Hugging Face's logo

limxdynamics
/
FluxVLAEngine

Safetensors
Model card Files
xet
 Community
8
ShenBW commited on
Commit
276fd0f
·
verified ·
1 Parent(s): 725bdac

Add GR00T N1.5 RoboTwin model metadata

Browse files
Files changed (10) hide show
  1. gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/config.json +419 -0
  2. gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/config.yaml +326 -0
  3. gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/dataset_statistics.json +156 -0
  4. gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/tokenizer/added_tokens.json +39 -0
  5. gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/tokenizer/chat_template.json +3 -0
  6. gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/tokenizer/merges.txt +0 -0
  7. gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/tokenizer/special_tokens_map.json +42 -0
  8. gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/tokenizer/tokenizer_config.json +344 -0
  9. gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/tokenizer/vocab.json +0 -0
  10. gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/vlm_backbone_config.json +115 -0
gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/config.json ADDED
@@ -0,0 +1,419 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_ROBOTWIN_TASK_LIST": [
3
+ "adjust_bottle",
4
+ "beat_block_hammer",
5
+ "blocks_ranking_rgb",
6
+ "blocks_ranking_size",
7
+ "click_alarmclock",
8
+ "click_bell",
9
+ "dump_bin_bigbin",
10
+ "grab_roller",
11
+ "handover_block",
12
+ "handover_mic",
13
+ "hanging_mug",
14
+ "lift_pot",
15
+ "move_can_pot",
16
+ "move_pillbottle_pad",
17
+ "move_playingcard_away",
18
+ "move_stapler_pad",
19
+ "open_laptop",
20
+ "open_microwave",
21
+ "pick_diverse_bottles",
22
+ "pick_dual_bottles",
23
+ "place_a2b_left",
24
+ "place_a2b_right",
25
+ "place_bread_basket",
26
+ "place_bread_skillet",
27
+ "place_burger_fries",
28
+ "place_can_basket",
29
+ "place_cans_plasticbox",
30
+ "place_container_plate",
31
+ "place_dual_shoes",
32
+ "place_empty_cup",
33
+ "place_fan",
34
+ "place_mouse_pad",
35
+ "place_object_basket",
36
+ "place_object_scale",
37
+ "place_object_stand",
38
+ "place_phone_stand",
39
+ "place_shoe",
40
+ "press_stapler",
41
+ "put_bottles_dustbin",
42
+ "put_object_cabinet",
43
+ "rotate_qrcode",
44
+ "scan_object",
45
+ "shake_bottle",
46
+ "shake_bottle_horizontally",
47
+ "stack_blocks_three",
48
+ "stack_blocks_two",
49
+ "stack_bowls_three",
50
+ "stack_bowls_two",
51
+ "stamp_seal",
52
+ "turn_switch"
53
+ ],
54
+ "eval": {
55
+ "dataset": {
56
+ "embodiment_id": 0,
57
+ "img_keys": [
58
+ "cam_high",
59
+ "cam_left_wrist",
60
+ "cam_right_wrist"
61
+ ],
62
+ "transforms": [
63
+ {
64
+ "max_len": 900,
65
+ "num_images": 3,
66
+ "tokenizer": {
67
+ "type": "PretrainedTokenizer"
68
+ },
69
+ "type": "ProcessPromptsWithImage"
70
+ },
71
+ {
72
+ "height": 224,
73
+ "type": "ResizeImages",
74
+ "width": 224
75
+ },
76
+ {
77
+ "means": [
78
+ [
79
+ 123.515625,
80
+ 116.04492188,
81
+ 103.59375
82
+ ],
83
+ [
84
+ 123.515625,
85
+ 116.04492188,
86
+ 103.59375
87
+ ],
88
+ [
89
+ 123.515625,
90
+ 116.04492188,
91
+ 103.59375
92
+ ]
93
+ ],
94
+ "stds": [
95
+ [
96
+ 58.27148438,
97
+ 57.02636719,
98
+ 57.27539062
99
+ ],
100
+ [
101
+ 58.27148438,
102
+ 57.02636719,
103
+ 57.27539062
104
+ ],
105
+ [
106
+ 58.27148438,
107
+ 57.02636719,
108
+ 57.27539062
109
+ ]
110
+ ],
111
+ "type": "NormalizeImages"
112
+ },
113
+ {
114
+ "action_key": "action",
115
+ "norm_type": "mean_std",
116
+ "state_dim": 64,
117
+ "state_key": "proprio",
118
+ "type": "NormalizeStatesAndActions"
119
+ }
120
+ ],
121
+ "type": "PrivateInferenceDataset"
122
+ },
123
+ "denormalize_action": {
124
+ "action_dim": 14,
125
+ "norm_type": "mean_std",
126
+ "type": "DenormalizePrivateAction"
127
+ },
128
+ "eval_chunk_size": 50,
129
+ "model_family": "groot",
130
+ "num_trials_per_task": 100,
131
+ "seed": 7,
132
+ "task_list": [
133
+ "adjust_bottle",
134
+ "beat_block_hammer",
135
+ "blocks_ranking_rgb",
136
+ "blocks_ranking_size",
137
+ "click_alarmclock",
138
+ "click_bell",
139
+ "dump_bin_bigbin",
140
+ "grab_roller",
141
+ "handover_block",
142
+ "handover_mic",
143
+ "hanging_mug",
144
+ "lift_pot",
145
+ "move_can_pot",
146
+ "move_pillbottle_pad",
147
+ "move_playingcard_away",
148
+ "move_stapler_pad",
149
+ "open_laptop",
150
+ "open_microwave",
151
+ "pick_diverse_bottles",
152
+ "pick_dual_bottles",
153
+ "place_a2b_left",
154
+ "place_a2b_right",
155
+ "place_bread_basket",
156
+ "place_bread_skillet",
157
+ "place_burger_fries",
158
+ "place_can_basket",
159
