Hugging Face's logo

pentacore-HRC2026
/
checkpoints

Safetensors
Model card Files
xet
 Community
1
NithanNguyen commited on
Commit
a9f16a0
·
1 Parent(s): a157127

ân

Browse files
Files changed (14) hide show
  1. Foam_Inlaying/v2/config.json +88 -0
  2. Packing_Box/v2.zip → Foam_Inlaying/v2/model.safetensors +2 -2
  3. Foam_Inlaying/v2/policy_postprocessor.json +32 -0
  4. Foam_Inlaying/v2/policy_postprocessor_step_0_unnormalizer_processor.safetensors +3 -0
  5. Foam_Inlaying/v2/policy_preprocessor.json +80 -0
  6. Foam_Inlaying/v2/policy_preprocessor_step_3_normalizer_processor.safetensors +3 -0
  7. Foam_Inlaying/v2/train_config.json +638 -0
  8. Packing_Box/v2/config.json +110 -0
  9. Packing_Box/v2/model.safetensors +3 -0
  10. Packing_Box/v2/policy_postprocessor.json +32 -0
  11. Packing_Box/v2/policy_postprocessor_step_0_unnormalizer_processor.safetensors +3 -0
  12. Packing_Box/v2/policy_preprocessor.json +80 -0
  13. Packing_Box/v2/policy_preprocessor_step_3_normalizer_processor.safetensors +3 -0
  14. Packing_Box/v2/train_config.json +311 -0
Foam_Inlaying/v2/config.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "act",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 62
9
+ ]
10
+ },
11
+ "observation.images.head_left": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 480,
16
+ 640
17
+ ]
18
+ },
19
+ "observation.images.head_right": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 480,
24
+ 640
25
+ ]
26
+ },
27
+ "observation.images.wrist_left": {
28
+ "type": "VISUAL",
29
+ "shape": [
30
+ 3,
31
+ 480,
32
+ 640
33
+ ]
34
+ },
35
+ "observation.images.wrist_right": {
36
+ "type": "VISUAL",
37
+ "shape": [
38
+ 3,
39
+ 480,
40
+ 640
41
+ ]
42
+ }
43
+ },
44
+ "output_features": {
45
+ "action": {
46
+ "type": "ACTION",
47
+ "shape": [
48
+ 20
49
+ ]
50
+ }
51
+ },
52
+ "device": "cuda",
53
+ "use_amp": false,
54
+ "use_peft": false,
55
+ "push_to_hub": false,
56
+ "repo_id": null,
57
+ "private": null,
58
+ "tags": null,
59
+ "license": null,
60
+ "pretrained_path": null,
61
+ "pretrained_revision": null,
62
+ "chunk_size": 100,
63
+ "n_action_steps": 100,
64
+ "normalization_mapping": {
65
+ "VISUAL": "MEAN_STD",
66
+ "STATE": "MEAN_STD",
67
+ "ACTION": "MEAN_STD"
68
+ },
69
+ "vision_backbone": "resnet18",
70
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
71
+ "replace_final_stride_with_dilation": false,
72
+ "pre_norm": false,
73
+ "dim_model": 512,
74
+ "n_heads": 8,
75
+ "dim_feedforward": 3200,
76
+ "feedforward_activation": "relu",
77
+ "n_encoder_layers": 4,
78
+ "n_decoder_layers": 1,
79
+ "use_vae": true,
80
+ "latent_dim": 32,
81
+ "n_vae_encoder_layers": 4,
82
+ "temporal_ensemble_coeff": null,
83
+ "dropout": 0.1,
84
+ "kl_weight": 10.0,
85
+ "optimizer_lr": 1e-05,
86
+ "optimizer_weight_decay": 0.0001,
87
+ "optimizer_lr_backbone": 1e-05
88
+ }
Packing_Box/v2.zip → Foam_Inlaying/v2/model.safetensors RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ff4cc2a61a2e0ecb10a742d91ec47c4c28894cdb6e96a54f5241d8b27d32f329
3
- size 1148298428
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6edf2c31516ae79026ed82ffaaba4571ef53dd90d76b356db46ffa2187a9b9e8
3
+ size 206986512
Foam_Inlaying/v2/policy_postprocessor.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "policy_postprocessor",
3
+ "steps": [
4
+ {
5
