Upload folder using huggingface_hub

task40-200k/config.json

@@ -0,0 +1,524 @@
+{
+  "_name_or_path": "/shared_work/markhsp/runnings/v3_task40_lr2e5/ckpt-50000",
+  "action_mode": "auto",
+  "architectures": [
+    "XVLA"
+  ],
+  "auto_map": {
+    "AutoConfig": "configuration_xvla.XVLAConfig",
+    "AutoModel": "modeling_xvla.XVLA"
+  },
+  "depth": 24,
+  "dim_time": 32,
+  "florence_config": {
+    "_attn_implementation_autoset": true,
+    "bos_token_id": 0,
+    "eos_token_id": 2,
+    "is_encoder_decoder": true,
+    "model_type": "florence2",
+    "pad_token_id": 1,
+    "text_config": {
+      "_attn_implementation_autoset": true,
+      "_name_or_path": "",
+      "activation_dropout": 0.1,
+      "activation_function": "gelu",
+      "add_bias_logits": false,
+      "add_cross_attention": false,
+      "add_final_layer_norm": false,
+      "architectures": null,
+      "attention_dropout": 0.1,
+      "bad_words_ids": null,
+      "begin_suppress_tokens": null,
+      "bos_token_id": 0,
+      "chunk_size_feed_forward": 0,
+      "classif_dropout": 0.1,
+      "classifier_dropout": 0.0,
+      "cross_attention_hidden_size": null,
+      "d_model": 1024,
+      "decoder_attention_heads": 16,
+      "decoder_ffn_dim": 4096,
+      "decoder_layerdrop": 0.0,
+      "decoder_layers": 12,
+      "decoder_start_token_id": 2,
+      "diversity_penalty": 0.0,
+      "do_sample": false,
+      "dropout": 0.1,
+      "early_stopping": true,
+      "encoder_attention_heads": 16,
+      "encoder_ffn_dim": 4096,
+      "encoder_layerdrop": 0.0,
+      "encoder_layers": 12,
+      "encoder_no_repeat_ngram_size": 0,
+      "eos_token_id": 2,
+      "exponential_decay_length_penalty": null,
+      "finetuning_task": null,
+      "forced_bos_token_id": 0,
+      "forced_eos_token_id": 2,
+      "gradient_checkpointing": false,
+      "id2label": {
+        "0": "LABEL_0",
+        "1": "LABEL_1",
+        "2": "LABEL_2"
+      },
+      "init_std": 0.02,
+      "is_decoder": false,
+      "is_encoder_decoder": true,
+      "label2id": {
+        "LABEL_0": 0,
+        "LABEL_1": 1,
+        "LABEL_2": 2
+      },
+      "length_penalty": 1.0,
+      "max_length": 20,
+      "max_position_embeddings": 4096,
+      "min_length": 0,
+      "model_type": "florence2_language",
+      "no_repeat_ngram_size": 3,
+      "normalize_before": false,
+      "num_beam_groups": 1,
+      "num_beams": 3,
+      "num_hidden_layers": 12,
+      "num_return_sequences": 1,
+      "output_attentions": false,
+      "output_hidden_states": false,
+      "output_scores": false,
+      "pad_token_id": 1,
+      "prefix": null,
+      "problem_type": null,
+      "pruned_heads": {},
+      "remove_invalid_values": false,
+      "repetition_penalty": 1.0,
+      "return_dict": true,
+      "return_dict_in_generate": false,
+      "scale_embedding": false,
+      "sep_token_id": null,
+      "suppress_tokens": null,
+      "task_specific_params": null,
+      "temperature": 1.0,
+      "tf_legacy_loss": false,
+      "tie_encoder_decoder": false,
+      "tie_word_embeddings": true,
+      "tokenizer_class": null,
+      "top_k": 50,
+      "top_p": 1.0,
+      "torch_dtype": null,
+      "torchscript": false,
+      "typical_p": 1.0,
+      "use_bfloat16": false,
+      "use_cache": true,
+      "vocab_size": 51289
+    },
+    "torch_dtype": "float32",
+    "vision_config": {
+      "_attn_implementation_autoset": false,
+      "_name_or_path": "",
+      "add_cross_attention": false,
+      "architectures": null,
+      "bad_words_ids": null,
+      "begin_suppress_tokens": null,
+      "bos_token_id": null,
+      "chunk_size_feed_forward": 0,
+      "cross_attention_hidden_size": null,
+      "decoder_start_token_id": null,
+      "depths": [
+        1,
+        1,
+        9,
+        1
+      ],
+      "dim_embed": [
+        256,
+        512,
+        1024,
+        2048
+      ],
+      "diversity_penalty": 0.0,
+      "do_sample": false,
+      "drop_path_rate": 0.1,
+      "early_stopping": false,
+      "enable_checkpoint": false,
+      "encoder_no_repeat_ngram_size": 0,
+      "eos_token_id": null,
+      "exponential_decay_length_penalty": null,
+      "finetuning_task": null,
+      "forced_bos_token_id": null,
+      "forced_eos_token_id": null,
+      "id2label": {
+        "0": "LABEL_0",
+        "1": "LABEL_1"
+      },
+      "image_feature_source": [
+        "spatial_avg_pool",
+        "temporal_avg_pool"
+      ],
+      "image_pos_embed": {
+        "max_pos_embeddings": 50,
+        "type": "learned_abs_2d"
+      },
+      "is_decoder": false,
+      "is_encoder_decoder": false,
+      "label2id": {
+        "LABEL_0": 0,
+        "LABEL_1": 1
+      },
+      "length_penalty": 1.0,
+      "max_length": 20,
+      "min_length": 0,
+      "model_type": "davit",
+      "no_repeat_ngram_size": 0,
+      "num_beam_groups": 1,
+      "num_beams": 1,
+      "num_groups": [
+        8,
+        16,
+        32,
+        64
+      ],
+      "num_heads": [
+        8,
+        16,
+        32,
+        64
+      ],
+      "num_return_sequences": 1,
+      "output_attentions": false,
+      "output_hidden_states": false,