+ "place_cans_plasticbox",
160
+ "place_container_plate",
161
+ "place_dual_shoes",
162
+ "place_empty_cup",
163
+ "place_fan",
164
+ "place_mouse_pad",
165
+ "place_object_basket",
166
+ "place_object_scale",
167
+ "place_object_stand",
168
+ "place_phone_stand",
169
+ "place_shoe",
170
+ "press_stapler",
171
+ "put_bottles_dustbin",
172
+ "put_object_cabinet",
173
+ "rotate_qrcode",
174
+ "scan_object",
175
+ "shake_bottle",
176
+ "shake_bottle_horizontally",
177
+ "stack_blocks_three",
178
+ "stack_blocks_two",
179
+ "stack_bowls_three",
180
+ "stack_bowls_two",
181
+ "stamp_seal",
182
+ "turn_switch"
183
+ ],
184
+ "type": "RobotwinEvalRunner"
185
+ },
186
+ "inference_model": {
187
+ "pretrained_name_or_path": "./checkpoints/GR00T-N1.5-3B",
188
+ "type": "LlavaVLA",
189
+ "vla_head": {
190
+ "action_dim": 32,
191
+ "diffusion_model_cfg": {
192
+ "attention_head_dim": 48,
193
+ "cross_attention_dim": 2048,
194
+ "dropout": 0.2,
195
+ "final_dropout": true,
196
+ "interleave_self_attention": true,
197
+ "norm_type": "ada_norm",
198
+ "num_attention_heads": 32,
199
+ "num_layers": 16,
200
+ "output_dim": 1024,
201
+ "positional_embeddings": null
202
+ },
203
+ "hidden_size": 1024,
204
+ "input_embedding_dim": 1536,
205
+ "max_input_seq_len": 900,
206
+ "num_inference_timesteps": 4,
207
+ "num_steps": 50,
208
+ "ori_action_dim": 14,
209
+ "state_dim": 64,
210
+ "type": "FlowMatchingInferenceHead"
211
+ },
212
+ "vlm_backbone": {
213
+ "type": "EagleInferenceBackbone",
214
+ "vlm_config": {
215
+ "max_input_seq_len": 900
216
+ },
217
+ "vlm_path": "fluxvla/models/third_party_models/eagle2_hg_model"
218
+ }
219
+ },
220
+ "model": {
221
+ "freeze_projector": false,
222
+ "freeze_vlm_backbone": false,
223
+ "name_mapping": {
224
+ "vla_head": "action_head",
225
+ "vlm_backbone.vlm": "backbone.eagle_model"
226
+ },
227
+ "pretrained_name_or_path": "./checkpoints/GR00T-N1.5-3B",
228
+ "type": "LlavaVLA",
229
+ "vla_head": {
230
+ "action_dim": 32,
231
+ "hidden_size": 1024,
232
+ "input_embedding_dim": 1536,
233
+ "num_inference_timesteps": 4,
234
+ "num_steps": 50,
235
+ "ori_action_dim": 14,
236
+ "state_dim": 64,
237
+ "type": "FlowMatchingHead"
238
+ },
239
+ "vlm_backbone": {
240
+ "type": "EagleBackbone",
241
+ "vlm_config": {
242
+ "max_input_seq_len": 900
243
+ },
244
+ "vlm_path": "fluxvla/models/third_party_models/eagle2_hg_model"
245
+ }
246
+ },
247
+ "runner": {
248
+ "change_key_name": false,
249
+ "collator": {
250
+ "keys": [
251
+ "states",
252
+ "timestamp",
253
+ "images",
254
+ "img_masks",
255
+ "lang_tokens",
256
+ "lang_masks",
257
+ "actions",
258
+ "action_masks",
259
+ "embodiment_ids"
260
+ ],
261
+ "meta_keys": [
262
+ "task_description",
263
+ "prompt",
264
+ "info",
265
+ "stats"
266
+ ],
267
+ "type": "DictCollator"
268
+ },
269
+ "enable_gradient_checkpointing": false,
270
+ "enable_mixed_precision_training": true,
271
+ "lr_scheduler": {
272
+ "type": "linear-warmup+cosine-decay",
273
+ "warmup_ratio": 0.03
274
+ },
275
+ "max_epochs": 3,
276
+ "max_grad_norm": 1.0,
277
+ "max_keep_ckpts": 3,
278
+ "metric": {
279
+ "active_trackers": [
280
+ "jsonl",
281
+ "wandb"
282
+ ],
283
+ "grad_accumulation_steps": 1,
284
+ "run_dir": "work_dirs",
285
+ "type": "VLAMetric",
286
+ "window_size": 1
287
+ },
288
+ "mixed_precision_dtype": "bf16",
289
+ "optimizer": {
290
+ "lr": 1e-05,
291
+ "paramwise_learning_rate": {
292
+ "vla_head": 0.0001
293
+ },
294
+ "type": "AdamW",
295
+ "weight_decay": 0.0
296
+ },
297
+ "sampler": null,
298
+ "tokenizer": {
299
+ "model_path": "fluxvla/models/third_party_models/eagle2_hg_model",
300
+ "type": "PretrainedTokenizer"
301
+ },
302
+ "type": "FSDPTrainRunner"
303
+ },
304
+ "train_dataloader": {
305
+ "dataset": {
306
+ "datasets": [
307
+ {
308
+ "action_window_size": 50,
309
+ "data_root_path": [
310
+ "datasets/robotwin_clean_lerobotv2.1",
311
+ "datasets/robotwin_randomized_lerobotv2.1"
312
+ ],
313
+ "transforms": [
314
+ {
315
+ "embodiment_id": 0,
316
+ "name_mappings": {
317
+ "observation.state": [
318
+ "states"
319
+ ]
320
+ },
321
+ "parquet_keys": [
322
+ "observation.state",
323
+ "timestamp",
324
+ "actions",
325
+ "info",
326
+ "stats",
327
+ "action_masks"
328
+ ],
329
+ "type": "ProcessParquetInputs",
330
+ "video_keys": [
331
+ "observation.images.cam_high",
332
+ "observation.images.cam_left_wrist",
333
+ "observation.images.cam_right_wrist"
334
+ ]
335
+ },
336
+ {
337
+ "type": "ParquetPrompter"
338
+ },
339
+ {
340
+ "max_len": 900,
341
+ "num_images": 3,
342
+ "tokenizer": {
343
+ "model_path": "fluxvla/models/third_party_models/eagle2_hg_model",
344