+ "registry_name": "unnormalizer_processor",
6
+ "config": {
7
+ "eps": 1e-08,
8
+ "features": {
9
+ "action": {
10
+ "type": "ACTION",
11
+ "shape": [
12
+ 20
13
+ ]
14
+ }
15
+ },
16
+ "norm_map": {
17
+ "VISUAL": "MEAN_STD",
18
+ "STATE": "MEAN_STD",
19
+ "ACTION": "MEAN_STD"
20
+ }
21
+ },
22
+ "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors"
23
+ },
24
+ {
25
+ "registry_name": "device_processor",
26
+ "config": {
27
+ "device": "cpu",
28
+ "float_dtype": null
29
+ }
30
+ }
31
+ ]
32
+ }
Foam_Inlaying/v2/policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dbd9783e28607d0ef2be5da64c9e98a02492b91dce4dc809b89baed03cb23f0d
3
+ size 12392
Foam_Inlaying/v2/policy_preprocessor.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "policy_preprocessor",
3
+ "steps": [
4
+ {
5
+ "registry_name": "rename_observations_processor",
6
+ "config": {
7
+ "rename_map": {}
8
+ }
9
+ },
10
+ {
11
+ "registry_name": "to_batch_processor",
12
+ "config": {}
13
+ },
14
+ {
15
+ "registry_name": "device_processor",
16
+ "config": {
17
+ "device": "cuda",
18
+ "float_dtype": null
19
+ }
20
+ },
21
+ {
22
+ "registry_name": "normalizer_processor",
23
+ "config": {
24
+ "eps": 1e-08,
25
+ "features": {
26
+ "observation.state": {
27
+ "type": "STATE",
28
+ "shape": [
29
+ 62
30
+ ]
31
+ },
32
+ "observation.images.head_left": {
33
+ "type": "VISUAL",
34
+ "shape": [
35
+ 3,
36
+ 480,
37
+ 640
38
+ ]
39
+ },
40
+ "observation.images.head_right": {
41
+ "type": "VISUAL",
42
+ "shape": [
43
+ 3,
44
+ 480,
45
+ 640
46
+ ]
47
+ },
48
+ "observation.images.wrist_left": {
49
+ "type": "VISUAL",
50
+ "shape": [
51
+ 3,
52
+ 480,
53
+ 640
54
+ ]
55
+ },
56
+ "observation.images.wrist_right": {
57
+ "type": "VISUAL",
58
+ "shape": [
59
+ 3,
60
+ 480,
61
+ 640
62
+ ]
63
+ },
64
+ "action": {
65
+ "type": "ACTION",
66
+ "shape": [
67
+ 20
68
+ ]
69
+ }
70
+ },
71
+ "norm_map": {
72
+ "VISUAL": "MEAN_STD",
73
+ "STATE": "MEAN_STD",
74
+ "ACTION": "MEAN_STD"
75
+ }
76
+ },
77
+ "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors"
78
+ }
79
+ ]
80
+ }
Foam_Inlaying/v2/policy_preprocessor_step_3_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:407d9014cc1f5ad77ea2934a5deb899e740b42b84852dbd6a72f6dfbe9131864
3
+ size 12408
Foam_Inlaying/v2/train_config.json ADDED
@@ -0,0 +1,638 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "ghrc2026-foam-inlaying",
4
+ "repo_type": "dataset",
5
+ "root": "/kaggle/input/datasets/nguynltrn/ghrc2026-foam-inlaying/Foam_Inlaying_Train",
6
+ "episodes": [
7
+ 0,
8
+ 1,
9
+ 2,
10
+ 3,
11
+ 4,
12
+ 5,
13
+ 6,
14
+ 7,
15
+ 8,
16
+ 9,
17
+ 10,
18
+ 11,
19
+ 12,
20
+ 13,
21
+ 14,
22
+ 15,
23
+ 16,
24
+ 17,
25
+ 18,
26
+ 19,
27
+ 20,
28
+ 21,
29
+ 22,
30
+ 23,
31
+ 24,
32
+ 25,
33
+ 26,
34
+ 27,
35
+ 28,
36
+ 29,
37
+ 30,
38
+ 31,
39
+ 32,
40
+ 33,
41
+ 34,
42
+ 35,
43
+ 36,
44
+ 37,
45
+ 38,
46
+ 39,
47
+ 40,
48
+ 41,
49
+ 42,
50
+ 43,
51
+ 44,
52
+ 45,
53
+ 46,
54
+ 47,
55
+ 48,
56
+ 49,
57
+ 50,
58
+ 51,
59
+ 52,
60
+ 53,
61
+ 54,
62
+ 55,
63
+ 56,
64
+ 57,
65
+ 58,
66
+ 59,
67
+ 60,
68
+ 61,
69
+ 62,
70
+ 63,
71
+ 64,
72
+ 65,
73
+ 66,
74
+ 67,
75
+ 68,
76
+ 69,
77
+ 70,
78
+ 71,
79
+ 72,
80
+ 73,
81
+ 74,
82
+ 75,
83
+ 76,
84
+ 77,
85
+ 78,
86
+ 79,
87
+ 80,
88
+ 81,
89
+ 82,
90
+ 83,
91
+ 84,
92
+ 85,
93
+ 86,
94
+ 87,
95
+ 88,
96
+ 89,
97
+ 90,
98
+ 91,