+      "output_scores": false,
+      "pad_token_id": null,
+      "patch_padding": [
+        3,
+        1,
+        1,
+        1
+      ],
+      "patch_prenorm": [
+        false,
+        true,
+        true,
+        true
+      ],
+      "patch_size": [
+        7,
+        3,
+        3,
+        3
+      ],
+      "patch_stride": [
+        4,
+        2,
+        2,
+        2
+      ],
+      "prefix": null,
+      "problem_type": null,
+      "projection_dim": 1024,
+      "pruned_heads": {},
+      "remove_invalid_values": false,
+      "repetition_penalty": 1.0,
+      "return_dict": true,
+      "return_dict_in_generate": false,
+      "sep_token_id": null,
+      "suppress_tokens": null,
+      "task_specific_params": null,
+      "temperature": 1.0,
+      "tf_legacy_loss": false,
+      "tie_encoder_decoder": false,
+      "tie_word_embeddings": true,
+      "tokenizer_class": null,
+      "top_k": 50,
+      "top_p": 1.0,
+      "torch_dtype": null,
+      "torchscript": false,
+      "typical_p": 1.0,
+      "use_bfloat16": false,
+      "visual_temporal_embedding": {
+        "max_temporal_embeddings": 100,
+        "type": "COSINE"
+      },
+      "window_size": 12
+    }
+  },
+  "hidden_size": 1024,
+  "len_soft_prompts": 32,
+  "max_action_dim": 23,
+  "max_len_seq": 512,
+  "mlp_ratio": 4.0,
+  "model_type": "xvla",
+  "num_actions": 30,
+  "num_domains": 30,
+  "num_heads": 16,
+  "num_objects": 172,
+  "num_skills": 34,
+  "num_tasks": 50,
+  "object_class_weights": [
+    0.4501,
+    0.4205,
+    0.3228,
+    0.3057,
+    0.3521,
+    7.3765,
+    0.2465,
+    0.5885,
+    0.4622,
+    0.3009,
+    0.281,
+    0.3189,
+    0.1221,
+    0.4479,
+    0.312,
+    0.4966,
+    0.3079,
+    0.2972,
+    0.5216,
+    0.1139,
+    0.2091,
+    0.2318,
+    0.1754,
+    0.2122,
+    0.3788,
+    0.2522,
+    0.2869,
+    0.2588,
+    0.5604,
+    0.2401,
+    0.4887,
+    0.2479,
+    0.4339,
+    0.2858,
+    0.4104,
+    0.1655,
+    0.5658,
+    0.1832,
+    0.4361,
+    0.3443,
+    0.4715,
+    0.4035,
+    0.3555,
+    0.5542,
+    0.5411,
+    0.3756,
+    0.6119,
+    0.3592,
+    0.3795,
+    0.17,
+    0.6553,
+    1.8098,
+    0.2901,
+    0.2248,
+    0.2027,
+    0.4599,
+    0.5536,
+    0.508,
+    0.2674,
+    0.5678,
+    0.3297,
+    0.4523,
+    0.2186,
+    0.2029,
+    0.4368,
+    0.432,
+    0.0822,
+    0.3718,
+    0.6172,
+    0.3748,
+    0.2685,
+    0.2913,
+    0.4405,
+    0.3391,
+    0.3676,
+    6.0925,
+    2.0803,
+    2.153,
+    2.2968,
+    2.1286,
+    12.6916,
+    0.5856,
+    5.4459,
+    2.6565,
+    13.1846,
+    23.4236,
+    0.478,
+    0.462,
+    0.1174,
+    0.4406,
+    0.2615,
+    0.2668,
+    0.2285,
+    0.3084,
+    0.2523,
+    0.3026,
+    0.2844,
+    0.5057,
+    0.343,
+    0.2129,
+    0.2497,
+    0.3409,
+    0.3574,
+    0.3983,
+    0.1783,
+    0.2182,
+    0.3948,
+    3.4148,
+    0.3134,
+    0.4798,
+    0.3458,
+    0.2877,
+    0.4385,
+    0.2716,
+    0.4203,
+    0.2383,
+    0.2431,
+    0.2523,
+    0.4795,
+    0.4197,
+    0.2102,
+    0.15,
+    0.3516,
+    1.1414,
+    0.1527,
+    0.4403,
+    0.188,
+    0.5024,
+    0.367,
+    0.4584,
+    0.293,
+    0.3649,
+    23.0659,
+    0.1816,
+    0.4641,
+    0.438,
+    0.2461,
+    2.3487,
+    0.3496,
+    0.405,
+    0.3314,
+    0.489,
+    0.4397,
+    0.4996,
+    0.1229,
+    3.6319,
+    0.4439,
+    0.2366,
+    0.2766,
+    0.2168,
+    0.3219,
+    0.3337,
+    0.4465,
+    0.7571,
+    0.257,
+    0.3779,
+    0.1382,
+    0.2154,
+    0.2358,
+    0.4292,
+    0.3343,
+    0.3047,
+    0.8751,
+    0.3987,
+    0.146,
+    0.2943,
+    0.8747,
+    0.1515,
+    0.1481,
+    0.5527,
+    2.8632,
+    0.3923
+  ],
+  "object_classifier_weight": 0.1,
+  "real_action_dim": 23,
+  "skill_class_weights": [
+    0.0859,
+    0.1021,
+    0.2138,
+    0.1674,
+    0.8973,
+    1.059,
+    1.6727,
+    0.9436,
+    0.1984,
+    1.1787,
+    0.2744,
+    0.621,
+    0.6252,
+    1.8634,
+    0.5304,
+    0.5203,
+    0.8339,
+    0.8988,
+    1.4175,
+    2.2557,
+    0.5537,
+    1.1661,
+    3.0415,
+    0.1975,
+    0.3228,
+    1.2419,
+    0.6963,
+    1.2705,
+    0.4855,
+    1.1792,
+    1.4312,
+    2.6619,
+    1.3428,
+    2.0497
+  ],
+  "skill_classifier_weight": 0.1,
+  "soft_prompt_length": 32,
+  "task_class_weights": [
+    2.173,
+    1.4191,
+    0.8104,
+    0.8292,
+    0.8919,
+    0.9968,
+    1.1146,
+    0.707,
+    1.029,
+    0.8308,
+    1.2263,
+    0.928,
+    1.0689,
+    0.6989,
+    0.8124,
+    0.8448,
+    0.8113,
+    1.0045,
+    0.9249,
+    0.9423,
+    0.8886,
+    0.7005,
+    1.1158,
+    0.6354,
+    0.6881,
+    0.8477,
+    0.5946,
+    0.7707,
+    0.7753,
+    0.6626,
+    1.0688,
+    1.0711,
+    1.0735,
+    0.9245,
+    2.0169,
+    1.6108,
+    0.8854,
+    1.4132,
+    1.2177,
+    1.0697,
+    1.7812,
+    0.8162,
+    1.2199,
+    0.7956,
+    0.9433,
+    1.0236,
+    1.1092,
+    0.8773,
+    0.6393,
+    0.6997
+  ],
+  "task_classifier_weight": 0.1,
+  "torch_dtype": "float32",
+  "transformers_version": "4.49.0",
+  "use_hetero_proj": false,
+  "use_proprio": true
+}