+ "type": "PretrainedTokenizer"
345
+ },
346
+ "type": "ProcessPromptsWithImage"
347
+ },
348
+ {
349
+ "height": 224,
350
+ "type": "ResizeImages",
351
+ "width": 224
352
+ },
353
+ {
354
+ "means": [
355
+ [
356
+ 123.515625,
357
+ 116.04492188,
358
+ 103.59375
359
+ ],
360
+ [
361
+ 123.515625,
362
+ 116.04492188,
363
+ 103.59375
364
+ ],
365
+ [
366
+ 123.515625,
367
+ 116.04492188,
368
+ 103.59375
369
+ ]
370
+ ],
371
+ "stds": [
372
+ [
373
+ 58.27148438,
374
+ 57.02636719,
375
+ 57.27539062
376
+ ],
377
+ [
378
+ 58.27148438,
379
+ 57.02636719,
380
+ 57.27539062
381
+ ],
382
+ [
383
+ 58.27148438,
384
+ 57.02636719,
385
+ 57.27539062
386
+ ]
387
+ ],
388
+ "type": "NormalizeImages"
389
+ },
390
+ {
391
+ "action_dim": 32,
392
+ "action_key": "action",
393
+ "norm_type": "mean_std",
394
+ "state_dim": 64,
395
+ "state_key": "proprio",
396
+ "type": "NormalizeStatesAndActions"
397
+ }
398
+ ],
399
+ "type": "ParquetDataset"
400
+ }
401
+ ],
402
+ "name_mappings": {
403
+ "observation.state": [
404
+ "proprio",
405
+ "action"
406
+ ]
407
+ },
408
+ "reshuffle_each_epoch": true,
409
+ "seed": 7,
410
+ "statistic_keys": [
411
+ "observation.state",
412
+ "timestamp"
413
+ ],
414
+ "type": "DistributedRepeatingDataset"
415
+ },
416
+ "per_device_batch_size": 8,
417
+ "per_device_num_workers": 4
418
+ }
419
+ }
gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/config.yaml ADDED
@@ -0,0 +1,326 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ _ROBOTWIN_TASK_LIST:
2
+ - adjust_bottle
3
+ - beat_block_hammer
4
+ - blocks_ranking_rgb
5
+ - blocks_ranking_size
6
+ - click_alarmclock
7
+ - click_bell
8
+ - dump_bin_bigbin
9
+ - grab_roller
10
+ - handover_block
11
+ - handover_mic
12
+ - hanging_mug
13
+ - lift_pot
14
+ - move_can_pot
15
+ - move_pillbottle_pad
16
+ - move_playingcard_away
17
+ - move_stapler_pad
18
+ - open_laptop
19
+ - open_microwave
20
+ - pick_diverse_bottles
21
+ - pick_dual_bottles
22
+ - place_a2b_left
23
+ - place_a2b_right
24
+ - place_bread_basket
25
+ - place_bread_skillet
26
+ - place_burger_fries
27
+ - place_can_basket
28
+ - place_cans_plasticbox
29
+ - place_container_plate
30
+ - place_dual_shoes
31
+ - place_empty_cup
32
+ - place_fan
33
+ - place_mouse_pad
34
+ - place_object_basket
35
+ - place_object_scale
36
+ - place_object_stand
37
+ - place_phone_stand
38
+ - place_shoe
39
+ - press_stapler
40
+ - put_bottles_dustbin
41
+ - put_object_cabinet
42
+ - rotate_qrcode
43
+ - scan_object
44
+ - shake_bottle
45
+ - shake_bottle_horizontally
46
+ - stack_blocks_three
47
+ - stack_blocks_two
48
+ - stack_bowls_three
49
+ - stack_bowls_two
50
+ - stamp_seal
51
+ - turn_switch
52
+ eval:
53
+ dataset:
54
+ embodiment_id: 0
55
+ img_keys:
56
+ - cam_high
57
+ - cam_left_wrist
58
+ - cam_right_wrist
59
+ transforms:
60
+ - max_len: 900
61
+ num_images: 3
62
+ tokenizer:
63
+ type: PretrainedTokenizer
64
+ type: ProcessPromptsWithImage
65
+ - height: 224
66
+ type: ResizeImages
67
+ width: 224
68
+ - means:
69
+ - - 123.515625
70
+ - 116.04492188
71
+ - 103.59375
72
+ - - 123.515625
73
+ - 116.04492188
74
+ - 103.59375
75
+ - - 123.515625
76
+ - 116.04492188
77
+ - 103.59375
78
+ stds:
79
+ - - 58.27148438
80
+ - 57.02636719
81
+ - 57.27539062
82
+ - - 58.27148438
83
+ - 57.02636719
84
+ - 57.27539062
85
+ - - 58.27148438
86
+ - 57.02636719
87
+ - 57.27539062
88
+ type: NormalizeImages
89
+ - action_key: action
90
+ norm_type: mean_std
91
+ state_dim: 64
92
+ state_key: proprio
93
+ type: NormalizeStatesAndActions
94
+ type: PrivateInferenceDataset
95
+ denormalize_action:
96
+ action_dim: 14
97
+ norm_type: mean_std
98
+ type: DenormalizePrivateAction
99
+ eval_chunk_size: 50
100
+ model_family: groot
101
+ num_trials_per_task: 100
102
+ seed: 7
103
+ task_list:
104
+ - adjust_bottle
105
+ - beat_block_hammer
106
+ - blocks_ranking_rgb
107
+ - blocks_ranking_size
108
+ - click_alarmclock
109
+ - click_bell
110
+ - dump_bin_bigbin
111
+ - grab_roller
112
+ - handover_block
113
+ - handover_mic
114
+ - hanging_mug
115
+ - lift_pot
116
+ - move_can_pot
117
+ - move_pillbottle_pad
118
+ - move_playingcard_away
119
+ - move_stapler_pad
120
+ - open_laptop
121
+ - open_microwave
122
+ - pick_diverse_bottles
123
+ - pick_dual_bottles
124
+ - place_a2b_left
125
+ - place_a2b_right
126
+ - place_bread_basket
127
+ - place_bread_skillet
128
+ - place_burger_fries
129
+ - place_can_basket
130
+ - place_cans_plasticbox
131
+ - place_container_plate
132
+ - place_dual_shoes
133
+ - place_empty_cup
134
+ - place_fan
135
+ - place_mouse_pad
136
+ - place_object_basket
137
+ - place_object_scale
138
+ - place_object_stand
139
