a-4fdb-945e-5a60db6487b0/ep2/fisheye_cam2.mp4
+b9dd769b-e31a-4fdb-945e-5a60db6487b0/ep2/fisheye_cam3.mp4
+b9dd769b-e31a-4fdb-945e-5a60db6487b0/ep2/stereo_left.mp4
+b9dd769b-e31a-4fdb-945e-5a60db6487b0/ep2/stereo_right.mp4
+c304368b-3b2c-46fc-8a19-c1de1e13dc75/ep4/annotation.hdf5
+c304368b-3b2c-46fc-8a19-c1de1e13dc75/ep4/fisheye_cam0.mp4
+c304368b-3b2c-46fc-8a19-c1de1e13dc75/ep4/fisheye_cam1.mp4
+c304368b-3b2c-46fc-8a19-c1de1e13dc75/ep4/fisheye_cam2.mp4
+c304368b-3b2c-46fc-8a19-c1de1e13dc75/ep4/fisheye_cam3.mp4
+c304368b-3b2c-46fc-8a19-c1de1e13dc75/ep4/stereo_left.mp4
+c304368b-3b2c-46fc-8a19-c1de1e13dc75/ep4/stereo_right.mp4
+ba045ed4-ef25-404d-b756-8dcbd45b18fa/ep2/annotation.hdf5
+ba045ed4-ef25-404d-b756-8dcbd45b18fa/ep2/fisheye_cam0.mp4
+ba045ed4-ef25-404d-b756-8dcbd45b18fa/ep2/fisheye_cam1.mp4
+ba045ed4-ef25-404d-b756-8dcbd45b18fa/ep2/fisheye_cam2.mp4
+ba045ed4-ef25-404d-b756-8dcbd45b18fa/ep2/fisheye_cam3.mp4
+ba045ed4-ef25-404d-b756-8dcbd45b18fa/ep2/stereo_left.mp4
+ba045ed4-ef25-404d-b756-8dcbd45b18fa/ep2/stereo_right.mp4
+2651c600-a427-42a5-bdc9-6e8ab74356eb/ep5/annotation.hdf5
+2651c600-a427-42a5-bdc9-6e8ab74356eb/ep5/fisheye_cam0.mp4
+2651c600-a427-42a5-bdc9-6e8ab74356eb/ep5/fisheye_cam1.mp4
+2651c600-a427-42a5-bdc9-6e8ab74356eb/ep5/fisheye_cam2.mp4
+2651c600-a427-42a5-bdc9-6e8ab74356eb/ep5/fisheye_cam3.mp4
+2651c600-a427-42a5-bdc9-6e8ab74356eb/ep5/stereo_left.mp4
+2651c600-a427-42a5-bdc9-6e8ab74356eb/ep5/stereo_right.mp4
+dab5e5d1-9d11-43ad-ac6f-205500ff5ad4/ep5/annotation.hdf5
+dab5e5d1-9d11-43ad-ac6f-205500ff5ad4/ep5/fisheye_cam0.mp4
+dab5e5d1-9d11-43ad-ac6f-205500ff5ad4/ep5/fisheye_cam1.mp4
+dab5e5d1-9d11-43ad-ac6f-205500ff5ad4/ep5/fisheye_cam2.mp4
+dab5e5d1-9d11-43ad-ac6f-205500ff5ad4/ep5/fisheye_cam3.mp4
+dab5e5d1-9d11-43ad-ac6f-205500ff5ad4/ep5/stereo_left.mp4
+dab5e5d1-9d11-43ad-ac6f-205500ff5ad4/ep5/stereo_right.mp4
+a2f2e433-d2c6-4ae1-80b9-c775a8428aa8/ep2/annotation.hdf5
+a2f2e433-d2c6-4ae1-80b9-c775a8428aa8/ep2/fisheye_cam0.mp4
+a2f2e433-d2c6-4ae1-80b9-c775a8428aa8/ep2/fisheye_cam1.mp4
+a2f2e433-d2c6-4ae1-80b9-c775a8428aa8/ep2/fisheye_cam2.mp4
+a2f2e433-d2c6-4ae1-80b9-c775a8428aa8/ep2/fisheye_cam3.mp4
+a2f2e433-d2c6-4ae1-80b9-c775a8428aa8/ep2/stereo_left.mp4
+a2f2e433-d2c6-4ae1-80b9-c775a8428aa8/ep2/stereo_right.mp4
+2f918260-6ba7-4132-8069-7972afedc1cf/ep3/annotation.hdf5
+2f918260-6ba7-4132-8069-7972afedc1cf/ep3/fisheye_cam0.mp4
+2f918260-6ba7-4132-8069-7972afedc1cf/ep3/fisheye_cam1.mp4
+2f918260-6ba7-4132-8069-7972afedc1cf/ep3/fisheye_cam2.mp4
+2f918260-6ba7-4132-8069-7972afedc1cf/ep3/fisheye_cam3.mp4
+2f918260-6ba7-4132-8069-7972afedc1cf/ep3/stereo_left.mp4
+2f918260-6ba7-4132-8069-7972afedc1cf/ep3/stereo_right.mp4
+4b02bb38-384a-438a-b5f9-6131d85c34b0/ep1/annotation.hdf5
+4b02bb38-384a-438a-b5f9-6131d85c34b0/ep1/fisheye_cam0.mp4
+4b02bb38-384a-438a-b5f9-6131d85c34b0/ep1/fisheye_cam1.mp4
+4b02bb38-384a-438a-b5f9-6131d85c34b0/ep1/fisheye_cam2.mp4
+4b02bb38-384a-438a-b5f9-6131d85c34b0/ep1/fisheye_cam3.mp4
+4b02bb38-384a-438a-b5f9-6131d85c34b0/ep1/stereo_left.mp4
+4b02bb38-384a-438a-b5f9-6131d85c34b0/ep1/stereo_right.mp4
+13ca0290-713b-43a8-a06c-305bde0cbf6c/ep2/annotation.hdf5
+13ca0290-713b-43a8-a06c-305bde0cbf6c/ep2/fisheye_cam0.mp4
+13ca0290-713b-43a8-a06c-305bde0cbf6c/ep2/fisheye_cam1.mp4
+13ca0290-713b-43a8-a06c-305bde0cbf6c/ep2/fisheye_cam2.mp4
+13ca0290-713b-43a8-a06c-305bde0cbf6c/ep2/fisheye_cam3.mp4
+13ca0290-713b-43a8-a06c-305bde0cbf6c/ep2/stereo_left.mp4
+13ca0290-713b-43a8-a06c-305bde0cbf6c/ep2/stereo_right.mp4
+0a5009c4-292b-40c6-b9ec-7c75cf54a112/ep1/annotation.hdf5
+0a5009c4-292b-40c6-b9ec-7c75cf54a112/ep1/fisheye_cam0.mp4
+0a5009c4-292b-40c6-b9ec-7c75cf54a112/ep1/fisheye_cam1.mp4
+0a5009c4-292b-40c6-b9ec-7c75cf54a112/ep1/fisheye_cam2.mp4
+0a5009c4-292b-40c6-b9ec-7c75cf54a112/ep1/fisheye_cam3.mp4
+0a5009c4-292b-40c6-b9ec-7c75cf54a112/ep1/stereo_left.mp4
+0a5009c4-292b-40c6-b9ec-7c75cf54a112/ep1/stereo_right.mp4
+b750fab3-7fbb-43a0-b451-c64c4d4a64da/ep1/annotation.hdf5
+b750fab3-7fbb-43a0-b451-c64c4d4a64da/ep1/fisheye_cam0.mp4
+b750fab3-7fbb-43a0-b451-c64c4d4a64da/ep1/fisheye_cam1.mp4
+b750fab3-7fbb-43a0-b451-c64c4d4a64da/ep1/fisheye_cam2.mp4
+b750fab3-7fbb-43a0-b451-c64c4d4a64da/ep1/fisheye_cam3.mp4
+b750fab3-7fbb-43a0-b451-c64c4d4a64da/ep1/stereo_left.mp4
+b750fab3-7fbb-43a0-b451-c64c4d4a64da/ep1/stereo_right.mp4
+9c7b2c19-5370-4a81-af68-6b76723f3c67/ep1/annotation.hdf5
+9c7b2c19-5370-4a81-af68-6b76723f3c67/ep1/fisheye_cam0.mp4
+9c7b2c19-5370-4a81-af68-6b76723f3c67/ep1/fisheye_cam1.mp4
+9c7b2c19-5370-4a81-af68-6b76723f3c67/ep1/fisheye_cam2.mp4
+9c7b2c19-5370-4a81-af68-6b76723f3c67/ep1/fisheye_cam3.mp4
+9c7b2c19-5370-4a81-af68-6b76723f3c67/ep1/stereo_left.mp4
+9c7b2c19-5370-4a81-af68-6b76723f3c67/ep1/stereo_right.mp4
+a39eaab1-ac8c-4c2d-9764-a566b773136e/ep2/annotation.hdf5
+a39eaab1-ac8c-4c2d-9764-a566b773136e/ep2/fisheye_cam0.mp4
+a39eaab1-ac8c-4c2d-9764-a566b773136e/ep2/fisheye_cam1.mp4
+a39eaab1-ac8c-4c2d-9764-a566b773136e/ep2/fisheye_cam2.mp4
+a39eaab1-ac8c-4c2d-9764-a566b773136e/ep2/fisheye_cam3.mp4
+a39eaab1-ac8c-4c2d-9764-a566b773136e/ep2/stereo_left.mp4
+a39eaab1-ac8c-4c2d-9764-a566b773136e/ep2/stereo_right.mp4
+53e64c93-8947-403e-ab35-3428c431aee8/ep2/annotation.hdf5
+53e64c93-8947-403e-ab35-3428c431aee8/ep2/fisheye_cam0.mp4
+53e64c93-8947-403e-ab35-3428c431aee8/ep2/fisheye_cam1.mp4
+53e64c93-8947-403e-ab35-3428c431aee8/ep2/fisheye_cam2.mp4
+53e64c93-8947-403e-ab35-3428c431aee8/ep2/fisheye_cam3.mp4
+53e64c93-8947-403e-ab35-3428c431aee8/ep2/stereo_left.mp4
+53e64c93-8947-403e-ab35-3428c431aee8/ep2/stereo_right.mp4
+7ba21b77-14c9-47c5-8942-ef2ef5e02784/ep3/annotation.hdf5
+7ba21b77-14c9-47c5-8942-ef2ef5e02784/ep3/fisheye_cam0.mp4
+7ba21b77-14c9-47c5-8942-ef2ef5e02784/ep3/fisheye_cam1.mp4
+7ba21b77-14c9-47c5-8942-ef2ef5e02784/ep3/fisheye_cam2.mp4
+7ba21b77-14c9-47c5-8942-ef2ef5e02784/ep3/fisheye_cam3.mp4
+7ba21b77-14c9-47c5-8942-ef2ef5e02784/ep3/stereo_left.mp4
+7ba21b77-14c9-47c5-8942-ef2ef5e02784/ep3/stereo_right.mp4
+d377bbb4-1bce-43ae-a41f-cefd23b8a8a2/ep2/annotation.hdf5
+d377bbb4-1bce-43ae-a41f-cefd23b8a8a2/ep2/fisheye_cam0.mp4
+d377bbb4-1bce-43ae-a41f-cefd23b8a8a2/ep2/fisheye_cam1.mp4
+d377bbb4-1bce-43ae-a41f-cefd23b8a8a2/ep2/fisheye_cam2.mp4
+d377bbb4-1bce-43ae-a41f-cefd23b8a8a2/ep2/fisheye_cam3.mp4
+d377bbb4-1bce-43ae-a41f-cefd23b8a8a2/ep2/stereo_left.mp4
+d377bbb4-1bce-43ae-a41f-cefd23b8a8a2/ep2/stereo_right.mp4
+c221c115-7b98-4d08-b5ab-653b5956c811/ep3/annotation.hdf5
+c221c115-7b98-4d08-b5ab-653b5956c811/ep3/fisheye_cam0.mp4
+c221c115-7b98-4d08-b5ab-653b5956c811/ep3/fisheye_cam1.mp4
+c221c115-7b98-4d08-b5ab-653b5956c811/ep3/fisheye_cam2.mp4
+c221c115-7b98-4d08-b5ab-653b5956c811/ep3/fisheye_cam3.mp4
+c221c115-7b98-4d08-b5ab-653b5956c811/ep3/stereo_left.mp4
+c221c115-7b98-4d08-b5ab-653b5956c811/ep3/stereo_right.mp4
+a90b8b61-cbba-40fa-948c-b3da79fec1cb/ep1/annotation.hdf5
+a90b8b61-cbba-40fa-948c-b3da79fec1cb/ep1/fisheye_cam0.mp4
+a90b8b61-cbba-40fa-948c-b3da79fec1cb/ep1/fisheye_cam1.mp4
+a90b8b61-cbba-40fa-948c-b3da79fec1cb/ep1/fisheye_cam2.mp4
+a90b8b61-cbba-40fa-948c-b3da79fec1cb/ep1/fisheye_cam3.mp4
+a90b8b61-cbba-40fa-948c-b3da79fec1cb/ep1/stereo_left.mp4
+a90b8b61-cbba-40fa-948c-b3da79fec1cb/ep1/stereo_right.mp4
+7258c332-1061-4c43-8242-f438d951db1f/ep1/annotation.hdf5
+7258c332-1061-4c43-8242-f438d951db1f/ep1/fisheye_cam0.mp4
+7258c332-1061-4c43-8242-f438d951db1f/ep1/fisheye_cam1.mp4
+7258c332-1061-4c43-8242-f438d951db1f/ep1/fisheye_cam2.mp4
+7258c332-1061-4c43-8242-f438d951db1f/ep1/fisheye_cam3.mp4
+7258c332-1061-4c43-8242-f438d951db1f/ep1/stereo_left.mp4
+7258c332-1061-4c43-8242-f438d951db1f/ep1/stereo_right.mp4
+22af1ff1-eaa9-4268-b96a-81a9fa35fe93/ep3/annotation.hdf5
+22af1ff1-eaa9-4268-b96a-81a9fa35fe93/ep3/fisheye_cam0.mp4
+22af1ff1-eaa9-4268-b96a-81a9fa35fe93/ep3/fisheye_cam1.mp4
+22af1ff1-eaa9-4268-b96a-81a9fa35fe93/ep3/fisheye_cam2.mp4
+22af1ff1-eaa9-4268-b96a-81a9fa35fe93/ep3/fisheye_cam3.mp4
+22af1ff1-eaa9-4268-b96a-81a9fa35fe93/ep3/stereo_left.mp4
+22af1ff1-eaa9-4268-b96a-81a9fa35fe93/ep3/stereo_right.mp4
+2680606b-f296-4daa-baf9-7d58c90444bc/ep2/annotation.hdf5
+2680606b-f296-4daa-baf9-7d58c90444bc/ep2/fisheye_cam0.mp4
+2680606b-f296-4daa-baf9-7d58c90444bc/ep2/fisheye_cam1.mp4
+2680606b-f296-4daa-baf9-7d58c90444bc/ep2/fisheye_cam2.mp4
+2680606b-f296-4daa-baf9-7d58c90444bc/ep2/fisheye_cam3.mp4
+2680606b-f296-4daa-baf9-7d58c90444bc/ep2/stereo_left.mp4
+2680606b-f296-4daa-baf9-7d58c90444bc/ep2/stereo_right.mp4
+21d33146-cd1f-40bd-b26b-0c64fd1d603f/ep2/annotation.hdf5
+21d33146-cd1f-40bd-b26b-0c64fd1d603f/ep2/fisheye_cam0.mp4
+21d33146-cd1f-40bd-b26b-0c64fd1d603f/ep2/fisheye_cam1.mp4
+21d33146-cd1f-40bd-b26b-0c64fd1d603f/ep2/fisheye_cam2.mp4
+21d33146-cd1f-40bd-b26b-0c64fd1d603f/ep2/fisheye_cam3.mp4
+21d33146-cd1f-40bd-b26b-0c64fd1d603f/ep2/stereo_left.mp4
+21d33146-cd1f-40bd-b26b-0c64fd1d603f/ep2/stereo_right.mp4
+48b299c1-6c08-40a5-9b9d-9d170647a15a/ep1/annotation.hdf5
+48b299c1-6c08-40a5-9b9d-9d170647a15a/ep1/fisheye_cam0.mp4
+48b299c1-6c08-40a5-9b9d-9d170647a15a/ep1/fisheye_cam1.mp4
+48b299c1-6c08-40a5-9b9d-9d170647a15a/ep1/fisheye_cam2.mp4
+48b299c1-6c08-40a5-9b9d-9d170647a15a/ep1/fisheye_cam3.mp4
+48b299c1-6c08-40a5-9b9d-9d170647a15a/ep1/stereo_left.mp4
+48b299c1-6c08-40a5-9b9d-9d170647a15a/ep1/stereo_right.mp4
+4ab31463-5a47-4904-82c4-d859a0506e04/ep4/annotation.hdf5
+4ab31463-5a47-4904-82c4-d859a0506e04/ep4/fisheye_cam0.mp4
+4ab31463-5a47-4904-82c4-d859a0506e04/ep4/fisheye_cam1.mp4
+4ab31463-5a47-4904-82c4-d859a0506e04/ep4/fisheye_cam2.mp4
+4ab31463-5a47-4904-82c4-d859a0506e04/ep4/fisheye_cam3.mp4
+4ab31463-5a47-4904-82c4-d859a0506e04/ep4/stereo_left.mp4
+4ab31463-5a47-4904-82c4-d859a0506e04/ep4/stereo_right.mp4
+175a573b-7294-43b4-9c14-09c29ab3fef0/ep2/annotation.hdf5
+175a573b-7294-43b4-9c14-09c29ab3fef0/ep2/fisheye_cam0.mp4
+175a573b-7294-43b4-9c14-09c29ab3fef0/ep2/fisheye_cam1.mp4
+175a573b-7294-43b4-9c14-09c29ab3fef0/ep2/fisheye_cam2.mp4
+175a573b-7294-43b4-9c14-09c29ab3fef0/ep2/fisheye_cam3.mp4
+175a573b-7294-43b4-9c14-09c29ab3fef0/ep2/stereo_left.mp4
+175a573b-7294-43b4-9c14-09c29ab3fef0/ep2/stereo_right.mp4
+34e4d6f3-f2b4-4de1-b837-15a372512a90/ep2/annotation.hdf5
+34e4d6f3-f2b4-4de1-b837-15a372512a90/ep2/fisheye_cam0.mp4
+34e4d6f3-f2b4-4de1-b837-15a372512a90/ep2/fisheye_cam1.mp4
+34e4d6f3-f2b4-4de1-b837-15a372512a90/ep2/fisheye_cam2.mp4
+34e4d6f3-f2b4-4de1-b837-15a372512a90/ep2/fisheye_cam3.mp4
+34e4d6f3-f2b4-4de1-b837-15a372512a90/ep2/stereo_left.mp4
+34e4d6f3-f2b4-4de1-b837-15a372512a90/ep2/stereo_right.mp4
+bdfae17b-1e56-4cc5-a4dc-f4fad8369aed/ep1/annotation.hdf5
+bdfae17b-1e56-4cc5-a4dc-f4fad8369aed/ep1/fisheye_cam0.mp4
+bdfae17b-1e56-4cc5-a4dc-f4fad8369aed/ep1/fisheye_cam1.mp4
+bdfae17b-1e56-4cc5-a4dc-f4fad8369aed/ep1/fisheye_cam2.mp4
+bdfae17b-1e56-4cc5-a4dc-f4fad8369aed/ep1/fisheye_cam3.mp4
+bdfae17b-1e56-4cc5-a4dc-f4fad8369aed/ep1/stereo_left.mp4
+bdfae17b-1e56-4cc5-a4dc-f4fad8369aed/ep1/stereo_right.mp4
+1a76d123-ed27-4c4a-a8d4-06f3aaeda454/ep1/annotation.hdf5
+1a76d123-ed27-4c4a-a8d4-06f3aaeda454/ep1/fisheye_cam0.mp4
+1a76d123-ed27-4c4a-a8d4-06f3aaeda454/ep1/fisheye_cam1.mp4
+1a76d123-ed27-4c4a-a8d4-06f3aaeda454/ep1/fisheye_cam2.mp4
+1a76d123-ed27-4c4a-a8d4-06f3aaeda454/ep1/fisheye_cam3.mp4
+1a76d123-ed27-4c4a-a8d4-06f3aaeda454/ep1/stereo_left.mp4
+1a76d123-ed27-4c4a-a8d4-06f3aaeda454/ep1/stereo_right.mp4
+aed79382-cf3f-4b5c-89bb-46c322c19350/ep3/annotation.hdf5
+aed79382-cf3f-4b5c-89bb-46c322c19350/ep3/fisheye_cam0.mp4
+aed79382-cf3f-4b5c-89bb-46c322c19350/ep3/fisheye_cam1.mp4
+aed79382-cf3f-4b5c-89bb-46c322c19350/ep3/fisheye_cam2.mp4
+aed79382-cf3f-4b5c-89bb-46c322c19350/ep3/fisheye_cam3.mp4
+aed79382-cf3f-4b5c-89bb-46c322c19350/ep3/stereo_left.mp4
+aed79382-cf3f-4b5c-89bb-46c322c19350/ep3/stereo_right.mp4
+0073a4be-ec41-47f3-92d6-f63cf89aec9b/ep4/annotation.hdf5
+0073a4be-ec41-47f3-92d6-f63cf89aec9b/ep4/fisheye_cam0.mp4
+0073a4be-ec41-47f3-92d6-f63cf89aec9b/ep4/fisheye_cam1.mp4
+0073a4be-ec41-47f3-92d6-f63cf89aec9b/ep4/fisheye_cam2.mp4
+0073a4be-ec41-47f3-92d6-f63cf89aec9b/ep4/fisheye_cam3.mp4
+0073a4be-ec41-47f3-92d6-f63cf89aec9b/ep4/stereo_left.mp4
+0073a4be-ec41-47f3-92d6-f63cf89aec9b/ep4/stereo_right.mp4
+4a12e289-e4ed-405c-b198-96f704e94276/ep4/annotation.hdf5
+4a12e289-e4ed-405c-b198-96f704e94276/ep4/fisheye_cam0.mp4
+4a12e289-e4ed-405c-b198-96f704e94276/ep4/fisheye_cam1.mp4
+4a12e289-e4ed-405c-b198-96f704e94276/ep4/fisheye_cam2.mp4
+4a12e289-e4ed-405c-b198-96f704e94276/ep4/fisheye_cam3.mp4
+4a12e289-e4ed-405c-b198-96f704e94276/ep4/stereo_left.mp4
+4a12e289-e4ed-405c-b198-96f704e94276/ep4/stereo_right.mp4
+1796b943-caad-43c6-b9bd-80b8d601f37d/ep1/annotation.hdf5
+1796b943-caad-43c6-b9bd-80b8d601f37d/ep1/fisheye_cam0.mp4
+1796b943-caad-43c6-b9bd-80b8d601f37d/ep1/fisheye_cam1.mp4
+1796b943-caad-43c6-b9bd-80b8d601f37d/ep1/fisheye_cam2.mp4
+1796b943-caad-43c6-b9bd-80b8d601f37d/ep1/fisheye_cam3.mp4
+1796b943-caad-43c6-b9bd-80b8d601f37d/ep1/stereo_left.mp4
+1796b943-caad-43c6-b9bd-80b8d601f37d/ep1/stereo_right.mp4
+a93a840b-2d22-4231-9504-87ec4d930aad/ep2/annotation.hdf5
+a93a840b-2d22-4231-9504-87ec4d930aad/ep2/fisheye_cam0.mp4
+a93a840b-2d22-4231-9504-87ec4d930aad/ep2/fisheye_cam1.mp4
+a93a840b-2d22-4231-9504-87ec4d930aad/ep2/fisheye_cam2.mp4
+a93a840b-2d22-4231-9504-87ec4d930aad/ep2/fisheye_cam3.mp4
+a93a840b-2d22-4231-9504-87ec4d930aad/ep2/stereo_left.mp4
+a93a840b-2d22-4231-9504-87ec4d930aad/ep2/stereo_right.mp4
+b2bd37ae-1261-4a9c-82ee-6d50fec3550b/ep1/annotation.hdf5
+b2bd37ae-1261-4a9c-82ee-6d50fec3550b/ep1/fisheye_cam0.mp4
+b2bd37ae-1261-4a9c-82ee-6d50fec3550b/ep1/fisheye_cam1.mp4
+b2bd37ae-1261-4a9c-82ee-6d50fec3550b/ep1/fisheye_cam2.mp4
+b2bd37ae-1261-4a9c-82ee-6d50fec3550b/ep1/fisheye_cam3.mp4
+b2bd37ae-1261-4a9c-82ee-6d50fec3550b/ep1/stereo_left.mp4
+b2bd37ae-1261-4a9c-82ee-6d50fec3550b/ep1/stereo_right.mp4
+b5d76a3d-f4a3-4950-ab9d-344caa247059/ep4/annotation.hdf5
+b5d76a3d-f4a3-4950-ab9d-344caa247059/ep4/fisheye_cam0.mp4
+b5d76a3d-f4a3-4950-ab9d-344caa247059/ep4/fisheye_cam1.mp4
+b5d76a3d-f4a3-4950-ab9d-344caa247059/ep4/fisheye_cam2.mp4
+b5d76a3d-f4a3-4950-ab9d-344caa247059/ep4/fisheye_cam3.mp4
+b5d76a3d-f4a3-4950-ab9d-344caa247059/ep4/stereo_left.mp4
+b5d76a3d-f4a3-4950-ab9d-344caa247059/ep4/stereo_right.mp4
+4318ba68-cc33-455d-bac1-6f8b66558708/ep1/annotation.hdf5
+4318ba68-cc33-455d-bac1-6f8b66558708/ep1/fisheye_cam0.mp4
+4318ba68-cc33-455d-bac1-6f8b66558708/ep1/fisheye_cam1.mp4
+4318ba68-cc33-455d-bac1-6f8b66558708/ep1/fisheye_cam2.mp4
+4318ba68-cc33-455d-bac1-6f8b66558708/ep1/fisheye_cam3.mp4
+4318ba68-cc33-455d-bac1-6f8b66558708/ep1/stereo_left.mp4
+4318ba68-cc33-455d-bac1-6f8b66558708/ep1/stereo_right.mp4
+373850be-6393-4d15-985d-283da6cbe3e7/ep2/annotation.hdf5
+373850be-6393-4d15-985d-283da6cbe3e7/ep2/fisheye_cam0.mp4
+373850be-6393-4d15-985d-283da6cbe3e7/ep2/fisheye_cam1.mp4
+373850be-6393-4d15-985d-283da6cbe3e7/ep2/fisheye_cam2.mp4
+373850be-6393-4d15-985d-283da6cbe3e7/ep2/fisheye_cam3.mp4
+373850be-6393-4d15-985d-283da6cbe3e7/ep2/stereo_left.mp4
+373850be-6393-4d15-985d-283da6cbe3e7/ep2/stereo_right.mp4
+b6579cb5-0a71-4ca6-8808-1e2700be05c7/ep3/annotation.hdf5
+b6579cb5-0a71-4ca6-8808-1e2700be05c7/ep3/fisheye_cam0.mp4
+b6579cb5-0a71-4ca6-8808-1e2700be05c7/ep3/fisheye_cam1.mp4
+b6579cb5-0a71-4ca6-8808-1e2700be05c7/ep3/fisheye_cam2.mp4
+b6579cb5-0a71-4ca6-8808-1e2700be05c7/ep3/fisheye_cam3.mp4
+b6579cb5-0a71-4ca6-8808-1e2700be05c7/ep3/stereo_left.mp4
+b6579cb5-0a71-4ca6-8808-1e2700be05c7/ep3/stereo_right.mp4
+37ee5802-66fb-4893-9173-fe97fe0e2000/ep2/annotation.hdf5
+37ee5802-66fb-4893-9173-fe97fe0e2000/ep2/fisheye_cam0.mp4
+37ee5802-66fb-4893-9173-fe97fe0e2000/ep2/fisheye_cam1.mp4
+37ee5802-66fb-4893-9173-fe97fe0e2000/ep2/fisheye_cam2.mp4
+37ee5802-66fb-4893-9173-fe97fe0e2000/ep2/fisheye_cam3.mp4
+37ee5802-66fb-4893-9173-fe97fe0e2000/ep2/stereo_left.mp4
+37ee5802-66fb-4893-9173-fe97fe0e2000/ep2/stereo_right.mp4