99
+ 92,
100
+ 93,
101
+ 94,
102
+ 95,
103
+ 96,
104
+ 97,
105
+ 98,
106
+ 99,
107
+ 100,
108
+ 101,
109
+ 102,
110
+ 103,
111
+ 104,
112
+ 105,
113
+ 106,
114
+ 107,
115
+ 108,
116
+ 109,
117
+ 110,
118
+ 111,
119
+ 112,
120
+ 113,
121
+ 114,
122
+ 115,
123
+ 116,
124
+ 117,
125
+ 118,
126
+ 119,
127
+ 120,
128
+ 121,
129
+ 122,
130
+ 123,
131
+ 124,
132
+ 125,
133
+ 126,
134
+ 127,
135
+ 128,
136
+ 129,
137
+ 130,
138
+ 131,
139
+ 132,
140
+ 133,
141
+ 134,
142
+ 135,
143
+ 136,
144
+ 137,
145
+ 138,
146
+ 139,
147
+ 140,
148
+ 141,
149
+ 142,
150
+ 143,
151
+ 144,
152
+ 145,
153
+ 146,
154
+ 147,
155
+ 148,
156
+ 149,
157
+ 150,
158
+ 151,
159
+ 152,
160
+ 153,
161
+ 154,
162
+ 155,
163
+ 156,
164
+ 157,
165
+ 158,
166
+ 159,
167
+ 160,
168
+ 161,
169
+ 162,
170
+ 163,
171
+ 164,
172
+ 165,
173
+ 166,
174
+ 167,
175
+ 168,
176
+ 169,
177
+ 170,
178
+ 171,
179
+ 172,
180
+ 173,
181
+ 174,
182
+ 175,
183
+ 176,
184
+ 177,
185
+ 178,
186
+ 179,
187
+ 180,
188
+ 181,
189
+ 182,
190
+ 183,
191
+ 184,
192
+ 185,
193
+ 186,
194
+ 187,
195
+ 188,
196
+ 189,
197
+ 190,
198
+ 191,
199
+ 192,
200
+ 193,
201
+ 194,
202
+ 195,
203
+ 196,
204
+ 197,
205
+ 198,
206
+ 199,
207
+ 200,
208
+ 201,
209
+ 202,
210
+ 203,
211
+ 204,
212
+ 205,
213
+ 206,
214
+ 207,
215
+ 208,
216
+ 209,
217
+ 210,
218
+ 211,
219
+ 212,
220
+ 213,
221
+ 214,
222
+ 215,
223
+ 216,
224
+ 217,
225
+ 218,
226
+ 219,
227
+ 220,
228
+ 221,
229
+ 222,
230
+ 223,
231
+ 224,
232
+ 225,
233
+ 226,
234
+ 227,
235
+ 228,
236
+ 229,
237
+ 230,
238
+ 231,
239
+ 232,
240
+ 233,
241
+ 234,
242
+ 235,
243
+ 236,
244
+ 237,
245
+ 238,
246
+ 239,
247
+ 240,
248
+ 241,
249
+ 242,
250
+ 243,
251
+ 244,
252
+ 245,
253
+ 246,
254
+ 247,
255
+ 248,
256
+ 249,
257
+ 250,
258
+ 251,
259
+ 252,
260
+ 253,
261
+ 254,
262
+ 255,
263
+ 256,
264
+ 257,
265
+ 258,
266
+ 259,
267
+ 260,
268
+ 261,
269
+ 262,
270
+ 263,
271
+ 264,
272
+ 265,
273
+ 266,
274
+ 267,
275
+ 268,
276
+ 269,
277
+ 270,
278
+ 271,
279
+ 272,
280
+ 273,
281
+ 274,
282
+ 275,
283
+ 276,
284
+ 277,
285
+ 278,
286
+ 279,
287
+ 280,
288
+ 281,
289
+ 282,
290
+ 283,
291
+ 284,
292
+ 285,
293
+ 286,
294
+ 287,
295
+ 288,
296
+ 289,
297
+ 290,
298
+ 291,
299
+ 292,
300
+ 293,
301
+ 294,
302
+ 295,
303
+ 296,
304
+ 297,
305
+ 298,
306
+ 299,
307
+ 300,
308
+ 301,
309
+ 302,
310
+ 303,
311
+ 304,
312
+ 305,
313
+ 306,
314
+ 307,
315
+ 308,
316
+ 309,
317
+ 310,
318
+ 311,
319
+ 312,
320
+ 313,
321
+ 314,
322
+ 315,
323
+ 316,
324
+ 317,
325
+ 318,
326
+ 319,
327
+ 320,
328
+ 321,
329
+ 322,
330
+ 323,
331
+ 324,
332
+ 325,
333
+ 326,
334
+ 327,
335
+ 328,
336
+ 329,
337
+ 330,
338
+ 331,
339
+ 332,
340
+ 333,
341
+ 334,
342
+ 335,
343
+ 336,
344
+ 337,
345
+ 338,
346
+ 339,
347
+ 340,
348
+ 341,
349
+ 342,
350
+ 343,
351
+ 344,
352
+ 345,
353
+ 346,
354
+ 347,
355
+ 348,
356
+ 349,
357
+ 350,
358
+ 351,
359
+ 352,
360
+ 353,
361
+ 354,
362
+ 355,
363
+ 356,
364
+ 357,
365
+ 358,
366
+ 359,
367
+ 360,
368
+ 361,
369
+ 362,
370
+ 363,
371
+ 364,
372
+ 365,
373
+ 366,
374
+ 367,
375
+ 368,
376
+ 369,
377
+ 370,
378
+ 371,
379
+ 372,
380
+ 373,
381
+ 374,
382
+ 375,
383
+ 376,
384
+ 377,
385
+ 378,
386
+ 379,
387
+ 380,
388
+ 381,
389
+ 382,
390
+ 383,
391
+ 384,
392
+ 385,
393
+ 386,
394
+ 387,
395
+ 388,
396
+ 389,
397
+ 390,
398
+ 391,
399
+ 392,
400
+ 393,
401
+ 394,