+ - place_phone_stand
140
+ - place_shoe
141
+ - press_stapler
142
+ - put_bottles_dustbin
143
+ - put_object_cabinet
144
+ - rotate_qrcode
145
+ - scan_object
146
+ - shake_bottle
147
+ - shake_bottle_horizontally
148
+ - stack_blocks_three
149
+ - stack_blocks_two
150
+ - stack_bowls_three
151
+ - stack_bowls_two
152
+ - stamp_seal
153
+ - turn_switch
154
+ type: RobotwinEvalRunner
155
+ inference_model:
156
+ pretrained_name_or_path: ./checkpoints/GR00T-N1.5-3B
157
+ type: LlavaVLA
158
+ vla_head:
159
+ action_dim: 32
160
+ diffusion_model_cfg:
161
+ attention_head_dim: 48
162
+ cross_attention_dim: 2048
163
+ dropout: 0.2
164
+ final_dropout: true
165
+ interleave_self_attention: true
166
+ norm_type: ada_norm
167
+ num_attention_heads: 32
168
+ num_layers: 16
169
+ output_dim: 1024
170
+ positional_embeddings: null
171
+ hidden_size: 1024
172
+ input_embedding_dim: 1536
173
+ max_input_seq_len: 900
174
+ num_inference_timesteps: 4
175
+ num_steps: 50
176
+ ori_action_dim: 14
177
+ state_dim: 64
178
+ type: FlowMatchingInferenceHead
179
+ vlm_backbone:
180
+ type: EagleInferenceBackbone
181
+ vlm_config:
182
+ max_input_seq_len: 900
183
+ vlm_path: fluxvla/models/third_party_models/eagle2_hg_model
184
+ model:
185
+ freeze_projector: false
186
+ freeze_vlm_backbone: false
187
+ name_mapping:
188
+ vla_head: action_head
189
+ vlm_backbone.vlm: backbone.eagle_model
190
+ pretrained_name_or_path: ./checkpoints/GR00T-N1.5-3B
191
+ type: LlavaVLA
192
+ vla_head:
193
+ action_dim: 32
194
+ hidden_size: 1024
195
+ input_embedding_dim: 1536
196
+ num_inference_timesteps: 4
197
+ num_steps: 50
198
+ ori_action_dim: 14
199
+ state_dim: 64
200
+ type: FlowMatchingHead
201
+ vlm_backbone:
202
+ type: EagleBackbone
203
+ vlm_config:
204
+ max_input_seq_len: 900
205
+ vlm_path: fluxvla/models/third_party_models/eagle2_hg_model
206
+ runner:
207
+ change_key_name: false
208
+ collator:
209
+ keys:
210
+ - states
211
+ - timestamp
212
+ - images
213
+ - img_masks
214
+ - lang_tokens
215
+ - lang_masks
216
+ - actions
217
+ - action_masks
218
+ - embodiment_ids
219
+ meta_keys:
220
+ - task_description
221
+ - prompt
222
+ - info
223
+ - stats
224
+ type: DictCollator
225
+ enable_gradient_checkpointing: false
226
+ enable_mixed_precision_training: true
227
+ lr_scheduler:
228
+ type: linear-warmup+cosine-decay
229
+ warmup_ratio: 0.03
230
+ max_epochs: 3
231
+ max_grad_norm: 1.0
232
+ max_keep_ckpts: 3
233
+ metric:
234
+ active_trackers:
235
+ - jsonl
236
+ - wandb
237
+ grad_accumulation_steps: 1
238
+ run_dir: work_dirs
239
+ type: VLAMetric
240
+ window_size: 1
241
+ mixed_precision_dtype: bf16
242
+ optimizer:
243
+ lr: 1.0e-05
244
+ paramwise_learning_rate:
245
+ vla_head: 0.0001
246
+ type: AdamW
247
+ weight_decay: 0.0
248
+ sampler: null
249
+ tokenizer:
250
+ model_path: fluxvla/models/third_party_models/eagle2_hg_model
251
+ type: PretrainedTokenizer
252
+ type: FSDPTrainRunner
253
+ train_dataloader:
254
+ dataset:
255
+ datasets:
256
+ - action_window_size: 50
257
+ data_root_path:
258
+ - datasets/robotwin_clean_lerobotv2.1
259
+ - datasets/robotwin_randomized_lerobotv2.1
260
+ transforms:
261
+ - embodiment_id: 0
262
+ name_mappings:
263
+ observation.state:
264
+ - states
265
+ parquet_keys:
266
+ - observation.state
267
+ - timestamp
268
+ - actions
269
+ - info
270
+ - stats
271
+ - action_masks
272
+ type: ProcessParquetInputs
273
+ video_keys:
274
+ - observation.images.cam_high
275
+ - observation.images.cam_left_wrist
276
+ - observation.images.cam_right_wrist
277
+ - type: ParquetPrompter
278
+ - max_len: 900
279
+ num_images: 3
280
+ tokenizer:
281
+ model_path: fluxvla/models/third_party_models/eagle2_hg_model
282
+ type: PretrainedTokenizer
283
+ type: ProcessPromptsWithImage
284
+ - height: 224
285
+ type: ResizeImages
286
+ width: 224
287
+ - means:
288
+ - - 123.515625
289
+ - 116.04492188
290
+ - 103.59375
291
+ - - 123.515625
292
+ - 116.04492188
293
+ - 103.59375
294
+ - - 123.515625
295
+ - 116.04492188
296
+ - 103.59375
297
+ stds:
298
+ - - 58.27148438
299
+ - 57.02636719
300
+ - 57.27539062
301
+ - - 58.27148438
302
+ - 57.02636719
303
+ - 57.27539062
304
+ - - 58.27148438
305
+ - 57.02636719
306
+ - 57.27539062
307
+ type: NormalizeImages
308
+ - action_dim: 32
309
+ action_key: action
310
+ norm_type: mean_std
311
+ state_dim: 64
312
+ state_key: proprio
313
+ type: NormalizeStatesAndActions
314
+ type: ParquetDataset
315
+ name_mappings:
316
+ observation.state:
317
+ - proprio
318
+ - action
319
+ reshuffle_each_epoch: true
320
+ seed: 7
321
+ statistic_keys:
322
+ - observation.state
323
+ - timestamp
324