results/omni_finetune/xperience10m_128_episode_selection.csv

@@ -0,0 +1,129 @@
+selection_rank,split,size_band,episode_path,top_level_session,episode_id,annotation_human,training_human,annotation_bytes,training_bytes_excluding_visualization_rrd,has_visualization_rrd,selection_score
+1,test,short,8a8e1b3c-607e-4ada-b3fd-fa639727e92c/ep1,8a8e1b3c-607e-4ada-b3fd-fa639727e92c,ep1,1.58 GiB,2.07 GiB,1692922632,2220826383,False,0.0168743
+2,train,short,27c9fc42-2bb4-4737-b09c-08d2dd88aed4/ep4,27c9fc42-2bb4-4737-b09c-08d2dd88aed4,ep4,1.58 GiB,2.06 GiB,1692213048,2215296867,False,0.01853652
+3,train,short,363abff7-e5ce-425e-a85c-5397186aefd2/ep3,363abff7-e5ce-425e-a85c-5397186aefd2,ep3,1.57 GiB,2.06 GiB,1689936188,2212544693,False,0.01988375
+4,train,short,0e4f59cc-c232-4ad0-9d0c-fed6e026422a/ep3,0e4f59cc-c232-4ad0-9d0c-fed6e026422a,ep3,1.57 GiB,2.08 GiB,1686278728,2233434022,False,0.02007735
+5,train,short,705435da-879d-456e-a28e-f15f86e75027/ep3,705435da-879d-456e-a28e-f15f86e75027,ep3,1.58 GiB,2.04 GiB,1694263288,2190558798,False,0.02126736
+6,train,short,65adf646-cb23-430e-9583-12482dd451a1/ep8,65adf646-cb23-430e-9583-12482dd451a1,ep8,1.58 GiB,2.04 GiB,1692898400,2191724036,False,0.02161242
+7,train,short,0474d134-f983-4590-9cf9-29bd1c6e33cb/ep3,0474d134-f983-4590-9cf9-29bd1c6e33cb,ep3,1.57 GiB,2.09 GiB,1682258880,2238843545,False,0.02162206
+8,test,short,33f7ae08-ac1d-4321-9cb9-eca79016b359/ep1,33f7ae08-ac1d-4321-9cb9-eca79016b359,ep1,1.58 GiB,2.03 GiB,1695175580,2178429798,False,0.02247304
+9,test,short,34f07a04-eb37-45a3-95ec-189ed5f4a85b/ep5,34f07a04-eb37-45a3-95ec-189ed5f4a85b,ep5,1.58 GiB,2.05 GiB,1697525816,2203402325,False,0.02249368
+10,val,short,2f579b88-12fb-41c2-aa2a-9b8f5ae8f918/ep1,2f579b88-12fb-41c2-aa2a-9b8f5ae8f918,ep1,1.58 GiB,2.05 GiB,1699834356,2198942634,False,0.02252887
+11,train,short,0ffa6540-b99f-4909-9f80-c1b3e4c14a85/ep6,0ffa6540-b99f-4909-9f80-c1b3e4c14a85,ep6,1.57 GiB,2.06 GiB,1688274456,2206993440,False,0.02276091
+12,train,short,c5e31792-0961-463c-bfb4-ff89030c2ac8/ep1,c5e31792-0961-463c-bfb4-ff89030c2ac8,ep1,1.59 GiB,2.10 GiB,1711092828,2258479672,False,0.02287723
+13,train,short,6dc018a4-0614-478c-9332-86bd6c4af9e8/ep2,6dc018a4-0614-478c-9332-86bd6c4af9e8,ep2,1.59 GiB,2.08 GiB,1705236060,2230296590,False,0.02316309
+14,train,short,c09b25c6-218c-4e06-95ff-d63eccf20f2b/ep8,c09b25c6-218c-4e06-95ff-d63eccf20f2b,ep8,1.58 GiB,2.07 GiB,1700863160,2222145094,False,0.02336672
+15,train,short,0d330006-684f-4e93-bf38-06d53be3fe5f/ep3,0d330006-684f-4e93-bf38-06d53be3fe5f,ep3,1.58 GiB,2.05 GiB,1700228808,2198320817,False,0.02392015
+16,test,short,cba9c19e-a55f-46e8-bda8-423cdcc2de5f/ep1,cba9c19e-a55f-46e8-bda8-423cdcc2de5f,ep1,1.59 GiB,2.06 GiB,1704207792,2207730397,False,0.02403295
+17,train,short,322958e6-2d26-40d7-af3d-cb16362c78f7/ep2,322958e6-2d26-40d7-af3d-cb16362c78f7,ep2,1.60 GiB,2.10 GiB,1713121844,2258882875,False,0.02431338
+18,val,short,7f443115-54a4-4d50-acfb-e05fe506c270/ep5,7f443115-54a4-4d50-acfb-e05fe506c270,ep5,1.58 GiB,2.06 GiB,1701878284,2207224952,False,0.02445194
+19,train,short,af8e9345-011f-4c97-a78e-ecb4a268220f/ep4,af8e9345-011f-4c97-a78e-ecb4a268220f,ep4,1.57 GiB,2.04 GiB,1687061400,2191905595,False,0.02445481
+20,train,short,7f723cf6-13a4-451d-96bf-d37cbb3f84f3/ep9,7f723cf6-13a4-451d-96bf-d37cbb3f84f3,ep9,1.57 GiB,2.05 GiB,1686618372,2201845958,False,0.0245621
+21,val,short,3ebbfff2-3a60-4cf4-af98-9db074a46e73/ep11,3ebbfff2-3a60-4cf4-af98-9db074a46e73,ep11,1.57 GiB,2.04 GiB,1688702932,2194932059,False,0.02493936
+22,train,short,90fe845b-df65-4297-838c-7924d4c3b6ad/ep9,90fe845b-df65-4297-838c-7924d4c3b6ad,ep9,1.57 GiB,2.03 GiB,1690513852,2181432621,False,0.02519393
+23,train,short,638630bb-caf0-4367-a990-7a3d02dd395a/ep2,638630bb-caf0-4367-a990-7a3d02dd395a,ep2,1.57 GiB,2.05 GiB,1682926752,2195916974,False,0.02529752
+24,train,short,17eea990-0d39-4cc8-a7c8-03fd56b4bf04/ep10,17eea990-0d39-4cc8-a7c8-03fd56b4bf04,ep10,1.57 GiB,2.05 GiB,1687161140,2199860564,False,0.02542125
+25,train,short,81d4c5c5-b164-419c-86e0-c26347284f83/ep9,81d4c5c5-b164-419c-86e0-c26347284f83,ep9,1.58 GiB,2.05 GiB,1701234708,2199961042,False,0.02579083
+26,train,short,85a956ae-65c6-4d11-9e7f-720d6d71ae5b/ep5,85a956ae-65c6-4d11-9e7f-720d6d71ae5b,ep5,1.58 GiB,2.03 GiB,1700584268,2179220839,False,0.02625212
+27,train,short,4b7cc33d-56cf-4029-bda8-f154bd146447/ep5,4b7cc33d-56cf-4029-bda8-f154bd146447,ep5,1.60 GiB,2.13 GiB,1718740864,2284163663,False,0.0266328
+28,val,short,5d994eb3-bc12-46dd-8984-da2d5f1a7055/ep9,5d994eb3-bc12-46dd-8984-da2d5f1a7055,ep9,1.57 GiB,2.03 GiB,1688462140,2177870595,False,0.02669922
+29,train,short,9dc8fc7c-977f-444a-9331-06d2dd7bf120/ep2,9dc8fc7c-977f-444a-9331-06d2dd7bf120,ep2,1.60 GiB,2.12 GiB,1719702876,2272168553,False,0.02685785
+30,train,short,59d0d93f-55c9-4a55-8cd4-e570cc0499d6/ep4,59d0d93f-55c9-4a55-8cd4-e570cc0499d6,ep4,1.59 GiB,2.05 GiB,1704840016,2197860946,False,0.02689026
+31,train,short,acb1b145-804e-41fe-915e-16ff7f59433a/ep7,acb1b145-804e-41fe-915e-16ff7f59433a,ep7,1.57 GiB,2.00 GiB,1690989056,2147820034,False,0.02703099
+32,train,short,0da1f6d2-6564-4f28-b248-1d1b71f9521b/ep1,0da1f6d2-6564-4f28-b248-1d1b71f9521b,ep1,1.56 GiB,2.04 GiB,1679345356,2194155137,True,0.02729302
+33,train,lower_mid,994e4c6c-660a-49f2-90a7-3ebbbb55eb9b/ep2,994e4c6c-660a-49f2-90a7-3ebbbb55eb9b,ep2,1.81 GiB,2.20 GiB,1942653204,2362652603,False,0.00185433
+34,train,lower_mid,53ccaadb-44df-4ddf-bc40-c1bcedb92faa/ep7,53ccaadb-44df-4ddf-bc40-c1bcedb92faa,ep7,1.81 GiB,2.21 GiB,1941724552,2370297326,False,0.00248427
+35,train,lower_mid,d5520c73-fa72-4e92-ad2a-4c3e48d8ad7c/ep6,d5520c73-fa72-4e92-ad2a-4c3e48d8ad7c,ep6,1.81 GiB,2.21 GiB,1941988076,2373419162,False,0.00248503
+36,val,lower_mid,53f2c164-09dd-48ca-ae69-4f5955a902ed/ep4,53f2c164-09dd-48ca-ae69-4f5955a902ed,ep4,1.81 GiB,2.20 GiB,1941348324,2366501574,False,0.00250071
+37,train,lower_mid,a00c1d99-de08-42bc-8623-712b9dde7fde/ep3,a00c1d99-de08-42bc-8623-712b9dde7fde,ep3,1.81 GiB,2.21 GiB,1940612164,2370379355,False,0.00259573
+38,val,lower_mid,d342e6e2-c381-4a1c-b0ac-e62a5c80624f/ep3,d342e6e2-c381-4a1c-b0ac-e62a5c80624f,ep3,1.81 GiB,2.22 GiB,1941941580,2380309709,False,0.00276839
+39,test,lower_mid,46f7dea3-76bd-4de4-a0fe-56c0c1e1a276/ep3,46f7dea3-76bd-4de4-a0fe-56c0c1e1a276,ep3,1.81 GiB,2.19 GiB,1941899132,2356751349,False,0.00297404
+40,train,lower_mid,adb1bb6f-6df3-4f46-b895-f045ac3e0772/ep2,adb1bb6f-6df3-4f46-b895-f045ac3e0772,ep2,1.81 GiB,2.20 GiB,1943209388,2357681735,False,0.00298093
+41,test,lower_mid,5399ef86-4df9-49bc-809f-8f4f92f9e659/ep6,5399ef86-4df9-49bc-809f-8f4f92f9e659,ep6,1.81 GiB,2.21 GiB,1941223988,2372639643,False,0.0032859
+42,train,lower_mid,5aeb0920-ab9f-4dc2-a261-747a678bf9cb/ep2,5aeb0920-ab9f-4dc2-a261-747a678bf9cb,ep2,1.80 GiB,2.21 GiB,1937449580,2369327064,False,0.00330289
+43,test,lower_mid,877779cd-25f3-4293-a3c4-39067dd9558c/ep4,877779cd-25f3-4293-a3c4-39067dd9558c,ep4,1.81 GiB,2.21 GiB,1938930176,2368478759,False,0.00336242
+44,train,lower_mid,9f84a0a7-8964-48c5-b7e1-8a1d28094c86/ep2,9f84a0a7-8964-48c5-b7e1-8a1d28094c86,ep2,1.81 GiB,2.22 GiB,1940541892,2384293317,False,0.00352063
+45,train,lower_mid,84106bb9-1cf8-47b6-8d11-fddbfe381e4f/ep5,84106bb9-1cf8-47b6-8d11-fddbfe381e4f,ep5,1.81 GiB,2.21 GiB,1940012900,2377037440,False,0.00354971
+46,test,lower_mid,ba18b7c1-21ff-45da-8452-41acce7fc8de/ep2,ba18b7c1-21ff-45da-8452-41acce7fc8de,ep2,1.81 GiB,2.20 GiB,1946202432,2364673381,False,0.00356755
+47,train,lower_mid,6fbdecec-d4d5-477a-b94b-2486adb0dbd8/ep6,6fbdecec-d4d5-477a-b94b-2486adb0dbd8,ep6,1.81 GiB,2.22 GiB,1941323844,2383647150,False,0.00372522
+48,train,lower_mid,c5e5183a-9eca-448a-9ec1-a0a40682b81a/ep4,c5e5183a-9eca-448a-9ec1-a0a40682b81a,ep4,1.81 GiB,2.21 GiB,1945464636,2376527573,False,0.00376342
+49,train,lower_mid,003dcaf0-edba-4787-ada0-187d2748f684/ep1,003dcaf0-edba-4787-ada0-187d2748f684,ep1,1.81 GiB,2.23 GiB,1943666280,2391307737,False,0.00383016
+50,train,lower_mid,55f345d5-aa45-4d75-829b-a6ec41dca2ed/ep7,55f345d5-aa45-4d75-829b-a6ec41dca2ed,ep7,1.81 GiB,2.20 GiB,1942904384,2366524153,False,0.00394459
+51,train,lower_mid,5e221284-f4af-4408-9177-25200282b6de/ep4,5e221284-f4af-4408-9177-25200282b6de,ep4,1.81 GiB,2.19 GiB,1941568560,2349841955,False,0.00397446
+52,train,lower_mid,676c3db3-464e-406a-be9f-8283518e8181/ep4,676c3db3-464e-406a-be9f-8283518e8181,ep4,1.81 GiB,2.21 GiB,1945942660,2368716864,False,0.00400177
+53,train,lower_mid,46b507a1-b1cb-4851-8558-9435de4abf68/ep2,46b507a1-b1cb-4851-8558-9435de4abf68,ep2,1.81 GiB,2.21 GiB,1947558424,2373393269,False,0.00400855
+54,val,lower_mid,3cd70c8e-4ef5-481c-aa61-88b0dd4dec1f/ep4,3cd70c8e-4ef5-481c-aa61-88b0dd4dec1f,ep4,1.81 GiB,2.20 GiB,1938699044,2363791408,True,0.00407752
+55,train,lower_mid,aa3ca3ec-56fd-4bbd-ad15-055eabc1838a/ep7,aa3ca3ec-56fd-4bbd-ad15-055eabc1838a,ep7,1.81 GiB,2.21 GiB,1943189644,2376888815,False,0.00414048
+56,train,lower_mid,82c95b73-ba14-4083-8929-ddeb85d22e9f/ep3,82c95b73-ba14-4083-8929-ddeb85d22e9f,ep3,1.81 GiB,2.18 GiB,1942621076,2343431906,False,0.00414466
+57,train,lower_mid,8533707e-d7e0-4d1a-99d1-fa154a8e62ad/ep5,8533707e-d7e0-4d1a-99d1-fa154a8e62ad,ep5,1.81 GiB,2.19 GiB,1941951276,2347121677,False,0.00418534
+58,train,lower_mid,81e9362f-fd63-4aec-92fa-1f489d6ac025/ep2,81e9362f-fd63-4aec-92fa-1f489d6ac025,ep2,1.81 GiB,2.19 GiB,1938830460,2353483140,False,0.00423431
+59,train,lower_mid,cd364707-88fc-43bd-a95b-61aee9f8bd20/ep5,cd364707-88fc-43bd-a95b-61aee9f8bd20,ep5,1.81 GiB,2.20 GiB,1942997976,2357334318,False,0.00436259
+60,val,lower_mid,c038defe-faa6-4fe9-aef6-84bd93726d1c/ep2,c038defe-faa6-4fe9-aef6-84bd93726d1c,ep2,1.81 GiB,2.18 GiB,1942665068,2341438270,False,0.00437939
+61,train,lower_mid,999100c1-b488-4ed1-a580-c0487f308e5d/ep5,999100c1-b488-4ed1-a580-c0487f308e5d,ep5,1.81 GiB,2.19 GiB,1942240680,2346530807,False,0.00456347
+62,train,lower_mid,bf622eaf-363e-40c4-8a04-1dfedd85a718/ep10,bf622eaf-363e-40c4-8a04-1dfedd85a718,ep10,1.81 GiB,2.21 GiB,1940552780,2370820558,False,0.00460553
+63,train,lower_mid,a3e06450-50f8-4894-a22f-1f891fad3809/ep7,a3e06450-50f8-4894-a22f-1f891fad3809,ep7,1.81 GiB,2.23 GiB,1942358600,2391494696,False,0.00471311