402
+ 395,
403
+ 396,
404
+ 397,
405
+ 398,
406
+ 399
407
+ ],
408
+ "image_transforms": {
409
+ "enable": false,
410
+ "max_num_transforms": 3,
411
+ "random_order": false,
412
+ "tfs": {
413
+ "brightness": {
414
+ "weight": 1.0,
415
+ "type": "ColorJitter",
416
+ "kwargs": {
417
+ "brightness": [
418
+ 0.8,
419
+ 1.2
420
+ ]
421
+ }
422
+ },
423
+ "contrast": {
424
+ "weight": 1.0,
425
+ "type": "ColorJitter",
426
+ "kwargs": {
427
+ "contrast": [
428
+ 0.8,
429
+ 1.2
430
+ ]
431
+ }
432
+ },
433
+ "saturation": {
434
+ "weight": 1.0,
435
+ "type": "ColorJitter",
436
+ "kwargs": {
437
+ "saturation": [
438
+ 0.5,
439
+ 1.5
440
+ ]
441
+ }
442
+ },
443
+ "hue": {
444
+ "weight": 1.0,
445
+ "type": "ColorJitter",
446
+ "kwargs": {
447
+ "hue": [
448
+ -0.05,
449
+ 0.05
450
+ ]
451
+ }
452
+ },
453
+ "sharpness": {
454
+ "weight": 1.0,
455
+ "type": "SharpnessJitter",
456
+ "kwargs": {
457
+ "sharpness": [
458
+ 0.5,
459
+ 1.5
460
+ ]
461
+ }
462
+ },
463
+ "affine": {
464
+ "weight": 1.0,
465
+ "type": "RandomAffine",
466
+ "kwargs": {
467
+ "degrees": [
468
+ -5.0,
469
+ 5.0
470
+ ],
471
+ "translate": [
472
+ 0.05,
473
+ 0.05
474
+ ]
475
+ }
476
+ }
477
+ }
478
+ },
479
+ "revision": null,
480
+ "use_imagenet_stats": true,
481
+ "video_backend": "torchcodec",
482
+ "return_uint8": false,
483
+ "depth_output_unit": "mm",
484
+ "streaming": false,
485
+ "eval_split": 0.0
486
+ },
487
+ "env": null,
488
+ "policy": {
489
+ "type": "act",
490
+ "n_obs_steps": 1,
491
+ "input_features": {
492
+ "observation.state": {
493
+ "type": "STATE",
494
+ "shape": [
495
+ 62
496
+ ]
497
+ },
498
+ "observation.images.head_left": {
499
+ "type": "VISUAL",
500
+ "shape": [
501
+ 3,
502
+ 480,
503
+ 640
504
+ ]
505
+ },
506
+ "observation.images.head_right": {
507
+ "type": "VISUAL",
508
+ "shape": [
509
+ 3,
510
+ 480,
511
+ 640
512
+ ]
513
+ },
514
+ "observation.images.wrist_left": {
515
+ "type": "VISUAL",
516
+ "shape": [
517
+ 3,
518
+ 480,
519
+ 640
520
+ ]
521
+ },
522
+ "observation.images.wrist_right": {
523
+ "type": "VISUAL",
524
+ "shape": [
525
+ 3,
526
+ 480,
527
+ 640
528
+ ]
529
+ }
530
+ },
531
+ "output_features": {
532
+ "action": {
533
+ "type": "ACTION",
534
+ "shape": [
535
+ 20
536
+ ]
537
+ }
538
+ },
539
+ "device": "cuda",
540
+ "use_amp": false,
541
+ "use_peft": false,
542
+ "push_to_hub": false,
543
+ "repo_id": null,
544
+ "private": null,
545
+ "tags": null,
546
+ "license": null,
547
+ "pretrained_path": null,
548
+ "pretrained_revision": null,
549
+ "chunk_size": 100,
550
+ "n_action_steps": 100,
551
+ "normalization_mapping": {
552
+ "VISUAL": "MEAN_STD",
553
+ "STATE": "MEAN_STD",
554
+ "ACTION": "MEAN_STD"
555
+ },
556
+ "vision_backbone": "resnet18",
557
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
558
+ "replace_final_stride_with_dilation": false,
559
+ "pre_norm": false,
560
+ "dim_model": 512,
561
+ "n_heads": 8,
562
+ "dim_feedforward": 3200,
563
+ "feedforward_activation": "relu",
564
+ "n_encoder_layers": 4,
565
+ "n_decoder_layers": 1,
566
+ "use_vae": true,
567
+ "latent_dim": 32,
568
+ "n_vae_encoder_layers": 4,
569
+ "temporal_ensemble_coeff": null,
570
+ "dropout": 0.1,
571
+ "kl_weight": 10.0,
572
+ "optimizer_lr": 1e-05,
573
+ "optimizer_weight_decay": 0.0001,
574
+ "optimizer_lr_backbone": 1e-05
575
+ },
576
+ "reward_model": null,