+ type: DistributedRepeatingDataset
325
+ per_device_batch_size: 8
326
+ per_device_num_workers: 4
gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/dataset_statistics.json ADDED
@@ -0,0 +1,156 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "private": {
3
+ "proprio": {
4
+ "mean": [
5
+ -0.23836564500637852,
6
+ 1.129817402830952,
7
+ 0.8068359079301254,
8
+ -0.31024214960817226,
9
+ 0.05489044029082383,
10
+ -0.03797947365911228,
11
+ 0.6655638025405416,
12
+ 0.21664742843549212,
13
+ 1.0790448055651551,
14
+ 0.7875585893326063,
15
+ -0.3211320959513645,
16
+ -0.017501951884490514,
17
+ 0.03539706856287519,
18
+ 0.6794509037859313
19
+ ],
20
+ "std": [
21
+ 0.41515260848742136,
22
+ 1.0045809755922652,
23
+ 0.7768178536932145,
24
+ 0.6732816910017168,
25
+ 0.2708186547486596,
26
+ 0.6197771059972782,
27
+ 0.45465163653888563,
28
+ 0.34920798291604505,
29
+ 1.0249604849195586,
30
+ 0.794372303822,
31
+ 0.6847885016239902,
32
+ 0.24437403723179849,
33
+ 0.6142938232023921,
34
+ 0.44978408667126524
35
+ ],
36
+ "min": [
37
+ -7.552278995513916,
38
+ -0.36354875564575195,
39
+ -0.18577136099338531,
40
+ -1.956291913986206,
41
+ -1.6801013946533203,
42
+ -7.678869724273682,
43
+ 0.0,
44
+ -8.539926528930664,
45
+ -0.6294453740119934,
46
+ -0.07775841653347015,
47
+ -2.1328067779541016,
48
+ -2.1285502910614014,
49
+ -8.228925704956055,
50
+ 0.0
51
+ ],
52
+ "max": [
53
+ 6.45659065246582,
54
+ 4.179152488708496,
55
+ 5.346591472625732,
56
+ 1.7942548990249634,
57
+ 1.8604620695114136,
58
+ 5.43813943862915,
59
+ 1.0,
60
+ 7.142920017242432,
61
+ 4.157068729400635,
62
+ 5.672450065612793,
63
+ 1.95806884765625,
64
+ 1.5663840770721436,
65
+ 5.278968811035156,
66
+ 1.0
67
+ ],
68
+ "q01": null,
69
+ "q99": null
70
+ },
71
+ "action": {
72
+ "mean": [
73
+ -0.23836564500637852,
74
+ 1.129817402830952,
75
+ 0.8068359079301254,
76
+ -0.31024214960817226,
77
+ 0.05489044029082383,
78
+ -0.03797947365911228,
79
+ 0.6655638025405416,
80
+ 0.21664742843549212,
81
+ 1.0790448055651551,
82
+ 0.7875585893326063,
83
+ -0.3211320959513645,
84
+ -0.017501951884490514,
85
+ 0.03539706856287519,
86
+ 0.6794509037859313
87
+ ],
88
+ "std": [
89
+ 0.41515260848742136,
90
+ 1.0045809755922652,
91
+ 0.7768178536932145,
92
+ 0.6732816910017168,
93
+ 0.2708186547486596,
94
+ 0.6197771059972782,
95
+ 0.45465163653888563,
96
+ 0.34920798291604505,
97
+ 1.0249604849195586,
98
+ 0.794372303822,
99
+ 0.6847885016239902,
100
+ 0.24437403723179849,
101
+ 0.6142938232023921,
102
+ 0.44978408667126524
103
+ ],
104
+ "min": [
105
+ -7.552278995513916,
106
+ -0.36354875564575195,
107
+ -0.18577136099338531,
108
+ -1.956291913986206,
109
+ -1.6801013946533203,
110
+ -7.678869724273682,
111
+ 0.0,
112
+ -8.539926528930664,
113
+ -0.6294453740119934,
114
+ -0.07775841653347015,
115
+ -2.1328067779541016,
116
+ -2.1285502910614014,
117
+ -8.228925704956055,
118
+ 0.0
119
+ ],
120
+ "max": [
121
+ 6.45659065246582,
122
+ 4.179152488708496,
123
+ 5.346591472625732,
124
+ 1.7942548990249634,
125
+ 1.8604620695114136,
126
+ 5.43813943862915,
127
+ 1.0,
128
+ 7.142920017242432,
129
+ 4.157068729400635,
130
+ 5.672450065612793,
131
+ 1.95806884765625,
132
+ 1.5663840770721436,
133
+ 5.278968811035156,
134
+ 1.0
135
+ ],
136
+ "q01": null,
137
+ "q99": null
138
+ },
139
+ "timestamp": {
140
+ "mean": [
141
+ 9.822102670522625
142
+ ],
143
+ "std": [
144
+ 8.539596174071674
145
+ ],
146
+ "min": [
147
+ 0.0
148
+ ],
149
+ "max": [
150
+ 72.2
151
+ ],
152
+ "q01": null,
153
+ "q99": null
154
+ }
155
+ }
156
+ }
gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/tokenizer/added_tokens.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</box>": 151673,
3
+ "</img>": 151671,
4
+ "</interval>": 151679,
5
+ "</quad>": 151675,
6
+ "</ref>": 151677,
7
+ "</think>": 151668,
8
+ "</tool_call>": 151658,
9
+ "</tool_response>": 151666,
10
+ "<IMG_CONTEXT>": 151669,
11
+ "<box>": 151672,
12
+ "<img>": 151670,
13
+ "<interval>": 151678,
14
+ "<quad>": 151674,
15
+ "<ref>": 151676,
16
+ "<think>": 151667,
17
+ "<tool_call>": 151657,
18
+ "<tool_response>": 151665,
19
+ "<|box_end|>": 151649,
20
+ "<|box_start|>": 151648,
21
+ "<|endoftext|>": 151643,
22
+ "<|file_sep|>": 151664,
23
+ "<|fim_middle|>": 151660,
24
+ "<|fim_pad|>": 151662,
25
+ "<|fim_prefix|>": 151659,
26
+ "<|fim_suffix|>": 151661,
27
+ "<|im_end|>": 151645,
28
+ "<|im_start|>": 151644,
29
+ "<|image_pad|>": 151655,
30
+ "<|object_ref_end|>": 151647,
31
+ "<|object_ref_start|>": 151646,
32
+ "<|quad_end|>": 151651,
33
+ "<|quad_start|>": 151650,
34
+ "<|repo_name|>": 151663,
35
+ "<|video_pad|>": 151656,
36
+ "<|vision_end|>": 151653,