+64,train,lower_mid,04d9f378-2812-4c7f-a269-1ad0bc43445f/ep4,04d9f378-2812-4c7f-a269-1ad0bc43445f,ep4,1.81 GiB,2.19 GiB,1939004344,2353759521,False,0.00473298
+65,test,upper_mid,a1012a57-385e-45a9-8a59-694a26fe92a5/ep1,a1012a57-385e-45a9-8a59-694a26fe92a5,ep1,1.84 GiB,2.21 GiB,1973167444,2373049631,False,0.00065677
+66,train,upper_mid,41e90813-8fff-4053-94f0-9537b1393bc2/ep1,41e90813-8fff-4053-94f0-9537b1393bc2,ep1,1.84 GiB,2.21 GiB,1972856680,2370490692,False,0.00077283
+67,train,upper_mid,bc53027e-d29a-473f-8f18-7b524b801e40/ep2,bc53027e-d29a-473f-8f18-7b524b801e40,ep2,1.84 GiB,2.21 GiB,1972811888,2371693205,False,0.00106675
+68,train,upper_mid,727f582a-fe2a-45c3-b506-216133a28c35/ep1,727f582a-fe2a-45c3-b506-216133a28c35,ep1,1.84 GiB,2.21 GiB,1973459728,2370046733,False,0.00107543
+69,test,upper_mid,9c553886-83c5-4dc4-be5c-dcb269b3a771/ep2,9c553886-83c5-4dc4-be5c-dcb269b3a771,ep2,1.84 GiB,2.20 GiB,1973095532,2367490162,False,0.00129493
+70,train,upper_mid,b1292f20-bfca-497b-a41a-e7fd62bbf913/ep2,b1292f20-bfca-497b-a41a-e7fd62bbf913,ep2,1.84 GiB,2.21 GiB,1972779488,2372181505,False,0.00132551
+71,train,upper_mid,b49d8a3c-7a1e-4ca3-8ce8-ae1cce0df001/ep2,b49d8a3c-7a1e-4ca3-8ce8-ae1cce0df001,ep2,1.84 GiB,2.21 GiB,1973103660,2372050429,False,0.00139624
+72,train,upper_mid,8692d5eb-19b2-4c49-9f88-da2ec373796e/ep3,8692d5eb-19b2-4c49-9f88-da2ec373796e,ep3,1.84 GiB,2.21 GiB,1972741192,2368031571,False,0.00164927
+73,train,upper_mid,54bf2b92-a5f8-440c-81fe-8f820bb90aab/ep3,54bf2b92-a5f8-440c-81fe-8f820bb90aab,ep3,1.84 GiB,2.20 GiB,1973088196,2366392177,False,0.00168738
+74,train,upper_mid,3eb8ef88-902d-4e84-8a72-a361c9d18647/ep3,3eb8ef88-902d-4e84-8a72-a361c9d18647,ep3,1.84 GiB,2.20 GiB,1973472960,2365619914,False,0.00170045
+75,train,upper_mid,2aee3428-f462-49e6-95c4-b803c1803e0b/ep3,2aee3428-f462-49e6-95c4-b803c1803e0b,ep3,1.84 GiB,2.21 GiB,1972400252,2374120001,False,0.00173083
+76,train,upper_mid,40611d4c-a36d-4ae3-8a71-da94b81cdce4/ep2,40611d4c-a36d-4ae3-8a71-da94b81cdce4,ep2,1.84 GiB,2.21 GiB,1973489504,2378282476,False,0.00173263
+77,train,upper_mid,832369c5-8fca-4591-940a-7a7a28206de2/ep4,832369c5-8fca-4591-940a-7a7a28206de2,ep4,1.84 GiB,2.21 GiB,1972751704,2370595704,False,0.00180383
+78,train,upper_mid,34353740-0ce5-4b2f-9394-f9fd9a19b8fe/ep3,34353740-0ce5-4b2f-9394-f9fd9a19b8fe,ep3,1.84 GiB,2.20 GiB,1973393888,2366120890,False,0.00181177
+79,val,upper_mid,4e3da960-c06f-4b2c-996d-fe6f04fa4e36/ep2,4e3da960-c06f-4b2c-996d-fe6f04fa4e36,ep2,1.84 GiB,2.20 GiB,1973171028,2363928444,False,0.00182852
+80,train,upper_mid,2a433534-63d5-48ad-b75e-a5607087b5df/ep2,2a433534-63d5-48ad-b75e-a5607087b5df,ep2,1.84 GiB,2.20 GiB,1973438696,2360403950,False,0.00182928
+81,train,upper_mid,480cb308-6ad0-4791-8bb6-f029f439548a/ep5,480cb308-6ad0-4791-8bb6-f029f439548a,ep5,1.84 GiB,2.21 GiB,1973432112,2369690621,False,0.00183373
+82,train,upper_mid,613e0f29-cc10-4d0c-b0ce-e39fe9c8a6b9/ep2,613e0f29-cc10-4d0c-b0ce-e39fe9c8a6b9,ep2,1.84 GiB,2.21 GiB,1973400184,2374898708,False,0.00189528
+83,train,upper_mid,3efd6a52-1457-4437-a3bf-9d5894614fe6/ep4,3efd6a52-1457-4437-a3bf-9d5894614fe6,ep4,1.84 GiB,2.21 GiB,1972496124,2370572574,False,0.00193615
+84,train,upper_mid,17901b9e-7d5d-475a-86f8-b4a9bc9a5709/ep2,17901b9e-7d5d-475a-86f8-b4a9bc9a5709,ep2,1.84 GiB,2.20 GiB,1973423120,2361379626,False,0.00193744
+85,val,upper_mid,628d6ea4-0a63-4d8f-9725-1e41d2a56b8c/ep5,628d6ea4-0a63-4d8f-9725-1e41d2a56b8c,ep5,1.84 GiB,2.21 GiB,1973004896,2373425438,False,0.00196873
+86,train,upper_mid,960e7265-e8f2-4bca-9c21-501e2c250ce8/ep2,960e7265-e8f2-4bca-9c21-501e2c250ce8,ep2,1.84 GiB,2.21 GiB,1973769708,2373277315,False,0.00199583
+87,val,upper_mid,70a56200-9f80-4f13-8adb-33c0f931e1de/ep2,70a56200-9f80-4f13-8adb-33c0f931e1de,ep2,1.84 GiB,2.22 GiB,1973135604,2379117413,False,0.00209202
+88,train,upper_mid,833144f1-bd03-4b93-bdc7-3d2e46aedcf1/ep2,833144f1-bd03-4b93-bdc7-3d2e46aedcf1,ep2,1.84 GiB,2.21 GiB,1973787260,2376248913,False,0.00210965
+89,val,upper_mid,573a5795-03bf-4f16-b4ce-bc09c2c56476/ep3,573a5795-03bf-4f16-b4ce-bc09c2c56476,ep3,1.84 GiB,2.20 GiB,1973069188,2367214820,False,0.00211846
+90,test,upper_mid,b9dd769b-e31a-4fdb-945e-5a60db6487b0/ep2,b9dd769b-e31a-4fdb-945e-5a60db6487b0,ep2,1.84 GiB,2.21 GiB,1972447060,2369200324,False,0.00212378
+91,train,upper_mid,c304368b-3b2c-46fc-8a19-c1de1e13dc75/ep4,c304368b-3b2c-46fc-8a19-c1de1e13dc75,ep4,1.84 GiB,2.20 GiB,1973473820,2364510520,False,0.00218545
+92,test,upper_mid,ba045ed4-ef25-404d-b756-8dcbd45b18fa/ep2,ba045ed4-ef25-404d-b756-8dcbd45b18fa,ep2,1.84 GiB,2.21 GiB,1972771328,2376145891,False,0.0022233
+93,train,upper_mid,2651c600-a427-42a5-bdc9-6e8ab74356eb/ep5,2651c600-a427-42a5-bdc9-6e8ab74356eb,ep5,1.84 GiB,2.21 GiB,1973448776,2370303034,False,0.00222695
+94,train,upper_mid,dab5e5d1-9d11-43ad-ac6f-205500ff5ad4/ep5,dab5e5d1-9d11-43ad-ac6f-205500ff5ad4,ep5,1.84 GiB,2.21 GiB,1972748528,2374102343,False,0.00230824
+95,train,upper_mid,a2f2e433-d2c6-4ae1-80b9-c775a8428aa8/ep2,a2f2e433-d2c6-4ae1-80b9-c775a8428aa8,ep2,1.84 GiB,2.20 GiB,1972763528,2364414703,False,0.00234163
+96,train,upper_mid,2f918260-6ba7-4132-8069-7972afedc1cf/ep3,2f918260-6ba7-4132-8069-7972afedc1cf,ep3,1.84 GiB,2.20 GiB,1973084988,2362737608,False,0.00234955
+97,test,long,4b02bb38-384a-438a-b5f9-6131d85c34b0/ep1,4b02bb38-384a-438a-b5f9-6131d85c34b0,ep1,1.85 GiB,2.21 GiB,1990012504,2371409116,False,0.00077761
+98,train,long,13ca0290-713b-43a8-a06c-305bde0cbf6c/ep2,13ca0290-713b-43a8-a06c-305bde0cbf6c,ep2,1.85 GiB,2.21 GiB,1990040016,2371822099,True,0.00087842
+99,train,long,0a5009c4-292b-40c6-b9ec-7c75cf54a112/ep1,0a5009c4-292b-40c6-b9ec-7c75cf54a112,ep1,1.85 GiB,2.21 GiB,1990088312,2373390647,False,0.00088678
+100,test,long,b750fab3-7fbb-43a0-b451-c64c4d4a64da/ep1,b750fab3-7fbb-43a0-b451-c64c4d4a64da,ep1,1.85 GiB,2.21 GiB,1990096752,2373451302,False,0.00093735
+101,val,long,9c7b2c19-5370-4a81-af68-6b76723f3c67/ep1,9c7b2c19-5370-4a81-af68-6b76723f3c67,ep1,1.85 GiB,2.21 GiB,1989981160,2373822452,False,0.00095494
+102,train,long,a39eaab1-ac8c-4c2d-9764-a566b773136e/ep2,a39eaab1-ac8c-4c2d-9764-a566b773136e,ep2,1.85 GiB,2.21 GiB,1990156336,2369630704,False,0.00110013
+103,train,long,53e64c93-8947-403e-ab35-3428c431aee8/ep2,53e64c93-8947-403e-ab35-3428c431aee8,ep2,1.85 GiB,2.21 GiB,1990008712,2370862777,False,0.00115728
+104,train,long,7ba21b77-14c9-47c5-8942-ef2ef5e02784/ep3,7ba21b77-14c9-47c5-8942-ef2ef5e02784,ep3,1.85 GiB,2.21 GiB,1989995224,2368912779,False,0.00117148
+105,train,long,d377bbb4-1bce-43ae-a41f-cefd23b8a8a2/ep2,d377bbb4-1bce-43ae-a41f-cefd23b8a8a2,ep2,1.85 GiB,2.21 GiB,1990012048,2373424025,False,0.00126935
+106,train,long,c221c115-7b98-4d08-b5ab-653b5956c811/ep3,c221c115-7b98-4d08-b5ab-653b5956c811,ep3,1.85 GiB,2.21 GiB,1989993400,2369721796,False,0.00129683
+107,train,long,a90b8b61-cbba-40fa-948c-b3da79fec1cb/ep1,a90b8b61-cbba-40fa-948c-b3da79fec1cb,ep1,1.85 GiB,2.22 GiB,1990079936,2379881614,False,0.00130509
+108,train,long,7258c332-1061-4c43-8242-f438d951db1f/ep1,7258c332-1061-4c43-8242-f438d951db1f,ep1,1.85 GiB,2.21 GiB,1990054400,2377373319,False,0.00135654
+109,train,long,22af1ff1-eaa9-4268-b96a-81a9fa35fe93/ep3,22af1ff1-eaa9-4268-b96a-81a9fa35fe93,ep3,1.85 GiB,2.20 GiB,1990011096,2367447286,True,0.00137323
+110,train,long,2680606b-f296-4daa-baf9-7d58c90444bc/ep2,2680606b-f296-4daa-baf9-7d58c90444bc,ep2,1.85 GiB,2.20 GiB,1990038936,2363718706,False,0.00137996
+111,train,long,21d33146-cd1f-40bd-b26b-0c64fd1d603f/ep2,21d33146-cd1f-40bd-b26b-0c64fd1d603f,ep2,1.85 GiB,2.21 GiB,1990013872,2374707510,True,0.00145494
+112,train,long,48b299c1-6c08-40a5-9b9d-9d170647a15a/ep1,48b299c1-6c08-40a5-9b9d-9d170647a15a,ep1,1.85 GiB,2.22 GiB,1990034848,2381579147,False,0.00146648
+113,train,long,4ab31463-5a47-4904-82c4-d859a0506e04/ep4,4ab31463-5a47-4904-82c4-d859a0506e04,ep4,1.85 GiB,2.21 GiB,1990021560,2369185550,False,0.00148294
+114,train,long,175a573b-7294-43b4-9c14-09c29ab3fef0/ep2,175a573b-7294-43b4-9c14-09c29ab3fef0,ep2,1.85 GiB,2.21 GiB,1989981736,2375022520,False,0.0014869
+115,val,long,34e4d6f3-f2b4-4de1-b837-15a372512a90/ep2,34e4d6f3-f2b4-4de1-b837-15a372512a90,ep2,1.85 GiB,2.21 GiB,1990037776,2377549966,False,0.00149902
+116,train,long,bdfae17b-1e56-4cc5-a4dc-f4fad8369aed/ep1,bdfae17b-1e56-4cc5-a4dc-f4fad8369aed,ep1,1.85 GiB,2.21 GiB,1990063928,2375884374,False,0.00152423
+117,train,long,1a76d123-ed27-4c4a-a8d4-06f3aaeda454/ep1,1a76d123-ed27-4c4a-a8d4-06f3aaeda454,ep1,1.85 GiB,2.21 GiB,1990022952,2376083841,False,0.00153874
+118,val,long,aed79382-cf3f-4b5c-89bb-46c322c19350/ep3,aed79382-cf3f-4b5c-89bb-46c322c19350,ep3,1.85 GiB,2.21 GiB,1990054104,2375391388,False,0.0015561
+119,train,long,0073a4be-ec41-47f3-92d6-f63cf89aec9b/ep4,0073a4be-ec41-47f3-92d6-f63cf89aec9b,ep4,1.85 GiB,2.21 GiB,1990077792,2369241060,True,0.00155619
+120,train,long,4a12e289-e4ed-405c-b198-96f704e94276/ep4,4a12e289-e4ed-405c-b198-96f704e94276,ep4,1.85 GiB,2.21 GiB,1990061128,2370545689,False,0.00156607
+121,test,long,1796b943-caad-43c6-b9bd-80b8d601f37d/ep1,1796b943-caad-43c6-b9bd-80b8d601f37d,ep1,1.85 GiB,2.20 GiB,1990037360,2366095438,False,0.00156651
+122,train,long,a93a840b-2d22-4231-9504-87ec4d930aad/ep2,a93a840b-2d22-4231-9504-87ec4d930aad,ep2,1.85 GiB,2.21 GiB,1990052256,2371401218,False,0.00157363
+123,train,long,b2bd37ae-1261-4a9c-82ee-6d50fec3550b/ep1,b2bd37ae-1261-4a9c-82ee-6d50fec3550b,ep1,1.85 GiB,2.21 GiB,1990064392,2375230348,False,0.00160525
+124,val,long,b5d76a3d-f4a3-4950-ab9d-344caa247059/ep4,b5d76a3d-f4a3-4950-ab9d-344caa247059,ep4,1.85 GiB,2.21 GiB,1990012120,2371352875,False,0.00160638
+125,train,long,4318ba68-cc33-455d-bac1-6f8b66558708/ep1,4318ba68-cc33-455d-bac1-6f8b66558708,ep1,1.85 GiB,2.22 GiB,1990014448,2382076965,False,0.00162055
+126,train,long,373850be-6393-4d15-985d-283da6cbe3e7/ep2,373850be-6393-4d15-985d-283da6cbe3e7,ep2,1.85 GiB,2.20 GiB,1990003832,2363123735,False,0.00163272
+127,test,long,b6579cb5-0a71-4ca6-8808-1e2700be05c7/ep3,b6579cb5-0a71-4ca6-8808-1e2700be05c7,ep3,1.85 GiB,2.21 GiB,1990036472,2373874306,False,0.00163865
+128,train,long,37ee5802-66fb-4893-9173-fe97fe0e2000/ep2,37ee5802-66fb-4893-9173-fe97fe0e2000,ep2,1.85 GiB,2.21 GiB,1990022864,2375119624,False,0.00166234