task40-200k/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f5f5a0b222c46b1a532f69702ac2853b6f9364d9a576f1632301de75ca9355f
+size 3554780612

task40-200k/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1de2a086be93a8e489739e90cdd87e2d227415bc76d9aa1506d8a0310234f943
+size 7100195053

task40-200k/preprocessor_config.json

@@ -0,0 +1,32 @@
+{
+  "auto_map": {
+    "AutoProcessor": "processing_xvla.XVLAProcessor"
+  },
+  "crop_size": {
+    "height": 224,
+    "width": 224
+  },
+  "do_center_crop": false,
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "image_processor_type": "CLIPImageProcessor",
+  "image_std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "processor_class": "XVLAProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 224,
+    "width": 224
+  }
+}

task40-200k/special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

task40-200k/state.json

@@ -0,0 +1 @@
+{"global_step": 200000}

task40-200k/tokenizer_config.json

@@ -0,0 +1,66 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50264": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "max_length": 50,
+  "model_max_length": 1024,
+  "pad_to_multiple_of": null,
+  "pad_token": "<pad>",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "processor_class": "XVLAProcessor",
+  "sep_token": "</s>",
+  "stride": 0,
+  "tokenizer_class": "BartTokenizer",
+  "trim_offsets": true,
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "<unk>"
+}