37
+ "<|vision_pad|>": 151654,
38
+ "<|vision_start|>": 151652
39
+ }
gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/tokenizer/chat_template.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}<image {{ image_count.value }}>{% endif %}<image-{{ image_count.value }}>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}<video {{ video_count.value }}>{% endif %}<video-{{ video_count.value }}>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
3
+ }
gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/tokenizer/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/tokenizer/special_tokens_map.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>",
16
+ "<IMG_CONTEXT>",
17
+ "<img>",
18
+ "</img>",
19
+ "<box>",
20
+ "</box>",
21
+ "<quad>",
22
+ "</quad>",
23
+ "<ref>",
24
+ "</ref>",
25
+ "<interval>",
26
+ "</interval>"
27
+ ],
28
+ "eos_token": {
29
+ "content": "<|im_end|>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false
34
+ },
35
+ "pad_token": {
36
+ "content": "<|endoftext|>",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false
41
+ }
42
+ }
gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/tokenizer/tokenizer_config.json ADDED
@@ -0,0 +1,344 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": false,
5
+ "added_tokens_decoder": {
6
+ "151643": {
7
+ "content": "<|endoftext|>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "151644": {
15
+ "content": "<|im_start|>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "151645": {
23
+ "content": "<|im_end|>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "151646": {
31
+ "content": "<|object_ref_start|>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "151647": {
39
+ "content": "<|object_ref_end|>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ },
46
+ "151648": {
47
+ "content": "<|box_start|>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": false,
51
+ "single_word": false,
52
+ "special": true
53
+ },
54
+ "151649": {
55
+ "content": "<|box_end|>",
56
+ "lstrip": false,
57
+ "normalized": false,
58
+ "rstrip": false,
59
+ "single_word": false,
60
+ "special": true
61
+ },
62
+ "151650": {
63
+ "content": "<|quad_start|>",
64
+ "lstrip": false,
65
+ "normalized": false,
66
+ "rstrip": false,
67
+ "single_word": false,
68
+ "special": true
69
+ },
70
+ "151651": {
71
+ "content": "<|quad_end|>",
72
+ "lstrip": false,
73
+ "normalized": false,
74
+ "rstrip": false,
75
+ "single_word": false,
76
+ "special": true
77
+ },
78
+ "151652": {
79
+ "content": "<|vision_start|>",
80
+ "lstrip": false,
81
+ "normalized": false,
82
+ "rstrip": false,
83
+ "single_word": false,
84
+ "special": true
85
+ },
86
+ "151653": {
87
+ "content": "<|vision_end|>",
88
+ "lstrip": false,
89
+ "normalized": false,
90
+ "rstrip": false,
91
+ "single_word": false,
92
+ "special": true
93
+ },
94
+ "151654": {
95
+ "content": "<|vision_pad|>",
96
+ "lstrip": false,
97
+ "normalized": false,
98
+ "rstrip": false,
99
+ "single_word": false,
100
+ "special": true
101
+ },
102
+ "151655": {
103
+ "content": "<|image_pad|>",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": false,
107
+ "single_word": false,
108
+ "special": true
109
+ },
110
+ "151656": {
111
+ "content": "<|video_pad|>",
112
+ "lstrip": false,
113
+ "normalized": false,
114
+ "rstrip": false,
115
+ "single_word": false,
116
+ "special": true
117
+ },
118
+ "151657": {
119
+ "content": "<tool_call>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": false,
123
+ "single_word": false,
124
+ "special": false
125
+ },
126
+ "151658": {
127
+ "content": "</tool_call>",
128
+ "lstrip": false,
129
+ "normalized": false,
130
+ "rstrip": false,
131
+ "single_word": false,
132
+ "special": false
133
+ },
134
+ "151659": {
135
+ "content": "<|fim_prefix|>",
136
+ "lstrip": false,
137
+ "normalized": false,
138
+ "rstrip": false,
139
+ "single_word": false,
140
+ "special": false
141
+ },
142
+ "151660": {
143
+ "content": "<|fim_middle|>",
144
+ "lstrip": false,
145
+ "normalized": false,
146
+ "rstrip": false,
147
+ "single_word": false,
148
+ "special": false
149
+ },
150
+ "151661": {
151
+ "content": "<|fim_suffix|>",
152
+ "lstrip": false,
153
+ "normalized": false,
154
+ "rstrip": false,
155
+ "single_word": false,
156
+ "special": false
157
+ },
158
+ "151662": {
159
+ "content": "<|fim_pad|>",
160
+ "lstrip": false,
161
+ "normalized": false,
162
+ "rstrip": false,
163
+ "single_word": false,
164
+ "special": false
165
+ },
166
+ "151663": {
167
+ "content": "<|repo_name|>",
168
+ "lstrip": false,
169
+ "normalized": false,
170
+ "rstrip": false,
171
+ "single_word": false,