scripts/build_public_surface_qa.py

@@ -148,7 +148,7 @@ def build_report() -> dict:
         "Xperience-10M",
         "12-task",
         "Qwen3-Omni",
-        "one public Xperience-10M sample episode",
+        "128-episode relay",
     ]
     hf_link_markers = [
         "https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite",

scripts/omni/analyze_xperience10m_hf_metadata.py

@@ -0,0 +1,442 @@
+#!/usr/bin/env python3
+"""Analyze the gated Xperience-10M HF repo without downloading dataset files."""
+
+from __future__ import annotations
+
+import argparse
+import getpass
+import json
+import os
+from collections import Counter, defaultdict
+from datetime import datetime, timezone
+from pathlib import Path
+from statistics import median
+from typing import Any
+
+from huggingface_hub import HfApi
+
+
+REQUIRED_EPISODE_FILES = [
+    "annotation.hdf5",
+    "fisheye_cam0.mp4",
+    "fisheye_cam1.mp4",
+    "fisheye_cam2.mp4",
+    "fisheye_cam3.mp4",
+    "stereo_left.mp4",
+    "stereo_right.mp4",
+]
+TRAINING_EXCLUDE = {"visualization.rrd"}
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("--repo-id", default="ropedia-ai/xperience-10m")
+    parser.add_argument("--output", type=Path, default=Path("results/omni_finetune/full_dataset_metadata_audit.json"))
+    parser.add_argument("--report-output", type=Path, default=Path("results/omni_finetune/FULL_DATASET_METADATA_AUDIT.md"))
+    parser.add_argument("--token", default=os.environ.get("HF_TOKEN", "").strip())
+    parser.add_argument("--top-n", type=int, default=20)
+    return parser.parse_args()
+
+
+def file_size(sibling: Any) -> int:
+    value = getattr(sibling, "size", None)
+    if isinstance(value, int):
+        return value
+    lfs = getattr(sibling, "lfs", None)
+    if isinstance(lfs, dict) and isinstance(lfs.get("size"), int):
+        return int(lfs["size"])
+    return 0
+
+
+def human_bytes(num: float | int) -> str:
+    value = float(num)
+    for unit in ["B", "KiB", "MiB", "GiB", "TiB", "PiB"]:
+        if abs(value) < 1024.0 or unit == "PiB":
+            return f"{value:.2f} {unit}"
+        value /= 1024.0
+    return f"{value:.2f} PiB"
+
+
+def pct(part: int, whole: int) -> float:
+    return round((part / whole * 100.0), 4) if whole else 0.0
+
+
+def episode_parent(path: str) -> str:
+    return str(Path(path).parent).replace("\\", "/")
+
+
+def summarize_sizes(values: list[int]) -> dict[str, Any]:
+    if not values:
+        return {"count": 0}
+    ordered = sorted(values)
+    q1 = ordered[len(ordered) // 4]
+    q3 = ordered[(len(ordered) * 3) // 4]
+    return {
+        "count": len(values),
+        "min_bytes": ordered[0],
+        "p25_bytes": q1,
+        "median_bytes": int(median(ordered)),
+        "p75_bytes": q3,
+        "max_bytes": ordered[-1],
+        "mean_bytes": int(sum(values) / len(values)),
+        "min_human": human_bytes(ordered[0]),
+        "median_human": human_bytes(median(ordered)),
+        "mean_human": human_bytes(sum(values) / len(values)),
+        "max_human": human_bytes(ordered[-1]),
+    }
+
+
+def near_size_files(files: list[dict[str, Any]], target: int, count: int) -> list[dict[str, Any]]:
+    ranked = sorted(files, key=lambda item: abs(int(item["bytes"]) - target))
+    return ranked[:count]
+
+
+def summarize_numbers(values: list[int]) -> dict[str, Any]:
+    if not values:
+        return {"count": 0}
+    ordered = sorted(values)
+    return {
+        "count": len(values),
+        "min": ordered[0],
+        "p25": ordered[len(ordered) // 4],
+        "median": int(median(ordered)),
+        "p75": ordered[(len(ordered) * 3) // 4],
+        "max": ordered[-1],
+        "mean": round(sum(values) / len(values), 2),
+    }
+
+
+def md_table(headers: list[str], rows: list[list[Any]]) -> list[str]:
+    lines = [
+        "| " + " | ".join(headers) + " |",
+        "| " + " | ".join("---" for _ in headers) + " |",
+    ]
+    lines.extend("| " + " | ".join(str(cell) for cell in row) + " |" for row in rows)
+    return lines
+
+
+def main() -> int:
+    args = parse_args()
+    token = args.token or getpass.getpass("HF token: ").strip()
+    if not token:
+        raise SystemExit("HF token is required for gated dataset metadata.")
+
+    api = HfApi(token=token)
+    info = api.repo_info(
+        repo_id=args.repo_id,
+        repo_type="dataset",
+        files_metadata=True,
+        token=token,
+    )
+    siblings = list(info.siblings or [])
+
+    files = []
+    total_bytes = 0
+    ext_counter: Counter[str] = Counter()
+    basename_counter: Counter[str] = Counter()
+    top_level_counter: Counter[str] = Counter()
+    by_parent: dict[str, dict[str, Any]] = defaultdict(lambda: {"files": {}, "bytes": 0})
+    by_top_level_bytes: Counter[str] = Counter()
+
+    for sibling in siblings:
+        path = str(getattr(sibling, "rfilename", ""))
+        if not path or path == ".gitattributes":
+            continue
+        size = file_size(sibling)
+        total_bytes += size
+        ext = Path(path).suffix.lower() or "<no_ext>"
+        name = Path(path).name
+        top = path.split("/", 1)[0]
+        ext_counter[ext] += 1
+        basename_counter[name] += 1
+        top_level_counter[top] += 1
+        by_top_level_bytes[top] += size
+        parent = episode_parent(path)
+        bucket = by_parent[parent]
+        bucket["files"][name] = {"path": path, "bytes": size}
+        bucket["bytes"] += size
+        files.append({"path": path, "bytes": size, "extension": ext, "basename": name, "top_level": top})
+
+    episode_records = []
+    for parent, bucket in by_parent.items():
+        present = set(bucket["files"])
+        if not (present & set(REQUIRED_EPISODE_FILES)):
+            continue
+        has_annotation = "annotation.hdf5" in present
+        has_fisheye_cam0 = "fisheye_cam0.mp4" in present
+        video_count = sum(1 for name in REQUIRED_EPISODE_FILES[1:] if name in present)
+        missing_required = [name for name in REQUIRED_EPISODE_FILES if name not in present]
+        training_bytes = sum(
+            meta["bytes"]
+            for name, meta in bucket["files"].items()
+            if name not in TRAINING_EXCLUDE
+        )
+        episode_records.append(
+            {
+                "episode_path": parent,
+                "episode_id": Path(parent).name,
+                "top_level_session": parent.split("/", 1)[0],
+                "file_count": len(present),
+                "total_bytes": int(bucket["bytes"]),
+                "training_bytes_excluding_visualization_rrd": int(training_bytes),
+                "has_annotation": has_annotation,
+                "has_fisheye_cam0": has_fisheye_cam0,
+                "video_count": video_count,
+                "has_all_six_videos": video_count == 6,
+                "is_degraded_valid": has_annotation and has_fisheye_cam0,
+                "is_complete": has_annotation and video_count == 6,
+                "has_visualization_rrd": "visualization.rrd" in present,
+                "missing_required_files": missing_required,
+            }
+        )
+
+    complete = [ep for ep in episode_records if ep["is_complete"]]
+    degraded = [ep for ep in episode_records if ep["is_degraded_valid"]]
+    incomplete = [ep for ep in episode_records if not ep["is_complete"]]
+    training_sizes = [ep["training_bytes_excluding_visualization_rrd"] for ep in complete]
+    episode_sizes = [ep["total_bytes"] for ep in episode_records]
+    complete_by_session: Counter[str] = Counter(ep["top_level_session"] for ep in complete)
+    degraded_by_session: Counter[str] = Counter(ep["top_level_session"] for ep in degraded)
+    episode_count_by_session: Counter[str] = Counter(ep["top_level_session"] for ep in episode_records)
+    video_count_hist = Counter(str(ep["video_count"]) for ep in episode_records)
+    rrd_bytes = sum(item["bytes"] for item in files if item["basename"] == "visualization.rrd")
+    all_complete_training_bytes = sum(ep["training_bytes_excluding_visualization_rrd"] for ep in complete)
+    median_32_bytes = int(median(training_sizes)) * 32 if training_sizes else 0
+    mean_32_bytes = int(sum(training_sizes) / len(training_sizes)) * 32 if training_sizes else 0
+
+    largest_files = sorted(files, key=lambda item: item["bytes"], reverse=True)[: args.top_n]
+    annotation_files = [item for item in files if item["basename"] == "annotation.hdf5"]
+    annotation_sizes = [item["bytes"] for item in annotation_files]
+    annotation_size_summary = summarize_sizes(annotation_sizes)
+    annotation_median = int(annotation_size_summary.get("median_bytes", 0))
+    largest_episodes = sorted(episode_records, key=lambda item: item["total_bytes"], reverse=True)[: args.top_n]
+    smallest_complete = sorted(complete, key=lambda item: item["training_bytes_excluding_visualization_rrd"])[: args.top_n]
+
+    selected_32 = []
+    for session, _count in sorted(complete_by_session.items()):
+        candidates = [ep for ep in complete if ep["top_level_session"] == session]
+        candidates.sort(key=lambda ep: ep["training_bytes_excluding_visualization_rrd"])
+        selected_32.append(candidates[0])
+        if len(selected_32) == 32:
+            break
+
+    payload = {
+        "status": "pass",
+        "generated_at_utc": datetime.now(timezone.utc).isoformat(timespec="seconds"),
+        "repo_id": args.repo_id,
+        "repo_sha": getattr(info, "sha", None),
+        "gated": getattr(info, "gated", None),
+        "last_modified": getattr(info, "last_modified", None).isoformat() if getattr(info, "last_modified", None) else None,
+        "card_data": getattr(info, "card_data", None).to_dict() if getattr(info, "card_data", None) and hasattr(getattr(info, "card_data", None), "to_dict") else None,
+        "summary": {
+            "sibling_count": len(siblings),
+            "file_count_excluding_gitattributes": len(files),
+            "total_bytes_from_file_metadata": total_bytes,
+            "total_human_from_file_metadata": human_bytes(total_bytes),
+            "training_bytes_excluding_visualization_rrd": total_bytes - rrd_bytes,
+            "training_human_excluding_visualization_rrd": human_bytes(total_bytes - rrd_bytes),
+            "visualization_rrd_bytes": rrd_bytes,
+            "visualization_rrd_human": human_bytes(rrd_bytes),
+            "top_level_session_count": len(top_level_counter),
+            "episode_like_folder_count": len(episode_records),
+            "annotation_hdf5_count": basename_counter["annotation.hdf5"],
+            "mp4_count": sum(count for name, count in basename_counter.items() if name.endswith(".mp4")),
+            "visualization_rrd_count": basename_counter["visualization.rrd"],
+            "complete_episode_count": len(complete),
+            "degraded_valid_episode_count": len(degraded),
+            "complete_episode_pct": pct(len(complete), len(episode_records)),
+            "degraded_valid_episode_pct": pct(len(degraded), len(episode_records)),
+            "complete_sessions": len(complete_by_session),
+            "degraded_valid_sessions": len(degraded_by_session),
+            "all_complete_episode_training_bytes_excluding_visualization_rrd": all_complete_training_bytes,
+            "all_complete_episode_training_human_excluding_visualization_rrd": human_bytes(all_complete_training_bytes),
+        },
+        "file_type_counts": dict(sorted(ext_counter.items())),
+        "basename_counts": dict(sorted(basename_counter.items())),
+        "video_count_histogram": dict(sorted(video_count_hist.items())),
+        "episode_count_per_session_summary": summarize_numbers(list(episode_count_by_session.values())),
+        "episode_size_summary": summarize_sizes(episode_sizes),
+        "annotation_file_size_summary": annotation_size_summary,
+        "complete_episode_training_size_summary": summarize_sizes(training_sizes),
+        "incomplete_episode_records": incomplete,
+        "pilot_scale_estimates": {
+            "windows_per_episode": 256,
+            "all_complete_episodes_windows_at_256_each": len(complete) * 256,
+            "episode_32_windows_at_256_each": 32 * 256,
+            "episode_100_windows_at_256_each": 100 * 256,
+            "episode_500_windows_at_256_each": 500 * 256,
+            "median_based_32_episode_training_bytes": median_32_bytes,
+            "median_based_32_episode_training_human": human_bytes(median_32_bytes),
+            "mean_based_32_episode_training_bytes": mean_32_bytes,
+            "mean_based_32_episode_training_human": human_bytes(mean_32_bytes),
+        },
+        "selected_32_smallest_one_per_session_estimate": {
+            "episode_count": len(selected_32),
+            "estimated_training_bytes_excluding_visualization_rrd": sum(
+                ep["training_bytes_excluding_visualization_rrd"] for ep in selected_32
+            ),
+            "estimated_training_human": human_bytes(
+                sum(ep["training_bytes_excluding_visualization_rrd"] for ep in selected_32)
+            ),
+            "episodes": selected_32,
+        },
+        "top_level_sessions_by_file_count_top_n": top_level_counter.most_common(args.top_n),
+        "top_level_sessions_by_bytes_top_n": [
+            {"session": session, "bytes": bytes_, "human": human_bytes(bytes_)}
+            for session, bytes_ in by_top_level_bytes.most_common(args.top_n)
+        ],
+        "largest_files_top_n": [
+            {**item, "human": human_bytes(item["bytes"])} for item in largest_files
+        ],
+        "smallest_annotation_files_top_n": [
+            {**item, "human": human_bytes(item["bytes"])}
+            for item in sorted(annotation_files, key=lambda item: item["bytes"])[: args.top_n]
+        ],
+        "median_annotation_files_top_n": [
+            {**item, "human": human_bytes(item["bytes"])}
+            for item in near_size_files(annotation_files, annotation_median, args.top_n)
+        ],
+        "largest_annotation_files_top_n": [
+            {**item, "human": human_bytes(item["bytes"])}
+            for item in sorted(annotation_files, key=lambda item: item["bytes"], reverse=True)[: args.top_n]
+        ],
+        "largest_episode_folders_top_n": [
+            {**item, "total_human": human_bytes(item["total_bytes"]), "training_human": human_bytes(item["training_bytes_excluding_visualization_rrd"])}
+            for item in largest_episodes
+        ],
+        "smallest_complete_episode_training_folders_top_n": [
+            {**item, "total_human": human_bytes(item["total_bytes"]), "training_human": human_bytes(item["training_bytes_excluding_visualization_rrd"])}
+            for item in smallest_complete
+        ],
+        "download_recommendation": {
+            "metadata_only_audit_requires_training_host": False,
+            "recommended_download_host": "Any HF-reachable relay host with enough scratch storage; transfer staged episodes to the training host if that host cannot access Hugging Face.",
+            "training_host_role": "training and local manifest validation after data is staged",
+            "exclude_files": sorted(TRAINING_EXCLUDE),
+            "minimum_pilot": "32 complete episodes from different top-level sessions if storage permits; degraded-valid episodes only for loader smoke tests.",
+        },
+    }
+
+    args.output.parent.mkdir(parents=True, exist_ok=True)
+    args.output.write_text(json.dumps(payload, indent=2) + "\n", encoding="utf-8")
+
+    summary = payload["summary"]
+    complete_sizes = payload["complete_episode_training_size_summary"]
+    annotation_sizes_report = payload["annotation_file_size_summary"]
+    pilot = payload["pilot_scale_estimates"]
+    selected_32_estimate = payload["selected_32_smallest_one_per_session_estimate"]
+    card_data = payload["card_data"] or {}
+    report = [
+        "# Xperience-10M HF Metadata Audit",
+        "",
+        "Metadata-only analysis of the gated Hugging Face dataset. No MP4, HDF5, RRD, or model files were downloaded.",
+        "",
+        "## Access and Source",
+        "",
+        f"- Repo: `{args.repo_id}`",
+        f"- Repo SHA: `{payload['repo_sha']}`",
+        f"- Last modified: `{payload['last_modified']}`",
+        f"- Gated mode: `{payload['gated']}`",