577
+ "output_dir": "/kaggle/working/act_foam_400_30k",
578
+ "job_name": "act",
579
+ "resume": false,
580
+ "seed": 1000,
581
+ "cudnn_deterministic": false,
582
+ "num_workers": 2,
583
+ "batch_size": 2,
584
+ "prefetch_factor": 4,
585
+ "persistent_workers": true,
586
+ "dataloader_multiprocessing_context": "spawn",
587
+ "steps": 30000,
588
+ "env_eval_freq": 20000,
589
+ "log_freq": 200,
590
+ "eval_steps": 0,
591
+ "max_eval_samples": 0,
592
+ "tolerance_s": 0.0001,
593
+ "save_checkpoint": true,
594
+ "save_freq": 2000,
595
+ "use_policy_training_preset": true,
596
+ "optimizer": {
597
+ "type": "adamw",
598
+ "lr": 1e-05,
599
+ "weight_decay": 0.0001,
600
+ "grad_clip_norm": 10.0,
601
+ "betas": [
602
+ 0.9,
603
+ 0.999
604
+ ],
605
+ "eps": 1e-08
606
+ },
607
+ "scheduler": null,
608
+ "eval": {
609
+ "n_episodes": 50,
610
+ "batch_size": 2,
611
+ "use_async_envs": true,
612
+ "recording": false,
613
+ "recording_repo_id": null,
614
+ "recording_private": false
615
+ },
616
+ "wandb": {
617
+ "enable": false,
618
+ "disable_artifact": false,
619
+ "project": "lerobot",
620
+ "entity": null,
621
+ "notes": null,
622
+ "run_id": null,
623
+ "mode": null,
624
+ "add_tags": true
625
+ },
626
+ "peft": null,
627
+ "job": {
628
+ "target": null,
629
+ "image": "huggingface/lerobot-gpu:latest",
630
+ "timeout": "2d",
631
+ "detach": false,
632
+ "tags": []
633
+ },
634
+ "save_checkpoint_to_hub": false,
635
+ "sample_weighting": null,
636
+ "rename_map": {},
637
+ "checkpoint_path": null
638
+ }
Packing_Box/v2/config.json ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "diffusion",
3
+ "n_obs_steps": 2,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 20
9
+ ]
10
+ },
11
+ "observation.images.head_left": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 480,
16
+ 640
17
+ ]
18
+ },
19
+ "observation.images.head_right": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 480,
24
+ 640
25
+ ]
26
+ },
27
+ "observation.images.wrist_left": {
28
+ "type": "VISUAL",
29
+ "shape": [
30
+ 3,
31
+ 480,
32
+ 640
33
+ ]
34
+ },
35
+ "observation.images.wrist_right": {
36
+ "type": "VISUAL",
37
+ "shape": [
38
+ 3,
39
+ 480,
40
+ 640
41
+ ]
42
+ }
43
+ },
44
+ "output_features": {
45
+ "action": {
46
+ "type": "ACTION",
47
+ "shape": [
48
+ 20
49
+ ]
50
+ }
51
+ },
52
+ "device": "cuda",
53
+ "use_amp": false,
54
+ "use_peft": false,
55
+ "push_to_hub": false,
56
+ "repo_id": null,
57
+ "private": null,
58
+ "tags": null,
59
+ "license": null,
60
+ "pretrained_path": null,
61
+ "pretrained_revision": null,
62
+ "horizon": 64,
63
+ "n_action_steps": 32,
64
+ "normalization_mapping": {
65
+ "VISUAL": "MEAN_STD",
66
+ "STATE": "MIN_MAX",
67
+ "ACTION": "MIN_MAX"
68
+ },
69
+ "drop_n_last_frames": 7,
70
+ "vision_backbone": "resnet18",
71
+ "resize_shape": null,
72
+ "crop_ratio": 1.0,
73
+ "crop_shape": null,
74
+ "crop_is_random": true,
75
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
76
+ "use_group_norm": false,
77
+ "spatial_softmax_num_keypoints": 32,
78
+ "use_separate_rgb_encoder_per_camera": true,
79
+ "down_dims": [
80
+ 512,
81
+ 1024,
82
+ 2048
83
+ ],
84
+ "kernel_size": 5,
85
+ "n_groups": 8,
86
+ "diffusion_step_embed_dim": 128,
87
+ "use_film_scale_modulation": true,
88
+ "gradient_checkpointing": false,
89
+ "noise_scheduler_type": "DDPM",