172
+ "special": false
173
+ },
174
+ "151664": {
175
+ "content": "<|file_sep|>",
176
+ "lstrip": false,
177
+ "normalized": false,
178
+ "rstrip": false,
179
+ "single_word": false,
180
+ "special": false
181
+ },
182
+ "151665": {
183
+ "content": "<tool_response>",
184
+ "lstrip": false,
185
+ "normalized": false,
186
+ "rstrip": false,
187
+ "single_word": false,
188
+ "special": false
189
+ },
190
+ "151666": {
191
+ "content": "</tool_response>",
192
+ "lstrip": false,
193
+ "normalized": false,
194
+ "rstrip": false,
195
+ "single_word": false,
196
+ "special": false
197
+ },
198
+ "151667": {
199
+ "content": "<think>",
200
+ "lstrip": false,
201
+ "normalized": false,
202
+ "rstrip": false,
203
+ "single_word": false,
204
+ "special": false
205
+ },
206
+ "151668": {
207
+ "content": "</think>",
208
+ "lstrip": false,
209
+ "normalized": false,
210
+ "rstrip": false,
211
+ "single_word": false,
212
+ "special": false
213
+ },
214
+ "151669": {
215
+ "content": "<IMG_CONTEXT>",
216
+ "lstrip": false,
217
+ "normalized": false,
218
+ "rstrip": false,
219
+ "single_word": false,
220
+ "special": true
221
+ },
222
+ "151670": {
223
+ "content": "<img>",
224
+ "lstrip": false,
225
+ "normalized": false,
226
+ "rstrip": false,
227
+ "single_word": false,
228
+ "special": true
229
+ },
230
+ "151671": {
231
+ "content": "</img>",
232
+ "lstrip": false,
233
+ "normalized": false,
234
+ "rstrip": false,
235
+ "single_word": false,
236
+ "special": true
237
+ },
238
+ "151672": {
239
+ "content": "<box>",
240
+ "lstrip": false,
241
+ "normalized": false,
242
+ "rstrip": false,
243
+ "single_word": false,
244
+ "special": true
245
+ },
246
+ "151673": {
247
+ "content": "</box>",
248
+ "lstrip": false,
249
+ "normalized": false,
250
+ "rstrip": false,
251
+ "single_word": false,
252
+ "special": true
253
+ },
254
+ "151674": {
255
+ "content": "<quad>",
256
+ "lstrip": false,
257
+ "normalized": false,
258
+ "rstrip": false,
259
+ "single_word": false,
260
+ "special": true
261
+ },
262
+ "151675": {
263
+ "content": "</quad>",
264
+ "lstrip": false,
265
+ "normalized": false,
266
+ "rstrip": false,
267
+ "single_word": false,
268
+ "special": true
269
+ },
270
+ "151676": {
271
+ "content": "<ref>",
272
+ "lstrip": false,
273
+ "normalized": false,
274
+ "rstrip": false,
275
+ "single_word": false,
276
+ "special": true
277
+ },
278
+ "151677": {
279
+ "content": "</ref>",
280
+ "lstrip": false,
281
+ "normalized": false,
282
+ "rstrip": false,
283
+ "single_word": false,
284
+ "special": true
285
+ },
286
+ "151678": {
287
+ "content": "<interval>",
288
+ "lstrip": false,
289
+ "normalized": false,
290
+ "rstrip": false,
291
+ "single_word": false,
292
+ "special": true
293
+ },
294
+ "151679": {
295
+ "content": "</interval>",
296
+ "lstrip": false,
297
+ "normalized": false,
298
+ "rstrip": false,
299
+ "single_word": false,
300
+ "special": true
301
+ }
302
+ },
303
+ "additional_special_tokens": [
304
+ "<|im_start|>",
305
+ "<|im_end|>",
306
+ "<|object_ref_start|>",
307
+ "<|object_ref_end|>",
308
+ "<|box_start|>",
309
+ "<|box_end|>",
310
+ "<|quad_start|>",
311
+ "<|quad_end|>",
312
+ "<|vision_start|>",
313
+ "<|vision_end|>",
314
+ "<|vision_pad|>",
315
+ "<|image_pad|>",
316
+ "<|video_pad|>",
317
+ "<IMG_CONTEXT>",
318
+ "<img>",
319
+ "</img>",
320
+ "<box>",
321
+ "</box>",
322
+ "<quad>",
323
+ "</quad>",
324
+ "<ref>",
325
+ "</ref>",
326
+ "<interval>",
327
+ "</interval>"
328
+ ],
329
+ "auto_map": {
330
+ "AutoProcessor": "processing_eagle2_5_vl.Eagle2_5_VLProcessor"
331
+ },
332
+ "bos_token": null,
333
+ "chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0].role == 'system' %}\n {{- messages[0].content + '\\n\\n' }}\n {%- endif %}\n {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0].role == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n {%- set index = (messages|length - 1) - loop.index0 %}\n {%- if ns.multi_step_tool and message.role == \"user\" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}\n {%- set ns.multi_step_tool = false %}\n {%- set ns.last_query_index = index %}\n {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {%- set content = message.content %}\n {%- set reasoning_content = '' %}\n {%- if message.reasoning_content is defined and message.reasoning_content is not none %}\n {%- set reasoning_content = message.reasoning_content %}\n {%- else %}\n {%- if '</think>' in message.content %}\n {%- set content = message.content.split('</think>')[-1].lstrip('\\n') %}\n {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n {%- endif %}\n {%- endif %}\n {%- if loop.index0 > ns.last_query_index %}\n {%- if loop.last or (not loop.last and reasoning_content) %}\n {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n {%- else %}\n {{- '<|im_start|>' + message.role + '\\n' + content }}\n {%- endif %}\n {%- else %}\n {{- '<|im_start|>' + message.role + '\\n' + content }}\n {%- endif %}\n {%- if message.tool_calls %}\n {%- for tool_call in message.tool_calls %}\n {%- if (loop.first and content) or (not loop.first) %}\n {{- '\\n' }}\n {%- endif %}\n {%- if tool_call.function %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {%- if tool_call.arguments is string %}\n {{- tool_call.arguments }}\n {%- else %}\n {{- tool_call.arguments | tojson }}\n {%- endif %}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n {%- if enable_thinking is defined and enable_thinking is false %}\n {{- '<think>\\n\\n</think>\\n\\n' }}\n {%- endif %}\n{%- endif %}",
334
+ "clean_up_tokenization_spaces": false,
335
+ "eos_token": "<|im_end|>",
336
+ "errors": "replace",
337
+ "extra_special_tokens": {},
338
+ "model_max_length": 16384,
339
+ "pad_token": "<|endoftext|>",
340
+ "processor_class": "Eagle2_5_VLProcessor",
341
+ "split_special_tokens": false,
342
+ "tokenizer_class": "Qwen2Tokenizer",
343
+ "unk_token": null
344
+ }
gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/tokenizer/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
gr00t_eagle_3b_robotwin_all_data_full_finetune_bs128/vlm_backbone_config.json ADDED
@@ -0,0 +1,115 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation": "flash_attention_2",
3
+ "_attn_implementation_autoset": false,
4
+ "architectures": [
5
+ "Eagle2_5_VLForConditionalGeneration"
6
+ ],
7
+ "attn_implementation": "flash_attention_2",
8
+ "auto_map": {
9
+ "AutoConfig": "configuration_eagle2_5_vl.Eagle2_5_VLConfig",
10
+ "AutoModel": "modeling_eagle2_5_vl.Eagle2_5_VLForConditionalGeneration"
11
+ },
12
+ "downsample_ratio": 0.5,
13
+ "dtype": "bfloat16",
14
+ "dynamic_image_size": true,
15
+ "force_image_size": 224,
16
+ "image_token_index": 151669,
17
+ "initializer_range": 0.02,
18
+ "loss_version": "efficient_v2_cp_head",
19
+ "max_dynamic_tiles": 12,
20
+ "min_dynamic_tiles": 1,
21
+ "mlp_checkpoint": false,
22
+ "mlp_connector_layers": 1,
23
+ "model_type": "eagle_2_5_vl",
24
+ "output_attentions": false,
25
+ "pad2square": false,
26
+ "select_layer": -1,
27
+ "template": "qwen3-chat",
28
+ "text_config": {
29
+ "_attn_implementation_autoset": true,
30
+ "_name_or_path": "Qwen/Qwen3-1.7B",
31
+ "architectures": [
32
+ "Qwen3ForCausalLM"
33
+ ],
34
+ "attention_bias": false,
35
+ "attention_dropout": 0,
36
+ "attn_implementation": "flash_attention_2",
37
+ "bos_token_id": 151643,
38
+ "dtype": "bfloat16",
39
+ "eos_token_id": 151645,
40
+ "head_dim": 128,
41
+ "hidden_act": "silu",
42
+ "hidden_size": 2048,
43
+ "initializer_range": 0.02,
44
+ "intermediate_size": 6144,
45
+ "layer_types": [
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention",
53
+ "full_attention",
54
+ "full_attention",
55
+ "full_attention",
56
+ "full_attention",
57
+ "full_attention",
58
+ "full_attention",
59
+ "full_attention",
60
+ "full_attention",
61
+ "full_attention",
62
+ "full_attention",
63
+ "full_attention",
64
+ "full_attention",
65
+ "full_attention",
66
+ "full_attention",
67
+ "full_attention",
68
+ "full_attention",
69
+ "full_attention",
70
+ "full_attention",
71
+ "full_attention",
72
+ "full_attention",
73
+ "full_attention"
74
+ ],
75
+ "max_position_embeddings": 40960,
76
+ "max_window_layers": 28,
77
+ "model_type": "qwen3",
78
+ "num_attention_heads": 16,
79
+ "num_hidden_layers": 12,
80
+ "num_key_value_heads": 8,
81
+ "pad_token_id": null,
82
+ "rms_norm_eps": 1e-06,
83
+ "rope_parameters": {
84
+ "rope_theta": 1000000,
85
+ "rope_type": "default"
86
+ },
87
+ "sliding_window": null,
88
+ "tie_word_embeddings": true,
89
+ "use_cache": false,
90
+ "use_sliding_window": false,
91
+ "vocab_size": 151680
92
+ },
93
+ "tie_word_embeddings": true,
94
+ "transformers_version": null,
95
+ "use_backbone_lora": 0,
96
+ "use_llm_lora": 0,
97
+ "use_pixel_shuffle": false,
98
+ "use_thumbnail": true,
99
+ "vision_config": {
100
+ "_attn_implementation_autoset": true,
101
+ "attention_dropout": 0,
102
+ "attn_implementation": "flash_attention_2",
103
+ "dtype": "bfloat16",
104
+ "hidden_act": "gelu_pytorch_tanh",
105
+ "hidden_size": 1152,
106
+ "image_size": 224,
107
+ "intermediate_size": 4304,
108
+ "layer_norm_eps": 1e-06,
109
+ "model_type": "siglip_vision_model",
110
+ "num_attention_heads": 16,
111
+ "num_channels": 3,
112
+ "num_hidden_layers": 27,
113
+ "patch_size": 14
114
+ }
115
+ }