+        f"- Pretty name: `{card_data.get('pretty_name', 'Xperience-10M')}`",
+        f"- License field: `{card_data.get('license', 'unknown')}`",
+        f"- HF size category: `{', '.join(card_data.get('size_categories', [])) or 'unknown'}`",
+        f"- Tags: `{', '.join(card_data.get('tags', []))}`",
+        "",
+        "## Current Hub File Metadata",
+        "",
+        *md_table(
+            ["Measure", "Value"],
+            [
+                ["Files listed by API", f"{summary['file_count_excluding_gitattributes']:,}"],
+                ["Total bytes from file metadata", f"{summary['total_human_from_file_metadata']} ({summary['total_bytes_from_file_metadata']:,} bytes)"],
+                ["Bytes excluding visualization.rrd", f"{summary['training_human_excluding_visualization_rrd']} ({summary['training_bytes_excluding_visualization_rrd']:,} bytes)"],
+                ["visualization.rrd bytes", f"{summary['visualization_rrd_human']} ({summary['visualization_rrd_bytes']:,} bytes)"],
+                ["Top-level session folders", f"{summary['top_level_session_count']:,}"],
+                ["Episode-like folders", f"{summary['episode_like_folder_count']:,}"],
+            ],
+        ),
+        "",
+        "## File Composition",
+        "",
+        *md_table(
+            ["File type", "Count"],
+            [[key, f"{value:,}"] for key, value in payload["file_type_counts"].items()],
+        ),
+        "",
+        "## Episode Completeness",
+        "",
+        *md_table(
+            ["Measure", "Value"],
+            [
+                ["annotation.hdf5 files", f"{summary['annotation_hdf5_count']:,}"],
+                ["MP4 files", f"{summary['mp4_count']:,}"],
+                ["visualization.rrd files", f"{summary['visualization_rrd_count']:,}"],
+                ["Complete episodes: annotation + all six MP4 views", f"{summary['complete_episode_count']:,} ({summary['complete_episode_pct']}%)"],
+                ["Degraded-valid episodes: annotation + fisheye_cam0", f"{summary['degraded_valid_episode_count']:,} ({summary['degraded_valid_episode_pct']}%)"],
+                ["Sessions with complete episodes", f"{summary['complete_sessions']:,}"],
+                ["Video-count histogram per episode", json.dumps(payload["video_count_histogram"], sort_keys=True)],
+            ],
+        ),
+        "",
+        "## Episode Size Distribution",
+        "",
+        *md_table(
+            ["Statistic", "Training bytes per complete episode, excluding visualization.rrd"],
+            [
+                ["Min", complete_sizes.get("min_human")],
+                ["P25", human_bytes(complete_sizes.get("p25_bytes", 0))],
+                ["Median", complete_sizes.get("median_human")],
+                ["P75", human_bytes(complete_sizes.get("p75_bytes", 0))],
+                ["Mean", complete_sizes.get("mean_human")],
+                ["Max", complete_sizes.get("max_human")],
+            ],
+        ),
+        "",
+        "## Annotation File Size Distribution",
+        "",
+        *md_table(
+            ["Statistic", "annotation.hdf5 size"],
+            [
+                ["Min", annotation_sizes_report.get("min_human")],
+                ["P25", human_bytes(annotation_sizes_report.get("p25_bytes", 0))],
+                ["Median", annotation_sizes_report.get("median_human")],
+                ["P75", human_bytes(annotation_sizes_report.get("p75_bytes", 0))],
+                ["Mean", annotation_sizes_report.get("mean_human")],
+                ["Max", annotation_sizes_report.get("max_human")],
+            ],
+        ),
+        "",
+        "## Pilot Scale Estimates",
+        "",
+        *md_table(
+            ["Pilot", "Episodes", "Max windows at 256/episode", "Storage estimate"],
+            [
+                ["32-episode smallest one-per-session", selected_32_estimate["episode_count"], pilot["episode_32_windows_at_256_each"], selected_32_estimate["estimated_training_human"]],
+                ["32-episode median-sized estimate", 32, pilot["episode_32_windows_at_256_each"], pilot["median_based_32_episode_training_human"]],
+                ["32-episode mean-sized estimate", 32, pilot["episode_32_windows_at_256_each"], pilot["mean_based_32_episode_training_human"]],
+                ["100-episode pilot", 100, pilot["episode_100_windows_at_256_each"], f"roughly {human_bytes(complete_sizes.get('median_bytes', 0) * 100)} at median episode size"],
+                ["500-episode pilot", 500, pilot["episode_500_windows_at_256_each"], f"roughly {human_bytes(complete_sizes.get('median_bytes', 0) * 500)} at median episode size"],
+                ["All complete visible HF episodes", summary["complete_episode_count"], pilot["all_complete_episodes_windows_at_256_each"], summary["all_complete_episode_training_human_excluding_visualization_rrd"]],
+            ],
+        ),
+        "",
+        "## Incomplete Episode Records",
+        "",
+        json.dumps(incomplete, indent=2) if incomplete else "None found.",
+        "",
+        "## Download and Compute Recommendation",
+        "",
+        "- This metadata audit can run on any machine with Hugging Face access.",
+        "- If the training host cannot reach Hugging Face, download on an HF-reachable relay host, then transfer staged episode folders to the training host.",
+        "- For training downloads, include `annotation.hdf5` plus the six MP4 streams; exclude `visualization.rrd` unless Rerun visualization is specifically needed.",
+        "- For the first real training pilot, prefer 32 complete episodes from different top-level sessions and avoid selecting only the tiny outlier episodes.",
+        "- The training host is used after staged data exists: manifest validation, preprocessing, LoRA training, and held-out evaluation.",
+    ]
+    args.report_output.write_text("\n".join(report) + "\n", encoding="utf-8")
+    print(f"PASS: wrote {args.output}")
+    print(f"PASS: wrote {args.report_output}")
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

scripts/omni/audit_staged_xperience10m_content.py

@@ -0,0 +1,253 @@
+#!/usr/bin/env python3
+"""Audit semantic content labels after Xperience-10M annotations are staged."""
+
+from __future__ import annotations
+
+import argparse
+import csv
+import json
+from collections import Counter, defaultdict
+from datetime import datetime, timezone
+from pathlib import Path
+from typing import Any
+
+import h5py
+
+
+CATEGORY_RULES = {
+    "food_and_drink": ["cook", "coffee", "drink", "food", "kitchen", "meal", "pour", "cup", "bottle"],
+    "dressing_and_hygiene": ["sock", "shoe", "dress", "wear", "bathroom", "toilet", "wash", "brush"],
+    "packing_and_organizing": ["pack", "organize", "bin", "box", "storage", "arrange", "sort"],
+    "shopping_and_retail": ["retail", "shop", "shelf", "aisle", "product", "store"],
+    "cleaning_and_housework": ["clean", "wipe", "sweep", "wash", "laundry", "trash"],
+    "navigation_and_locomotion": ["walk", "move through", "navigate", "stair", "hallway", "corridor"],
+    "tool_or_device_use": ["tool", "device", "phone", "computer", "laptop", "button", "switch"],
+    "object_manipulation": ["pick", "place", "grasp", "hold", "open", "close", "move", "lift"],
+}
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("--data-root", type=Path, required=True)
+    parser.add_argument("--selection-json", type=Path, default=None)
+    parser.add_argument("--output-json", type=Path, default=Path("results/omni_finetune/staged_content_audit.json"))
+    parser.add_argument("--output-csv", type=Path, default=Path("results/omni_finetune/staged_content_audit.csv"))
+    parser.add_argument("--report-output", type=Path, default=Path("results/omni_finetune/STAGED_CONTENT_AUDIT.md"))
+    return parser.parse_args()
+
+
+def load_selection(path: Path | None) -> dict[str, dict[str, Any]]:
+    if path is None or not path.exists():
+        return {}
+    payload = json.loads(path.read_text(encoding="utf-8"))
+    return {ep["episode_path"]: ep for ep in payload.get("selected_episodes", [])}
+
+
+def parse_caption(annotation: Path) -> dict[str, Any]:
+    with h5py.File(annotation, "r") as h5:
+        if "caption" not in h5:
+            return {"parse_status": "missing"}
+        raw = h5["caption"][()]
+    text = raw.decode("utf-8", errors="replace") if isinstance(raw, bytes) else str(raw)
+    try:
+        data = json.loads(text)
+    except Exception as exc:
+        return {"parse_status": "failed", "error": str(exc), "json_bytes": len(text.encode("utf-8"))}
+
+    config = data.get("config", {}) if isinstance(data, dict) else {}
+    segments = data.get("segments", []) if isinstance(data, dict) else []
+    if not isinstance(segments, list):
+        segments = []
+
+    subtasks: list[str] = []
+    actions: list[str] = []
+    objects: list[str] = []
+    interactions: list[str] = []
+    for segment in segments:
+        if not isinstance(segment, dict):
+            continue
+        if segment.get("Sub Task"):
+            subtasks.append(str(segment["Sub Task"]))
+        current_actions = segment.get("Current Action", [])
+        if isinstance(current_actions, list):
+            for action in current_actions:
+                if isinstance(action, dict):
+                    if action.get("label"):
+                        actions.append(str(action["label"]))
+                    if action.get("description"):
+                        interactions.append(str(action["description"]))
+        object_map = segment.get("objects", {})
+        if isinstance(object_map, dict):
+            for names in object_map.values():
+                if isinstance(names, list):
+                    objects.extend(str(name) for name in names)
+        interaction_map = segment.get("interaction", {})
+        if isinstance(interaction_map, dict):
+            interactions.extend(str(value) for value in interaction_map.values())
+
+    main_task = str(config.get("Main Task", ""))
+    global_summary = str(data.get("global_summary", "")) if isinstance(data, dict) else ""
+    return {
+        "parse_status": "ok",
+        "json_bytes": len(text.encode("utf-8")),
+        "main_task": main_task,
+        "global_summary": global_summary,
+        "segment_count": len(segments),
+        "subtasks": sorted(set(subtasks)),
+        "actions": sorted(set(actions)),
+        "objects": sorted(set(objects)),
+        "interaction_preview": " ".join(interactions)[:500],
+    }
+
+
+def derive_category(record: dict[str, Any]) -> str:
+    text = " ".join(
+        [
+            record.get("main_task", ""),
+            record.get("global_summary", ""),
+            " ".join(record.get("subtasks", [])),
+            " ".join(record.get("actions", [])),
+            " ".join(record.get("objects", [])),
+            record.get("interaction_preview", ""),
+        ]
+    ).lower()
+    scores = {
+        category: sum(1 for keyword in keywords if keyword in text)
+        for category, keywords in CATEGORY_RULES.items()
+    }
+    category, score = max(scores.items(), key=lambda item: item[1])
+    return category if score > 0 else "uncategorized"
+
+
+def infer_episode_key(annotation: Path, data_root: Path) -> str:
+    parent = annotation.parent
+    try:
+        return parent.relative_to(data_root).as_posix()
+    except ValueError:
+        return parent.as_posix()
+
+
+def md_table(headers: list[str], rows: list[list[Any]]) -> list[str]:
+    lines = [
+        "| " + " | ".join(headers) + " |",
+        "| " + " | ".join("---" for _ in headers) + " |",
+    ]
+    lines.extend("| " + " | ".join(str(cell) for cell in row) + " |" for row in rows)
+    return lines
+
+
+def write_csv(path: Path, rows: list[dict[str, Any]]) -> None:
+    path.parent.mkdir(parents=True, exist_ok=True)
+    fields = [
+        "episode_path",
+        "split",
+        "size_band",
+        "category",
+        "main_task",
+        "segment_count",
+        "actions",
+        "objects",
+        "annotation_path",
+    ]
+    with path.open("w", newline="", encoding="utf-8") as handle:
+        writer = csv.DictWriter(handle, fieldnames=fields)
+        writer.writeheader()
+        for row in rows:
+            writer.writerow(
+                {
+                    "episode_path": row["episode_path"],
+                    "split": row.get("split", ""),
+                    "size_band": row.get("size_band", ""),
+                    "category": row["category"],
+                    "main_task": row.get("main_task", ""),
+                    "segment_count": row.get("segment_count", 0),
+                    "actions": "; ".join(row.get("actions", [])),
+                    "objects": "; ".join(row.get("objects", [])),
+                    "annotation_path": row["annotation_path"],
+                }
+            )
+
+
+def main() -> int:
+    args = parse_args()
+    data_root = args.data_root.expanduser().resolve()
+    selection = load_selection(args.selection_json)
+    annotations = sorted(data_root.rglob("annotation.hdf5"))
+    rows: list[dict[str, Any]] = []
+    for annotation in annotations:
+        episode_path = infer_episode_key(annotation, data_root)
+        parsed = parse_caption(annotation)
+        selected_meta = selection.get(episode_path, {})
+        record = {
+            "episode_path": episode_path,
+            "annotation_path": str(annotation),
+            "split": selected_meta.get("split", ""),
+            "size_band": selected_meta.get("size_band", ""),
+            **parsed,
+        }
+        record["category"] = derive_category(record) if parsed.get("parse_status") == "ok" else "unparsed"
+        rows.append(record)
+
+    category_counts = Counter(row["category"] for row in rows)
+    split_category_counts: dict[str, Counter] = defaultdict(Counter)
+    for row in rows:
+        split_category_counts[row.get("split", "")][row["category"]] += 1
+
+    payload = {
+        "status": "pass",
+        "generated_at_utc": datetime.now(timezone.utc).isoformat(timespec="seconds"),
+        "data_root": str(data_root),
+        "selection_json": str(args.selection_json) if args.selection_json else None,
+        "episode_count": len(rows),
+        "category_counts": dict(category_counts.most_common()),
+        "split_category_counts": {split or "unknown": dict(counts.most_common()) for split, counts in split_category_counts.items()},
+        "rows": rows,
+        "note": "Categories are keyword-derived from caption text and should be reviewed before final training claims.",
+    }
+    args.output_json.parent.mkdir(parents=True, exist_ok=True)
+    args.output_json.write_text(json.dumps(payload, indent=2) + "\n", encoding="utf-8")
+    write_csv(args.output_csv, rows)
+
+    report = [
+        "# Xperience-10M Staged Content Audit",
+        "",
+        "This report parses staged `annotation.hdf5` files and derives coarse content categories from caption text.",
+        "",
+        f"- Data root: `{data_root}`",
+        f"- Episodes parsed: {len(rows)}",
+        "",
+        "## Category Counts",
+        "",
+        *md_table(["Category", "Episodes"], [[cat, count] for cat, count in category_counts.most_common()]),
+        "",
+        "## Split x Category",
+        "",
+    ]
+    categories = sorted(category_counts)
+    report.extend(
+        md_table(
+            ["Split", *categories],
+            [
+                [split or "unknown", *[counts.get(category, 0) for category in categories]]
+                for split, counts in sorted(split_category_counts.items())
+            ],
+        )
+    )
+    report.extend(
+        [
+            "",
+            "## Next Action",
+            "",
+            "If one category dominates train, val, or test, swap episodes from the staged pool before starting model fine-tuning.",
+        ]
+    )
+    args.report_output.write_text("\n".join(report) + "\n", encoding="utf-8")
+    print(json.dumps({"episode_count": len(rows), "category_counts": payload["category_counts"]}, indent=2))
+    print(f"PASS: wrote {args.output_json}")
+    print(f"PASS: wrote {args.output_csv}")
+    print(f"PASS: wrote {args.report_output}")
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