90
+ "num_train_timesteps": 100,
91
+ "beta_schedule": "squaredcos_cap_v2",
92
+ "beta_start": 0.0001,
93
+ "beta_end": 0.02,
94
+ "prediction_type": "epsilon",
95
+ "clip_sample": true,
96
+ "clip_sample_range": 1.0,
97
+ "num_inference_steps": null,
98
+ "compile_model": false,
99
+ "compile_mode": "reduce-overhead",
100
+ "do_mask_loss_for_padding": false,
101
+ "optimizer_lr": 0.0001,
102
+ "optimizer_betas": [
103
+ 0.95,
104
+ 0.999
105
+ ],
106
+ "optimizer_eps": 1e-08,
107
+ "optimizer_weight_decay": 1e-06,
108
+ "scheduler_name": "cosine",
109
+ "scheduler_warmup_steps": 500
110
+ }
Packing_Box/v2/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bce8b66a225b05fc9eb4184b3c654d9fc0d08668347bdbd31fd051a27726d8e7
3
+ size 1233869336
Packing_Box/v2/policy_postprocessor.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "policy_postprocessor",
3
+ "steps": [
4
+ {
5
+ "registry_name": "unnormalizer_processor",
6
+ "config": {
7
+ "eps": 1e-08,
8
+ "features": {
9
+ "action": {
10
+ "type": "ACTION",
11
+ "shape": [
12
+ 20
13
+ ]
14
+ }
15
+ },
16
+ "norm_map": {
17
+ "VISUAL": "MEAN_STD",
18
+ "STATE": "MIN_MAX",
19
+ "ACTION": "MIN_MAX"
20
+ }
21
+ },
22
+ "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors"
23
+ },
24
+ {
25
+ "registry_name": "device_processor",
26
+ "config": {
27
+ "device": "cpu",
28
+ "float_dtype": null
29
+ }
30
+ }
31
+ ]
32
+ }
Packing_Box/v2/policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7aab13c0aa66139b734c51287458a364eb4c70e2f0027f17fb9b52e4f1770bd
3
+ size 10880
Packing_Box/v2/policy_preprocessor.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "policy_preprocessor",
3
+ "steps": [
4
+ {
5
+ "registry_name": "rename_observations_processor",
6
+ "config": {
7
+ "rename_map": {}
8
+ }
9
+ },
10
+ {
11
+ "registry_name": "to_batch_processor",
12
+ "config": {}
13
+ },
14
+ {
15
+ "registry_name": "device_processor",
16
+ "config": {
17
+ "device": "cuda",
18
+ "float_dtype": null
19
+ }
20
+ },
21
+ {
22
+ "registry_name": "normalizer_processor",
23
+ "config": {
24
+ "eps": 1e-08,
25
+ "features": {
26
+ "observation.state": {
27
+ "type": "STATE",
28
+ "shape": [
29
+ 20
30
+ ]
31
+ },
32
+ "observation.images.head_left": {
33
+ "type": "VISUAL",
34
+ "shape": [
35
+ 3,
36
+ 480,
37
+ 640
38
+ ]
39
+ },
40
+ "observation.images.head_right": {
41
+ "type": "VISUAL",
42
+ "shape": [
43
+ 3,
44
+ 480,
45
+ 640
46
+ ]
47
+ },
48
+ "observation.images.wrist_left": {
49
+ "type": "VISUAL",
50
+ "shape": [
51
+ 3,
52
+ 480,
53
+ 640
54
+ ]
55
+ },
56
+ "observation.images.wrist_right": {
57
+ "type": "VISUAL",
58
+ "shape": [
59
+ 3,
60
+ 480,
61
+ 640
62
+ ]
63
+ },
64
+ "action": {
65
+ "type": "ACTION",
66
+ "shape": [
67
+ 20
68
+ ]
69
+ }
70
+ },
71
+ "norm_map": {
72
+ "VISUAL": "MEAN_STD",
73
+ "STATE": "MIN_MAX",
74
+ "ACTION": "MIN_MAX"
75
+ }
76
+ },
77
+ "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors"
78
+ }
79
+ ]
80
+ }
Packing_Box/v2/policy_preprocessor_step_3_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4721bff7322b17bbd62dd3a04951054cbcf998dbe6c6f5e025375c26f68b3488
3
+ size 10896
Packing_Box/v2/train_config.json ADDED
@@ -0,0 +1,311 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "local_dataset",
4
+ "repo_type": "dataset",
5