scripts/omni/probe_xperience10m_annotation_records.py

@@ -0,0 +1,354 @@
+#!/usr/bin/env python3
+"""Download and inspect minimal Xperience-10M annotation.hdf5 files.
+
+This probe intentionally downloads only annotation files, not videos or RRD
+viewer files. Raw annotations are cached outside the repo by default.
+"""
+
+from __future__ import annotations
+
+import argparse
+import getpass
+import json
+import os
+from collections import Counter, defaultdict
+from datetime import datetime, timezone
+from pathlib import Path
+from typing import Any
+
+import h5py
+import numpy as np
+from huggingface_hub import hf_hub_download
+
+
+DEFAULT_ANNOTATIONS = [
+    "9cecac72-8874-4b97-9541-18d4858f8e43/ep10/annotation.hdf5",
+]
+TEXT_RELATED_TERMS = ("caption", "action", "interaction", "object", "subtask", "task")
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("--repo-id", default="ropedia-ai/xperience-10m")
+    parser.add_argument("--filenames", nargs="+", default=DEFAULT_ANNOTATIONS)
+    parser.add_argument(
+        "--cache-dir",
+        type=Path,
+        default=Path(os.environ.get("XPERIENCE10M_ANNOTATION_PROBE_CACHE", "xperience10m_annotation_probe_cache")),
+    )
+    parser.add_argument("--output", type=Path, default=Path("results/omni_finetune/annotation_record_probe.json"))
+    parser.add_argument("--report-output", type=Path, default=Path("results/omni_finetune/ANNOTATION_RECORD_PROBE.md"))
+    parser.add_argument("--token", default=os.environ.get("HF_TOKEN", "").strip())
+    parser.add_argument("--sample-values", type=int, default=3)
+    parser.add_argument("--local-files-only", action="store_true", help="Use the HF cache and do not contact the Hub.")
+    return parser.parse_args()
+
+
+def human_bytes(num: float | int) -> str:
+    value = float(num)
+    for unit in ["B", "KiB", "MiB", "GiB", "TiB"]:
+        if abs(value) < 1024.0 or unit == "TiB":
+            return f"{value:.2f} {unit}"
+        value /= 1024.0
+    return f"{value:.2f} TiB"
+
+
+def json_safe(value: Any) -> Any:
+    if isinstance(value, np.generic):
+        return value.item()
+    if isinstance(value, bytes):
+        return value.decode("utf-8", errors="replace")
+    if isinstance(value, np.ndarray):
+        return value.tolist()
+    return value
+
+
+def sample_dataset(ds: h5py.Dataset, limit: int) -> list[Any]:
+    if ds.shape == ():
+        raw = ds[()]
+        return [json_safe(raw)]
+    if not ds.shape or ds.shape[0] == 0:
+        return []
+    count = min(limit, int(ds.shape[0]))
+    samples: list[Any] = []
+    for idx in range(count):
+        try:
+            if ds.dtype.kind in {"S", "O", "U"}:
+                raw = ds.asstr()[idx]
+            else:
+                raw = ds[idx]
+            if isinstance(raw, np.ndarray) and raw.size > 12:
+                raw = raw.reshape(-1)[:12]
+            samples.append(json_safe(raw))
+        except Exception as exc:  # pragma: no cover - defensive HDF5 read path
+            samples.append(f"<sample failed: {exc}>")
+            break
+    return samples
+
+
+def inspect_annotation(path: Path, sample_limit: int) -> dict[str, Any]:
+    datasets: list[dict[str, Any]] = []
+    related_datasets: list[dict[str, Any]] = []
+    top_group_stats: dict[str, dict[str, Any]] = defaultdict(
+        lambda: {"dataset_count": 0, "max_first_dim": 0, "first_dim_values": Counter()}
+    )
+
+    caption_json_summary = None
+    with h5py.File(path, "r") as h5:
+        top_level_keys = sorted(h5.keys())
+
+        def visitor(name: str, obj: Any) -> None:
+            if not isinstance(obj, h5py.Dataset):
+                return
+            shape = tuple(int(dim) for dim in obj.shape)
+            first_dim = int(shape[0]) if shape else None
+            top = name.split("/", 1)[0]
+            stats = top_group_stats[top]
+            stats["dataset_count"] += 1
+            if first_dim is not None:
+                stats["max_first_dim"] = max(int(stats["max_first_dim"]), first_dim)
+                stats["first_dim_values"][str(first_dim)] += 1
+
+            record = {
+                "path": name,
+                "shape": list(shape),
+                "dtype": str(obj.dtype),
+                "first_dim": first_dim,
+                "storage_bytes": int(obj.id.get_storage_size()),
+                "storage_human": human_bytes(obj.id.get_storage_size()),
+            }
+            datasets.append(record)
+
+            lowered = name.lower()
+            if any(term in lowered for term in TEXT_RELATED_TERMS):
+                related = dict(record)
+                related["sample_values"] = sample_dataset(obj, sample_limit)
+                related_datasets.append(related)
+
+        h5.visititems(visitor)
+        if "caption" in h5 and isinstance(h5["caption"], h5py.Dataset):
+            caption_json_summary = summarize_caption_json(h5["caption"])
+
+    top_stats_out = {
+        key: {
+            "dataset_count": int(value["dataset_count"]),
+            "max_first_dim": int(value["max_first_dim"]),
+            "first_dim_values": dict(value["first_dim_values"].most_common(10)),
+        }
+        for key, value in sorted(top_group_stats.items())
+    }
+    dataset_first_dims = Counter(
+        str(item["first_dim"]) for item in datasets if item["first_dim"] is not None
+    )
+    max_first_dim_dataset = max(datasets, key=lambda item: item["first_dim"] or -1) if datasets else None
+    return {
+        "cache_note": "annotation file cached outside the published repo",
+        "local_bytes": path.stat().st_size,
+        "local_human": human_bytes(path.stat().st_size),
+        "top_level_keys": top_level_keys,
+        "dataset_count": len(datasets),
+        "dataset_first_dim_histogram_top20": dict(dataset_first_dims.most_common(20)),
+        "top_group_stats": top_stats_out,
+        "max_first_dim_dataset": max_first_dim_dataset,
+        "text_action_interaction_related_datasets": related_datasets,
+        "caption_json_summary": caption_json_summary,
+    }
+
+
+def summarize_caption_json(ds: h5py.Dataset) -> dict[str, Any] | None:
+    try:
+        raw = ds[()]
+        text = raw.decode("utf-8", errors="replace") if isinstance(raw, bytes) else str(raw)
+        data = json.loads(text)
+    except Exception as exc:  # pragma: no cover - defensive parse path
+        return {"parse_status": "failed", "error": str(exc)}
+
+    segments = data.get("segments", [])
+    if not isinstance(segments, list):
+        segments = []
+
+    sub_tasks = []
+    action_labels = []
+    object_names = []
+    object_frame_count = 0
+    interaction_frame_count = 0
+    sampled_frame_count = 0
+
+    for segment in segments:
+        if not isinstance(segment, dict):
+            continue
+        if segment.get("Sub Task"):
+            sub_tasks.append(str(segment["Sub Task"]))
+        actions = segment.get("Current Action", [])
+        if isinstance(actions, list):
+            for action in actions:
+                if isinstance(action, dict) and action.get("label"):
+                    action_labels.append(str(action["label"]))
+        objects = segment.get("objects", {})
+        if isinstance(objects, dict):
+            object_frame_count += len(objects)
+            for names in objects.values():
+                if isinstance(names, list):
+                    object_names.extend(str(name) for name in names)
+        interaction = segment.get("interaction", {})
+        if isinstance(interaction, dict):
+            interaction_frame_count += len(interaction)
+        sampled_frames = segment.get("sampled_frames", {})
+        if isinstance(sampled_frames, dict):
+            sampled_frame_count += len(sampled_frames)
+
+    config = data.get("config", {}) if isinstance(data, dict) else {}
+    return {
+        "parse_status": "ok",
+        "json_bytes": len(text.encode("utf-8")),
+        "top_keys": list(data.keys()) if isinstance(data, dict) else [],
+        "config": config,
+        "segment_count": len(segments),
+        "current_action_count": len(action_labels),
+        "unique_sub_task_count": len(set(sub_tasks)),
+        "unique_action_label_count": len(set(action_labels)),
+        "object_frame_count": object_frame_count,
+        "interaction_frame_count": interaction_frame_count,
+        "sampled_frame_count": sampled_frame_count,
+        "unique_object_count": len(set(object_names)),
+        "sub_tasks": sorted(set(sub_tasks))[:20],
+        "action_labels": sorted(set(action_labels))[:20],
+        "objects": sorted(set(object_names))[:30],
+        "global_summary_preview": str(data.get("global_summary", ""))[:240] if isinstance(data, dict) else "",
+    }
+
+
+def md_table(headers: list[str], rows: list[list[Any]]) -> list[str]:
+    lines = [
+        "| " + " | ".join(headers) + " |",
+        "| " + " | ".join("---" for _ in headers) + " |",
+    ]
+    lines.extend("| " + " | ".join(str(cell) for cell in row) + " |" for row in rows)
+    return lines
+
+
+def main() -> int:
+    args = parse_args()
+    token = args.token
+    if not args.local_files_only and not token:
+        token = getpass.getpass("HF token: ").strip()
+    if not args.local_files_only and not token:
+        raise SystemExit("HF token is required for gated dataset annotation probing.")
+
+    args.cache_dir.mkdir(parents=True, exist_ok=True)
+    probes = []
+    for filename in args.filenames:
+        local_path = hf_hub_download(
+            repo_id=args.repo_id,
+            repo_type="dataset",
+            filename=filename,
+            cache_dir=args.cache_dir,
+            token=token or None,
+            local_files_only=args.local_files_only,
+        )
+        local = Path(local_path)
+        probes.append(
+            {
+                "repo_filename": filename,
+                "inspection": inspect_annotation(local, args.sample_values),
+            }
+        )
+
+    payload = {
+        "status": "pass",
+        "generated_at_utc": datetime.now(timezone.utc).isoformat(timespec="seconds"),
+        "repo_id": args.repo_id,
+        "download_policy": "annotation.hdf5 only; no videos or visualization.rrd downloaded",
+        "cache_note": "raw annotation files were cached outside the published repo",
+        "probes": probes,
+    }
+
+    args.output.parent.mkdir(parents=True, exist_ok=True)
+    args.output.write_text(json.dumps(payload, indent=2) + "\n", encoding="utf-8")
+
+    report = [
+        "# Xperience-10M Annotation Record Probe",
+        "",
+        "Minimal-cost probe. Downloaded only `annotation.hdf5`; no MP4 or `visualization.rrd` files were downloaded.",
+        "",
+        f"- Repo: `{args.repo_id}`",
+        f"- Probe count: {len(probes)}",
+        "- Raw annotation cache: outside the published repo",
+        f"- Local files only: `{args.local_files_only}`",
+        "",
+    ]
+    for probe in probes:
+        inspection = probe["inspection"]
+        max_ds = inspection.get("max_first_dim_dataset") or {}
+        report.extend(
+            [
+                f"## {probe['repo_filename']}",
+                "",
+                f"- Downloaded annotation size: {inspection['local_human']} ({inspection['local_bytes']:,} bytes)",
+                f"- HDF5 top-level keys: `{', '.join(inspection['top_level_keys'])}`",
+                f"- HDF5 dataset count: {inspection['dataset_count']:,}",
+                f"- Largest first-dimension dataset: `{max_ds.get('path')}` with first dimension `{max_ds.get('first_dim')}`",
+                "",
+                "### Caption JSON Summary",
+                "",
+            ]
+        )
+        caption_summary = inspection.get("caption_json_summary") or {}
+        report.extend(
+            md_table(
+                ["Measure", "Value"],
+                [
+                    ["Parse status", caption_summary.get("parse_status")],
+                    ["JSON bytes", f"{caption_summary.get('json_bytes', 0):,}"],
+                    ["Segment count", caption_summary.get("segment_count")],
+                    ["Current-action count", caption_summary.get("current_action_count")],
+                    ["Object-frame count", caption_summary.get("object_frame_count")],
+                    ["Interaction-frame count", caption_summary.get("interaction_frame_count")],
+                    ["Sampled-frame count", caption_summary.get("sampled_frame_count")],
+                    ["Unique subtasks", caption_summary.get("unique_sub_task_count")],
+                    ["Unique action labels", caption_summary.get("unique_action_label_count")],
+                    ["Unique objects", caption_summary.get("unique_object_count")],
+                    ["Action labels", json.dumps(caption_summary.get("action_labels", []), ensure_ascii=False)],
+                    ["Objects", json.dumps(caption_summary.get("objects", []), ensure_ascii=False)],
+                ],
+            )
+        )
+        report.extend(
+            [
+                "",
+                "### Top Groups",
+                "",
+                *md_table(
+                    ["Group", "Dataset count", "Max first dimension", "First-dim histogram top values"],
+                    [
+                        [
+                            group,
+                            stats["dataset_count"],
+                            stats["max_first_dim"],
+                            json.dumps(stats["first_dim_values"], sort_keys=False),
+                        ]
+                        for group, stats in inspection["top_group_stats"].items()
+                    ],
+                ),
+                "",
+                "### Caption / Action / Interaction Related Datasets",
+                "",
+            ]
+        )
+        related_rows = []
+        for item in inspection["text_action_interaction_related_datasets"]:
+            sample = json.dumps(item.get("sample_values", []), ensure_ascii=False)
+            if len(sample) > 160:
+                sample = sample[:157] + "..."
+            related_rows.append([item["path"], item["shape"], item["dtype"], item["first_dim"], sample])
+        report.extend(md_table(["Dataset", "Shape", "Dtype", "First dim", "Sample values"], related_rows or [["None", "", "", "", ""]]))
+        report.append("")
+
+    args.report_output.write_text("\n".join(report) + "\n", encoding="utf-8")
+    print(f"PASS: wrote {args.output}")
+    print(f"PASS: wrote {args.report_output}")
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