+ "root": "/root/.cache/huggingface/hub/datasets--UBTECH-Robotics--challenge2026_dataset/snapshots/33df931d1f27951a39a3a4d1280080620f932ae6/Packing_Box",
6
+ "episodes": null,
7
+ "exclude_episodes": null,
8
+ "image_transforms": {
9
+ "enable": false,
10
+ "max_num_transforms": 3,
11
+ "random_order": false,
12
+ "tfs": {
13
+ "brightness": {
14
+ "weight": 1.0,
15
+ "type": "ColorJitter",
16
+ "kwargs": {
17
+ "brightness": [
18
+ 0.8,
19
+ 1.2
20
+ ]
21
+ }
22
+ },
23
+ "contrast": {
24
+ "weight": 1.0,
25
+ "type": "ColorJitter",
26
+ "kwargs": {
27
+ "contrast": [
28
+ 0.8,
29
+ 1.2
30
+ ]
31
+ }
32
+ },
33
+ "saturation": {
34
+ "weight": 1.0,
35
+ "type": "ColorJitter",
36
+ "kwargs": {
37
+ "saturation": [
38
+ 0.5,
39
+ 1.5
40
+ ]
41
+ }
42
+ },
43
+ "hue": {
44
+ "weight": 1.0,
45
+ "type": "ColorJitter",
46
+ "kwargs": {
47
+ "hue": [
48
+ -0.05,
49
+ 0.05
50
+ ]
51
+ }
52
+ },
53
+ "sharpness": {
54
+ "weight": 1.0,
55
+ "type": "SharpnessJitter",
56
+ "kwargs": {
57
+ "sharpness": [
58
+ 0.5,
59
+ 1.5
60
+ ]
61
+ }
62
+ },
63
+ "affine": {
64
+ "weight": 1.0,
65
+ "type": "RandomAffine",
66
+ "kwargs": {
67
+ "degrees": [
68
+ -5.0,
69
+ 5.0
70
+ ],
71
+ "translate": [
72
+ 0.05,
73
+ 0.05
74
+ ]
75
+ }
76
+ }
77
+ }
78
+ },
79
+ "revision": null,
80
+ "use_imagenet_stats": true,
81
+ "video_backend": "torchcodec",
82
+ "return_uint8": false,
83
+ "depth_output_unit": "mm",
84
+ "streaming": false,
85
+ "eval_split": 0.05
86
+ },
87
+ "env": null,
88
+ "policy": {
89
+ "type": "diffusion",
90
+ "n_obs_steps": 2,
91
+ "input_features": {
92
+ "observation.state": {
93
+ "type": "STATE",
94
+ "shape": [
95
+ 20
96
+ ]
97
+ },
98
+ "observation.images.head_left": {
99
+ "type": "VISUAL",
100
+ "shape": [
101
+ 3,
102
+ 480,
103
+ 640
104
+ ]
105
+ },
106
+ "observation.images.head_right": {
107
+ "type": "VISUAL",
108
+ "shape": [
109
+ 3,
110
+ 480,
111
+ 640
112
+ ]
113
+ },
114
+ "observation.images.wrist_left": {
115
+ "type": "VISUAL",
116
+ "shape": [
117
+ 3,
118
+ 480,
119
+ 640
120
+ ]
121
+ },
122
+ "observation.images.wrist_right": {
123
+ "type": "VISUAL",
124
+ "shape": [
125
+ 3,
126
+ 480,
127
+ 640
128
+ ]
129
+ }
130
+ },
131
+ "output_features": {
132
+ "action": {
133
+ "type": "ACTION",
134
+ "shape": [
135
+ 20
136
+ ]
137
+ }
138
+ },
139
+ "device": "cuda",
140
+ "use_amp": false,
141
+ "use_peft": false,
142
+ "push_to_hub": false,
143
+ "repo_id": null,
144
+ "private": null,
145
+ "tags": null,
146
+ "license": null,
147
+ "pretrained_path": null,
148
+ "pretrained_revision": null,
149
+ "horizon": 64,
150
+ "n_action_steps": 32,
151
+ "normalization_mapping": {
152
+ "VISUAL": "MEAN_STD",
153
+ "STATE": "MIN_MAX",
154
+ "ACTION": "MIN_MAX"
155
+ },
156
+ "drop_n_last_frames": 7,
157
+ "vision_backbone": "resnet18",
158
+ "resize_shape": null,
159
+ "crop_ratio": 1.0,
160
+ "crop_shape": null,
161
+ "crop_is_random": true,
162
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
163
+ "use_group_norm": false,
164
+ "spatial_softmax_num_keypoints": 32,
165
+ "use_separate_rgb_encoder_per_camera": true,
166
+ "down_dims": [
167
+ 512,
168
+ 1024,
169
+ 2048
170
+ ],
171
+ "kernel_size": 5,
172
+ "n_groups": 8,
173
+ "diffusion_step_embed_dim": 128,
174
+ "use_film_scale_modulation": true,
175
+ "gradient_checkpointing": false,
176
+ "noise_scheduler_type": "DDPM",
177