scripts/omni/relay_xperience10m_selection.py

@@ -0,0 +1,327 @@
+#!/usr/bin/env python3
+"""Download selected Xperience-10M episodes in relay-sized batches.
+
+Intended host: an HF-reachable relay machine with limited disk. The script
+downloads one batch, optionally rsyncs it to a training host, writes progress
+records, and can delete the local batch after transfer.
+"""
+
+from __future__ import annotations
+
+import argparse
+import getpass
+import json
+import os
+import shlex
+import shutil
+import subprocess
+import time
+from dataclasses import dataclass
+from datetime import datetime, timezone
+from pathlib import Path
+from typing import Any
+
+from huggingface_hub import hf_hub_download
+
+
+REQUIRED_FILES = [
+    "annotation.hdf5",
+    "fisheye_cam0.mp4",
+    "fisheye_cam1.mp4",
+    "fisheye_cam2.mp4",
+    "fisheye_cam3.mp4",
+    "stereo_left.mp4",
+    "stereo_right.mp4",
+]
+
+
+@dataclass
+class Batch:
+    index: int
+    episodes: list[dict[str, Any]]
+
+    @property
+    def bytes(self) -> int:
+        return sum(int(ep["training_bytes_excluding_visualization_rrd"]) for ep in self.episodes)
+
+    @property
+    def file_count(self) -> int:
+        return sum(len(ep["download_files"]) for ep in self.episodes)
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("--repo-id", default="ropedia-ai/xperience-10m")
+    parser.add_argument("--selection-json", type=Path, required=True)
+    parser.add_argument("--relay-root", type=Path, required=True)
+    parser.add_argument("--batch-max-gib", type=float, default=24.0)
+    parser.add_argument("--batch-max-episodes", type=int, default=8)
+    parser.add_argument("--start-batch", type=int, default=0)
+    parser.add_argument("--max-batches", type=int, default=0, help="0 means all remaining batches.")
+    parser.add_argument("--workers", type=int, default=1, help="Reserved for future use; downloads are sequential for disk safety.")
+    parser.add_argument("--dry-run", action="store_true")
+    parser.add_argument("--token", default=os.environ.get("HF_TOKEN", "").strip())
+    parser.add_argument("--progress-jsonl", type=Path, default=Path("relay_progress.jsonl"))
+    parser.add_argument("--summary-json", type=Path, default=Path("relay_summary.json"))
+    parser.add_argument("--transfer-host", default="", help="Remote destination, e.g. user@training-host")
+    parser.add_argument("--transfer-root", default="", help="Remote directory that receives session/episode folders.")
+    parser.add_argument("--ssh-key", type=Path, default=Path.home() / ".ssh" / "xperience10m_relay_ed25519")
+    parser.add_argument("--ssh-extra", default="-o BatchMode=yes -o StrictHostKeyChecking=accept-new")
+    parser.add_argument("--delete-after-transfer", action="store_true")
+    parser.add_argument("--validate-only", action="store_true", help="Do not download; validate selected files already under relay-root.")
+    return parser.parse_args()
+
+
+def utc_now() -> str:
+    return datetime.now(timezone.utc).isoformat(timespec="seconds")
+
+
+def human_bytes(num: float | int) -> str:
+    value = float(num)
+    for unit in ["B", "KiB", "MiB", "GiB", "TiB"]:
+        if abs(value) < 1024.0 or unit == "TiB":
+            return f"{value:.2f} {unit}"
+        value /= 1024.0
+    return f"{value:.2f} TiB"
+
+
+def append_jsonl(path: Path, record: dict[str, Any]) -> None:
+    path.parent.mkdir(parents=True, exist_ok=True)
+    with path.open("a", encoding="utf-8") as handle:
+        handle.write(json.dumps(record, sort_keys=True) + "\n")
+
+
+def load_selection(path: Path) -> list[dict[str, Any]]:
+    payload = json.loads(path.read_text(encoding="utf-8"))
+    episodes = payload.get("selected_episodes")
+    if not isinstance(episodes, list) or not episodes:
+        raise ValueError(f"No selected_episodes found in {path}")
+    for ep in episodes:
+        missing = [key for key in ("episode_path", "download_files", "training_bytes_excluding_visualization_rrd") if key not in ep]
+        if missing:
+            raise ValueError(f"Selection episode missing {missing}: {ep}")
+    return episodes
+
+
+def make_batches(episodes: list[dict[str, Any]], max_bytes: int, max_episodes: int) -> list[Batch]:
+    batches: list[Batch] = []
+    current: list[dict[str, Any]] = []
+    current_bytes = 0
+    for ep in episodes:
+        ep_bytes = int(ep["training_bytes_excluding_visualization_rrd"])
+        would_exceed_bytes = current and current_bytes + ep_bytes > max_bytes
+        would_exceed_count = current and len(current) >= max_episodes
+        if would_exceed_bytes or would_exceed_count:
+            batches.append(Batch(index=len(batches), episodes=current))
+            current = []
+            current_bytes = 0
+        current.append(ep)
+        current_bytes += ep_bytes
+    if current:
+        batches.append(Batch(index=len(batches), episodes=current))
+    return batches
+
+
+def local_file(root: Path, filename: str) -> Path:
+    return root / filename
+
+
+def validate_batch(root: Path, batch: Batch) -> dict[str, Any]:
+    missing = []
+    size_mismatches = []
+    total_bytes = 0
+    for ep in batch.episodes:
+        expected_by_name = {
+            "annotation.hdf5": int(ep["annotation_bytes"]),
+        }
+        for filename in ep["download_files"]:
+            path = local_file(root, filename)
+            if not path.exists():
+                missing.append(filename)
+                continue
+            actual = path.stat().st_size
+            total_bytes += actual
+            expected = expected_by_name.get(Path(filename).name)
+            if expected is not None and actual != expected:
+                size_mismatches.append({"path": filename, "expected": expected, "actual": actual})
+    return {
+        "ok": not missing and not size_mismatches,
+        "missing": missing,
+        "size_mismatches": size_mismatches,
+        "local_bytes": total_bytes,
+        "local_human": human_bytes(total_bytes),
+    }
+
+
+def download_batch(repo_id: str, token: str, root: Path, batch: Batch, progress_path: Path) -> None:
+    for ep in batch.episodes:
+        for filename in ep["download_files"]:
+            start = time.time()
+            append_jsonl(
+                progress_path,
+                {
+                    "time": utc_now(),
+                    "event": "download_start",
+                    "batch": batch.index,
+                    "episode_path": ep["episode_path"],
+                    "path": filename,
+                },
+            )
+            hf_hub_download(
+                repo_id=repo_id,
+                repo_type="dataset",
+                filename=filename,
+                local_dir=str(root),
+                token=token,
+            )
+            local = local_file(root, filename)
+            append_jsonl(
+                progress_path,
+                {
+                    "time": utc_now(),
+                    "event": "download_done",
+                    "batch": batch.index,
+                    "episode_path": ep["episode_path"],
+                    "path": filename,
+                    "bytes": local.stat().st_size if local.exists() else 0,
+                    "seconds": round(time.time() - start, 3),
+                },
+            )
+
+
+def run_command(cmd: list[str], progress_path: Path, event_prefix: str, batch_index: int, dry_run: bool) -> None:
+    append_jsonl(progress_path, {"time": utc_now(), "event": f"{event_prefix}_start", "batch": batch_index, "cmd": cmd})
+    if dry_run:
+        append_jsonl(progress_path, {"time": utc_now(), "event": f"{event_prefix}_dry_run", "batch": batch_index})
+        return
+    subprocess.run(cmd, check=True)
+    append_jsonl(progress_path, {"time": utc_now(), "event": f"{event_prefix}_done", "batch": batch_index})
+
+
+def transfer_batch(args: argparse.Namespace, batch_root: Path, batch: Batch) -> None:
+    if not args.transfer_host or not args.transfer_root:
+        return
+    ssh_cmd = f"ssh -i {shlex.quote(str(args.ssh_key))} {args.ssh_extra}"
+    mkdir_cmd = [
+        "ssh",
+        "-i",
+        str(args.ssh_key),
+        *shlex.split(args.ssh_extra),
+        args.transfer_host,
+        f"mkdir -p {shlex.quote(args.transfer_root)}",
+    ]
+    run_command(mkdir_cmd, args.progress_jsonl, "remote_mkdir", batch.index, args.dry_run)
+    rsync_cmd = [
+        "rsync",
+        "-avP",
+        "--partial",
+        "--append-verify",
+        "--exclude",
+        "visualization.rrd",
+        "-e",
+        ssh_cmd,
+        f"{batch_root}/",
+        f"{args.transfer_host}:{args.transfer_root}/",
+    ]
+    run_command(rsync_cmd, args.progress_jsonl, "rsync", batch.index, args.dry_run)
+
+
+def write_summary(path: Path, payload: dict[str, Any]) -> None:
+    path.parent.mkdir(parents=True, exist_ok=True)
+    path.write_text(json.dumps(payload, indent=2) + "\n", encoding="utf-8")
+
+
+def main() -> int:
+    args = parse_args()
+    if args.workers != 1:
+        print("NOTE: --workers is reserved; using sequential downloads for relay disk safety.")
+
+    token = args.token
+    if not args.dry_run and not args.validate_only and not token:
+        token = getpass.getpass("HF token: ").strip()
+    if not args.dry_run and not args.validate_only and not token:
+        raise SystemExit("HF token is required unless --dry-run or --validate-only is set.")
+
+    args.relay_root = args.relay_root.expanduser().resolve()
+    args.progress_jsonl = (args.relay_root / args.progress_jsonl).resolve() if not args.progress_jsonl.is_absolute() else args.progress_jsonl
+    args.summary_json = (args.relay_root / args.summary_json).resolve() if not args.summary_json.is_absolute() else args.summary_json
+    args.relay_root.mkdir(parents=True, exist_ok=True)
+
+    episodes = load_selection(args.selection_json)
+    batches = make_batches(episodes, int(args.batch_max_gib * 1024**3), args.batch_max_episodes)
+    selected_batches = batches[args.start_batch :]
+    if args.max_batches > 0:
+        selected_batches = selected_batches[: args.max_batches]
+
+    summary = {
+        "status": "running" if selected_batches else "nothing_to_do",
+        "generated_at_utc": utc_now(),
+        "repo_id": args.repo_id,
+        "selection_json": str(args.selection_json),
+        "relay_root": str(args.relay_root),
+        "batch_max_gib": args.batch_max_gib,
+        "batch_max_episodes": args.batch_max_episodes,
+        "total_batches": len(batches),
+        "scheduled_batches": [batch.index for batch in selected_batches],
+        "scheduled_episode_count": sum(len(batch.episodes) for batch in selected_batches),
+        "scheduled_bytes": sum(batch.bytes for batch in selected_batches),
+        "scheduled_human": human_bytes(sum(batch.bytes for batch in selected_batches)),
+        "transfer_host": args.transfer_host,
+        "transfer_root": args.transfer_root,
+        "delete_after_transfer": args.delete_after_transfer,
+        "dry_run": args.dry_run,
+        "validate_only": args.validate_only,
+    }
+    write_summary(args.summary_json, summary)
+
+    for batch in selected_batches:
+        batch_root = args.relay_root / f"batch_{batch.index:04d}"
+        batch_root.mkdir(parents=True, exist_ok=True)
+        append_jsonl(
+            args.progress_jsonl,
+            {
+                "time": utc_now(),
+                "event": "batch_start",
+                "batch": batch.index,
+                "episode_count": len(batch.episodes),
+                "expected_bytes": batch.bytes,
+                "expected_human": human_bytes(batch.bytes),
+                "batch_root": str(batch_root),
+            },
+        )
+        if args.dry_run:
+            append_jsonl(
+                args.progress_jsonl,
+                {
+                    "time": utc_now(),
+                    "event": "batch_planned",
+                    "batch": batch.index,
+                    "episode_paths": [ep["episode_path"] for ep in batch.episodes],
+                    "files": [filename for ep in batch.episodes for filename in ep["download_files"]],
+                    "validation": "skipped_for_dry_run",
+                },
+            )
+            transfer_batch(args, batch_root, batch)
+            append_jsonl(args.progress_jsonl, {"time": utc_now(), "event": "batch_done", "batch": batch.index})
+            continue
+        if not args.dry_run and not args.validate_only:
+            download_batch(args.repo_id, token, batch_root, batch, args.progress_jsonl)
+        validation = validate_batch(batch_root, batch)
+        append_jsonl(args.progress_jsonl, {"time": utc_now(), "event": "batch_validated", "batch": batch.index, **validation})
+        if not validation["ok"]:
+            raise SystemExit(f"Batch {batch.index} validation failed: {validation}")
+        transfer_batch(args, batch_root, batch)
+        if args.delete_after_transfer and args.transfer_host and args.transfer_root and not args.dry_run:
+            shutil.rmtree(batch_root)
+            append_jsonl(args.progress_jsonl, {"time": utc_now(), "event": "batch_deleted", "batch": batch.index, "batch_root": str(batch_root)})
+        append_jsonl(args.progress_jsonl, {"time": utc_now(), "event": "batch_done", "batch": batch.index})
+
+    summary["status"] = "complete"
+    summary["completed_at_utc"] = utc_now()
+    write_summary(args.summary_json, summary)
+    print(json.dumps(summary, indent=2))
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