+ "num_train_timesteps": 100,
178
+ "beta_schedule": "squaredcos_cap_v2",
179
+ "beta_start": 0.0001,
180
+ "beta_end": 0.02,
181
+ "prediction_type": "epsilon",
182
+ "clip_sample": true,
183
+ "clip_sample_range": 1.0,
184
+ "num_inference_steps": null,
185
+ "compile_model": false,
186
+ "compile_mode": "reduce-overhead",
187
+ "do_mask_loss_for_padding": false,
188
+ "optimizer_lr": 0.0001,
189
+ "optimizer_betas": [
190
+ 0.95,
191
+ 0.999
192
+ ],
193
+ "optimizer_eps": 1e-08,
194
+ "optimizer_weight_decay": 1e-06,
195
+ "scheduler_name": "cosine",
196
+ "scheduler_warmup_steps": 500
197
+ },
198
+ "reward_model": null,
199
+ "output_dir": "/kaggle/working/diffusion_policy_task4",
200
+ "job_name": "diffusion",
201
+ "resume": false,
202
+ "seed": 1000,
203
+ "cudnn_deterministic": false,
204
+ "num_workers": 2,
205
+ "batch_size": 4,
206
+ "prefetch_factor": 4,
207
+ "persistent_workers": true,
208
+ "dataloader_multiprocessing_context": "spawn",
209
+ "steps": 15000,
210
+ "env_eval_freq": 20000,
211
+ "log_freq": 50,
212
+ "eval_steps": 0,
213
+ "max_eval_samples": 0,
214
+ "tolerance_s": 0.0001,
215
+ "save_checkpoint": true,
216
+ "save_freq": 15000,
217
+ "checkpoint_format": "safetensors",
218
+ "use_policy_training_preset": true,
219
+ "optimizer": {
220
+ "type": "adam",
221
+ "lr": 0.0001,
222
+ "weight_decay": 1e-06,
223
+ "grad_clip_norm": 10.0,
224
+ "betas": [
225
+ 0.95,
226
+ 0.999
227
+ ],
228
+ "eps": 1e-08
229
+ },
230
+ "scheduler": {
231
+ "type": "diffuser",
232
+ "num_warmup_steps": 500,
233
+ "name": "cosine"
234
+ },
235
+ "parallelism": {
236
+ "dp_replicate": 1,
237
+ "dp_shard": 1,
238
+ "context_parallel": {
239
+ "ring_degree": 1,
240
+ "ulysses_degree": 1
241
+ },
242
+ "cfg_parallel": 1
243
+ },
244
+ "accelerator": {
245
+ "mixed_precision": "no",
246
+ "gradient_accumulation": {
247
+ "steps": 1
248
+ },
249
+ "fsdp": {
250
+ "reshard_after_forward": true,
251
+ "wrap_modules": null,
252
+ "min_num_params": null,
253
+ "cpu_offload": false,
254
+ "ignored_modules": null
255
+ },
256
+ "ddp": {
257
+ "find_unused_parameters": true,
258
+ "gradient_as_bucket_view": false,
259
+ "static_graph": false
260
+ },
261
+ "compile": {
262
+ "enabled": false,
263
+ "backend": "inductor",
264
+ "mode": null,
265
+ "regional": true
266
+ },
267
+ "activation_checkpointing": {
268
+ "mode": "none"
269
+ }
270
+ },
271
+ "eval": {
272
+ "n_episodes": 50,
273
+ "batch_size": 2,
274
+ "use_async_envs": true,
275
+ "recording": false,
276
+ "recording_repo_id": null,
277
+ "recording_private": false
278
+ },
279
+ "ema": {
280
+ "enable": false,
281
+ "decay": null,
282
+ "update_after_step": 0,
283
+ "inv_gamma": 1.0,
284
+ "power": 0.75,
285
+ "min_decay": 0.0,
286
+ "max_decay": 0.9999,
287
+ "use_for_eval": true
288
+ },
289
+ "wandb": {
290
+ "enable": false,
291
+ "disable_artifact": false,
292
+ "project": "lerobot",
293
+ "entity": null,
294
+ "notes": null,
295
+ "run_id": null,
296
+ "mode": null,
297
+ "add_tags": true
298
+ },
299
+ "peft": null,
300
+ "job": {
301
+ "target": null,
302
+ "image": "huggingface/lerobot-gpu:latest",
303
+ "timeout": "2d",
304
+ "detach": false,
305
+ "tags": []
306
+ },
307
+ "save_checkpoint_to_hub": false,
308
+ "sample_weighting": null,
309
+ "rename_map": {},
310
+ "checkpoint_path": null
311
+ }