scripts/omni/select_xperience10m_pilot_episodes.py

@@ -0,0 +1,499 @@
+#!/usr/bin/env python3
+"""Select a metadata-balanced Xperience-10M pilot subset.
+
+The selector uses Hugging Face file metadata only. It does not download episode
+data. Content-category balancing is deferred until annotations are staged,
+because category text lives inside annotation.hdf5 files.
+"""
+
+from __future__ import annotations
+
+import argparse
+import csv
+import getpass
+import hashlib
+import json
+import os
+import re
+from collections import Counter, defaultdict
+from datetime import datetime, timezone
+from pathlib import Path
+from statistics import median
+from typing import Any
+
+from huggingface_hub import HfApi
+
+
+REQUIRED_FILES = [
+    "annotation.hdf5",
+    "fisheye_cam0.mp4",
+    "fisheye_cam1.mp4",
+    "fisheye_cam2.mp4",
+    "fisheye_cam3.mp4",
+    "stereo_left.mp4",
+    "stereo_right.mp4",
+]
+EXCLUDED_TRAINING_FILES = {"visualization.rrd"}
+SIZE_BANDS = ["short", "lower_mid", "upper_mid", "long"]
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("--repo-id", default="ropedia-ai/xperience-10m")
+    parser.add_argument("--target-episodes", type=int, default=128)
+    parser.add_argument("--seed", type=int, default=7)
+    parser.add_argument("--train-fraction", type=float, default=0.75)
+    parser.add_argument("--val-fraction", type=float, default=0.125)
+    parser.add_argument("--test-fraction", type=float, default=0.125)
+    parser.add_argument("--drop-bottom-annotation-percentile", type=float, default=0.05)
+    parser.add_argument("--drop-bottom-training-percentile", type=float, default=0.05)
+    parser.add_argument("--min-annotation-gib", type=float, default=0.5)
+    parser.add_argument("--windows-per-episode", type=int, default=256)
+    parser.add_argument("--output-json", type=Path, default=Path("results/omni_finetune/xperience10m_128_episode_selection.json"))
+    parser.add_argument("--output-csv", type=Path, default=Path("results/omni_finetune/xperience10m_128_episode_selection.csv"))
+    parser.add_argument("--download-list-output", type=Path, default=Path("results/omni_finetune/xperience10m_128_episode_download_files.txt"))
+    parser.add_argument("--report-output", type=Path, default=Path("results/omni_finetune/XPERIENCE10M_128_EPISODE_SELECTION.md"))
+    parser.add_argument("--token", default=os.environ.get("HF_TOKEN", "").strip())
+    return parser.parse_args()
+
+
+def file_size(sibling: Any) -> int:
+    value = getattr(sibling, "size", None)
+    if isinstance(value, int):
+        return value
+    lfs = getattr(sibling, "lfs", None)
+    if isinstance(lfs, dict) and isinstance(lfs.get("size"), int):
+        return int(lfs["size"])
+    return 0
+
+
+def human_bytes(num: float | int) -> str:
+    value = float(num)
+    for unit in ["B", "KiB", "MiB", "GiB", "TiB"]:
+        if abs(value) < 1024.0 or unit == "TiB":
+            return f"{value:.2f} {unit}"
+        value /= 1024.0
+    return f"{value:.2f} TiB"
+
+
+def quantile(values: list[int], q: float) -> int:
+    if not values:
+        return 0
+    ordered = sorted(values)
+    if len(ordered) == 1:
+        return ordered[0]
+    pos = min(max(q, 0.0), 1.0) * (len(ordered) - 1)
+    lo = int(pos)
+    hi = min(lo + 1, len(ordered) - 1)
+    frac = pos - lo
+    return int(round(ordered[lo] * (1.0 - frac) + ordered[hi] * frac))
+
+
+def summarize_sizes(values: list[int]) -> dict[str, Any]:
+    if not values:
+        return {"count": 0}
+    ordered = sorted(values)
+    return {
+        "count": len(ordered),
+        "min_bytes": ordered[0],
+        "p05_bytes": quantile(ordered, 0.05),
+        "p25_bytes": quantile(ordered, 0.25),
+        "median_bytes": int(median(ordered)),
+        "p75_bytes": quantile(ordered, 0.75),
+        "p95_bytes": quantile(ordered, 0.95),
+        "max_bytes": ordered[-1],
+        "mean_bytes": int(sum(ordered) / len(ordered)),
+        "min_human": human_bytes(ordered[0]),
+        "p05_human": human_bytes(quantile(ordered, 0.05)),
+        "p25_human": human_bytes(quantile(ordered, 0.25)),
+        "median_human": human_bytes(median(ordered)),
+        "p75_human": human_bytes(quantile(ordered, 0.75)),
+        "p95_human": human_bytes(quantile(ordered, 0.95)),
+        "max_human": human_bytes(ordered[-1]),
+        "mean_human": human_bytes(sum(ordered) / len(ordered)),
+    }
+
+
+def stable_hash(seed: int, text: str) -> str:
+    return hashlib.sha256(f"{seed}:{text}".encode("utf-8")).hexdigest()
+
+
+def stable_float(seed: int, text: str) -> float:
+    return int(stable_hash(seed, text)[:12], 16) / float(16**12)
+
+
+def episode_number(episode_id: str) -> int | None:
+    match = re.fullmatch(r"ep(\d+)", episode_id)
+    return int(match.group(1)) if match else None
+
+
+def size_band(annotation_bytes: int, q25: int, q50: int, q75: int) -> str:
+    if annotation_bytes <= q25:
+        return "short"
+    if annotation_bytes <= q50:
+        return "lower_mid"
+    if annotation_bytes <= q75:
+        return "upper_mid"
+    return "long"
+
+
+def build_episode_records(siblings: list[Any]) -> list[dict[str, Any]]:
+    by_parent: dict[str, dict[str, Any]] = defaultdict(lambda: {"files": {}, "bytes": 0})
+    for sibling in siblings:
+        path = str(getattr(sibling, "rfilename", ""))
+        if not path or path == ".gitattributes":
+            continue
+        name = Path(path).name
+        parent = Path(path).parent.as_posix()
+        if not parent:
+            continue
+        size = file_size(sibling)
+        bucket = by_parent[parent]
+        bucket["files"][name] = {"path": path, "bytes": size}
+        bucket["bytes"] += size
+
+    records = []
+    for parent, bucket in by_parent.items():
+        files = bucket["files"]
+        present = set(files)
+        if "annotation.hdf5" not in present:
+            continue
+        has_all_six_videos = all(name in present for name in REQUIRED_FILES[1:])
+        training_bytes = sum(
+            meta["bytes"]
+            for name, meta in files.items()
+            if name not in EXCLUDED_TRAINING_FILES
+        )
+        records.append(
+            {
+                "episode_path": parent,
+                "episode_id": Path(parent).name,
+                "episode_number": episode_number(Path(parent).name),
+                "top_level_session": parent.split("/", 1)[0],
+                "file_count": len(present),
+                "total_bytes": int(bucket["bytes"]),
+                "training_bytes_excluding_visualization_rrd": int(training_bytes),
+                "annotation_bytes": int(files["annotation.hdf5"]["bytes"]),
+                "video_bytes": int(sum(files[name]["bytes"] for name in REQUIRED_FILES[1:] if name in files)),
+                "has_annotation": True,
+                "has_all_six_videos": has_all_six_videos,
+                "has_visualization_rrd": "visualization.rrd" in present,
+                "missing_required_files": [name for name in REQUIRED_FILES if name not in present],
+                "download_files": [files[name]["path"] for name in REQUIRED_FILES if name in files],
+            }
+        )
+    return records
+
+
+def choose_target_counts(target: int) -> dict[str, int]:
+    base = target // len(SIZE_BANDS)
+    remainder = target % len(SIZE_BANDS)
+    return {
+        band: base + (1 if idx < remainder else 0)
+        for idx, band in enumerate(SIZE_BANDS)
+    }
+
+
+def select_balanced(records: list[dict[str, Any]], target: int, seed: int) -> list[dict[str, Any]]:
+    counts = choose_target_counts(target)
+    by_band: dict[str, list[dict[str, Any]]] = {band: [] for band in SIZE_BANDS}
+    band_medians = {
+        band: median([record["annotation_bytes"] for record in records if record["size_band"] == band])
+        for band in SIZE_BANDS
+        if any(record["size_band"] == band for record in records)
+    }
+
+    # Keep the best representative episode per session per band. This prevents
+    # one long session from dominating the sample.
+    session_band_best: dict[tuple[str, str], dict[str, Any]] = {}
+    global_training_median = median([record["training_bytes_excluding_visualization_rrd"] for record in records])
+    for record in records:
+        band = record["size_band"]
+        band_median = float(band_medians.get(band, record["annotation_bytes"]) or 1.0)
+        size_score = abs(record["annotation_bytes"] - band_median) / band_median
+        training_score = abs(record["training_bytes_excluding_visualization_rrd"] - global_training_median) / float(global_training_median or 1.0)
+        ep_num = record["episode_number"]
+        index_score = 0.0 if ep_num is None else min(ep_num / 64.0, 1.0) * 0.02
+        tie = stable_float(seed, record["episode_path"]) * 0.001
+        record["selection_score"] = round(float(size_score + 0.25 * training_score + index_score + tie), 8)
+        key = (record["top_level_session"], band)
+        current = session_band_best.get(key)
+        if current is None or record["selection_score"] < current["selection_score"]:
+            session_band_best[key] = record
+
+    for record in session_band_best.values():
+        by_band[record["size_band"]].append(record)
+    for band in SIZE_BANDS:
+        by_band[band].sort(key=lambda item: (item["selection_score"], stable_hash(seed, item["episode_path"])))
+
+    selected: list[dict[str, Any]] = []
+    used_sessions: set[str] = set()
+    selected_by_band = Counter()
+    for band in SIZE_BANDS:
+        for record in by_band[band]:
+            if selected_by_band[band] >= counts[band]:
+                break
+            if record["top_level_session"] in used_sessions:
+                continue
+            selected.append(record)
+            used_sessions.add(record["top_level_session"])
+            selected_by_band[band] += 1
+
+    if len(selected) < target:
+        remaining = [
+            record
+            for band in SIZE_BANDS
+            for record in by_band[band]
+            if record["top_level_session"] not in used_sessions
+        ]
+        remaining.sort(key=lambda item: (item["selection_score"], stable_hash(seed, item["episode_path"])))
+        for record in remaining:
+            selected.append(record)
+            used_sessions.add(record["top_level_session"])
+            selected_by_band[record["size_band"]] += 1
+            if len(selected) >= target:
+                break
+
+    if len(selected) < target:
+        raise RuntimeError(f"Only selected {len(selected)} unique-session episodes; target is {target}.")
+    return selected[:target]
+
+
+def assign_splits(selected: list[dict[str, Any]], seed: int, train_fraction: float, val_fraction: float, test_fraction: float) -> None:
+    total_fraction = train_fraction + val_fraction + test_fraction
+    if abs(total_fraction - 1.0) > 1e-6:
+        raise ValueError(f"Split fractions must sum to 1.0, got {total_fraction}")
+
+    for band in SIZE_BANDS:
+        band_records = [record for record in selected if record["size_band"] == band]
+        band_records.sort(key=lambda item: stable_hash(seed + 101, item["episode_path"]))
+        n = len(band_records)
+        val_n = int(round(n * val_fraction))
+        test_n = int(round(n * test_fraction))
+        train_n = n - val_n - test_n
+        for idx, record in enumerate(band_records):
+            if idx < train_n:
+                split = "train"
+            elif idx < train_n + val_n:
+                split = "val"
+            else:
+                split = "test"
+            record["split"] = split
+
+
+def md_table(headers: list[str], rows: list[list[Any]]) -> list[str]:
+    lines = [
+        "| " + " | ".join(headers) + " |",
+        "| " + " | ".join("---" for _ in headers) + " |",
+    ]
+    lines.extend("| " + " | ".join(str(cell) for cell in row) + " |" for row in rows)
+    return lines
+
+
+def write_csv(path: Path, rows: list[dict[str, Any]]) -> None:
+    path.parent.mkdir(parents=True, exist_ok=True)
+    fields = [
+        "selection_rank",
+        "split",
+        "size_band",
+        "episode_path",
+        "top_level_session",
+        "episode_id",
+        "annotation_human",
+        "training_human",
+        "annotation_bytes",
+        "training_bytes_excluding_visualization_rrd",
+        "has_visualization_rrd",
+        "selection_score",
+    ]
+    with path.open("w", newline="", encoding="utf-8") as handle:
+        writer = csv.DictWriter(handle, fieldnames=fields)
+        writer.writeheader()
+        for row in rows:
+            writer.writerow({field: row.get(field) for field in fields})
+
+
+def main() -> int:
+    args = parse_args()
+    token = args.token or getpass.getpass("HF token: ").strip()
+    if not token:
+        raise SystemExit("HF token is required for gated dataset metadata.")
+
+    api = HfApi(token=token)
+    info = api.repo_info(args.repo_id, repo_type="dataset", files_metadata=True, token=token)
+    records = build_episode_records(list(info.siblings or []))
+    complete = [record for record in records if record["has_all_six_videos"]]
+
+    annotation_sizes = [record["annotation_bytes"] for record in complete]
+    training_sizes = [record["training_bytes_excluding_visualization_rrd"] for record in complete]
+    q25 = quantile(annotation_sizes, 0.25)
+    q50 = quantile(annotation_sizes, 0.50)
+    q75 = quantile(annotation_sizes, 0.75)
+    min_annotation = max(
+        int(args.min_annotation_gib * (1024**3)),
+        quantile(annotation_sizes, args.drop_bottom_annotation_percentile),
+    )
+    min_training = quantile(training_sizes, args.drop_bottom_training_percentile)
+
+    candidates = []
+    rejected = Counter()
+    for record in complete:
+        if record["annotation_bytes"] < min_annotation:
+            rejected["annotation_too_small"] += 1
+            continue
+        if record["training_bytes_excluding_visualization_rrd"] < min_training:
+            rejected["training_too_small"] += 1
+            continue
+        record = dict(record)
+        record["size_band"] = size_band(record["annotation_bytes"], q25, q50, q75)
+        record["annotation_human"] = human_bytes(record["annotation_bytes"])
+        record["training_human"] = human_bytes(record["training_bytes_excluding_visualization_rrd"])
+        candidates.append(record)
+
+    selected = select_balanced(candidates, args.target_episodes, args.seed)
+    selected.sort(key=lambda item: (SIZE_BANDS.index(item["size_band"]), item["selection_score"], item["episode_path"]))
+    assign_splits(selected, args.seed, args.train_fraction, args.val_fraction, args.test_fraction)
+    for idx, record in enumerate(selected, start=1):
+        record["selection_rank"] = idx
+
+    selected_download_files = [
+        filename
+        for record in selected
+        for filename in record["download_files"]
+    ]
+    split_counts = Counter(record["split"] for record in selected)
+    band_counts = Counter(record["size_band"] for record in selected)
+    split_band_counts = Counter((record["split"], record["size_band"]) for record in selected)
+    selected_sessions = {record["top_level_session"] for record in selected}
+    train_sessions = {record["top_level_session"] for record in selected if record["split"] == "train"}
+    val_sessions = {record["top_level_session"] for record in selected if record["split"] == "val"}
+    test_sessions = {record["top_level_session"] for record in selected if record["split"] == "test"}
+
+    payload = {
+        "status": "pass",
+        "generated_at_utc": datetime.now(timezone.utc).isoformat(timespec="seconds"),
+        "repo_id": args.repo_id,
+        "repo_sha": getattr(info, "sha", None),
+        "selection_type": "metadata_balanced_first_pass",
+        "target_episodes": args.target_episodes,
+        "seed": args.seed,
+        "rules": {
+            "complete_episode_required_files": REQUIRED_FILES,
+            "excluded_training_files": sorted(EXCLUDED_TRAINING_FILES),
+            "one_episode_per_top_level_session": True,
+            "drop_bottom_annotation_percentile": args.drop_bottom_annotation_percentile,
+            "drop_bottom_training_percentile": args.drop_bottom_training_percentile,
+            "min_annotation_bytes": min_annotation,
+            "min_annotation_human": human_bytes(min_annotation),
+            "min_training_bytes": min_training,
+            "min_training_human": human_bytes(min_training),
+            "content_category_status": "not directly visible in HF metadata; refine after annotations are downloaded and captions are parsed",
+        },
+        "available_complete_episodes": len(complete),
+        "candidate_episodes_after_filters": len(candidates),
+        "rejected_counts": dict(rejected),
+        "annotation_size_summary_complete": summarize_sizes(annotation_sizes),
+        "training_size_summary_complete": summarize_sizes(training_sizes),
+        "selected_summary": {
+            "episode_count": len(selected),
+            "unique_session_count": len(selected_sessions),
+            "split_counts": dict(split_counts),
+            "size_band_counts": dict(band_counts),
+            "split_band_counts": {f"{split}/{band}": count for (split, band), count in split_band_counts.items()},
+            "estimated_download_bytes_excluding_visualization_rrd": sum(record["training_bytes_excluding_visualization_rrd"] for record in selected),
+            "estimated_download_human_excluding_visualization_rrd": human_bytes(sum(record["training_bytes_excluding_visualization_rrd"] for record in selected)),
+            "estimated_annotation_bytes": sum(record["annotation_bytes"] for record in selected),
+            "estimated_annotation_human": human_bytes(sum(record["annotation_bytes"] for record in selected)),
+            "estimated_windows_at_configured_limit": len(selected) * args.windows_per_episode,
+            "windows_per_episode": args.windows_per_episode,
+            "train_sessions_overlap_val": sorted(train_sessions & val_sessions),
+            "train_sessions_overlap_test": sorted(train_sessions & test_sessions),
+            "val_sessions_overlap_test": sorted(val_sessions & test_sessions),
+        },
+        "selected_episodes": selected,
+        "download_files": selected_download_files,
+    }
+
+    args.output_json.parent.mkdir(parents=True, exist_ok=True)
+    args.output_json.write_text(json.dumps(payload, indent=2) + "\n", encoding="utf-8")
+    write_csv(args.output_csv, selected)
+    args.download_list_output.write_text("\n".join(selected_download_files) + "\n", encoding="utf-8")
+
+    summary = payload["selected_summary"]
+    report = [
+        "# Xperience-10M 128-Episode Metadata-Balanced Selection",
+        "",
+        "This is a download plan, not a trained model result. It uses Hugging Face file metadata only and downloads no raw episode data.",
+        "",
+        "## Why This Selection",
+        "",
+        "- Use only complete episodes: `annotation.hdf5` plus six MP4 streams.",
+        "- Exclude `visualization.rrd` from the training download plan.",
+        "- Avoid tiny annotation outliers that are likely one-segment examples.",
+        "- Use one episode per top-level session to reduce leakage and overfitting to one capture session.",
+        "- Balance across four annotation-size bands as a proxy for duration/content richness before category labels are available.",
+        "- Split by session into train/val/test.",
+        "",
+        "## Selection Summary",
+        "",
+        *md_table(
+            ["Measure", "Value"],
+            [
+                ["Selected episodes", summary["episode_count"]],
+                ["Unique sessions", summary["unique_session_count"]],
+                ["Split counts", json.dumps(summary["split_counts"], sort_keys=True)],
+                ["Size-band counts", json.dumps(summary["size_band_counts"], sort_keys=True)],
+                ["Estimated training download, no RRD", summary["estimated_download_human_excluding_visualization_rrd"]],
+                ["Estimated annotation bytes", summary["estimated_annotation_human"]],
+                ["Estimated windows at 256/episode", summary["estimated_windows_at_configured_limit"]],
+                ["Session leakage train/val", len(summary["train_sessions_overlap_val"])],
+                ["Session leakage train/test", len(summary["train_sessions_overlap_test"])],
+                ["Session leakage val/test", len(summary["val_sessions_overlap_test"])],
+            ],
+        ),
+        "",
+        "## Filters",
+        "",
+        *md_table(
+            ["Rule", "Value"],
+            [
+                ["Available complete episodes", len(complete)],
+                ["Candidates after filters", len(candidates)],
+                ["Minimum annotation size", payload["rules"]["min_annotation_human"]],
+                ["Minimum training size", payload["rules"]["min_training_human"]],
+                ["Rejected counts", json.dumps(payload["rejected_counts"], sort_keys=True)],
+            ],
+        ),
+        "",
+        "## Split x Size Band",
+        "",
+        *md_table(
+            ["Split", *SIZE_BANDS],
+            [
+                [split, *[split_band_counts.get((split, band), 0) for band in SIZE_BANDS]]
+                for split in ["train", "val", "test"]
+            ],
+        ),
+        "",
+        "## Important Limitation",
+        "",
+        "HF metadata does not expose semantic content categories. This selection is the best first-pass balance before downloading. After the selected annotations are staged, parse `Main Task`, `Sub Task`, `Current Action`, objects, and interaction text; then swap episodes if one content cluster dominates.",
+        "",
+        "## Output Files",
+        "",
+        f"- JSON: `{args.output_json}`",
+        f"- CSV: `{args.output_csv}`",
+        f"- Download file list: `{args.download_list_output}`",
+    ]
+    args.report_output.write_text("\n".join(report) + "\n", encoding="utf-8")
+
+    print(json.dumps(payload["selected_summary"], indent=2))
+    print(f"PASS: wrote {args.output_json}")
+    print(f"PASS: wrote {args.output_csv}")
+    print(f"PASS: wrote {args.download_list_output}")
+    print(f"PASS: wrote {args.report_output}")
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

scripts/validate_publication_package.py

@@ -101,9 +101,9 @@ CARD_FRESHNESS_EXPECTATIONS = [
             "build_research_takeaways.py",
             "cc-by-nc-4.0",
             "12,103 episode folders",
-            "task-first 12-task infographic",
-            "native responsive modality atlas",
-            "interactive scrub/play task walkthrough storyboard",
+            "Ropedia Xperience-10M 12-task infographic",
+            "responsive native modality atlas",
+            "interactive scrub/play walkthrough storyboard",
             "website HTML",
             "task_surface_integrity.json",
             "rendered_site_check.json",
@@ -178,9 +178,9 @@ CARD_FRESHNESS_EXPECTATIONS = [
             "build_research_takeaways.py",
             "cc-by-nc-4.0",
             "12,103 episode folders",
-            "task-first 12-head",
-            "responsive modality atlas",
-            "interactive scrub/play storyboard",
+            "Ropedia Xperience-10M 12-task infographic",
+            "responsive native modality atlas",
+            "interactive scrub/play walkthrough storyboard",
             "website HTML",
             "task_surface_integrity.json",
             "rendered_site_check.json",

scripts/validate_source_alignment.py

@@ -92,7 +92,7 @@ CURRENT_PROJECT_LIMIT_MARKERS = [
     "neural rendering",
     "policy learning",
     "cross-episode generalization",
-    "real 32-episode Qwen3-Omni model quality",
+    "real held-out multi-episode Qwen3-Omni model quality",
 ]
 
 PRESENTATION_MARKERS = {
@@ -155,7 +155,7 @@ HF_PRESENTATION_MARKERS = {
         "HOMIE Toolkit",
         "Rerun 0.29.0",
         "12,103 episode folders",
-        "source-listing facts only",
+        "upstream listing metadata only",
         "limited in diversity",
     ],
     "artifacts/README.md": [
@@ -192,7 +192,7 @@ HF_PRESENTATION_MARKERS = {
         "Toolkit",
         "Rerun 0.29.0",
         "12,103 episode folders",
-        "upstream metadata facts",
+        "upstream listing metadata only",
         "limited in diversity",
     ],
 }