Hugging Face's logo

enyolanev-bit
/
diffusion-pusht-sample-efficient

Robotics
LeRobot
Safetensors
imitation-learning
diffusion-policy
pusht
sample-efficiency
Model card Files
xet
 Community
enyolanev-bit commited on
Commit
5c6e074
·
verified ·
1 Parent(s): 5b9b136

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. enhanced-200demos/config.json +84 -0
  2. enhanced-200demos/model.safetensors +3 -0
  3. enhanced-200demos/policy_postprocessor.json +32 -0
  4. enhanced-200demos/policy_postprocessor_step_0_unnormalizer_processor.safetensors +3 -0
  5. enhanced-200demos/policy_preprocessor.json +56 -0
  6. enhanced-200demos/policy_preprocessor_step_3_normalizer_processor.safetensors +3 -0
  7. enhanced-200demos/train_config.json +462 -0
enhanced-200demos/config.json ADDED
@@ -0,0 +1,84 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "diffusion",
3
+ "n_obs_steps": 2,
4
+ "input_features": {
5
+ "observation.image": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 96,
10
+ 96
11
+ ]
12
+ },
13
+ "observation.state": {
14
+ "type": "STATE",
15
+ "shape": [
16
+ 2
17
+ ]
18
+ }
19
+ },
20
+ "output_features": {
21
+ "action": {
22
+ "type": "ACTION",
23
+ "shape": [
24
+ 2
25
+ ]
26
+ }
27
+ },
28
+ "device": "cuda",
29
+ "use_amp": false,
30
+ "use_peft": false,
31
+ "push_to_hub": false,
32
+ "repo_id": null,
33
+ "private": null,
34
+ "tags": null,
35
+ "license": null,
36
+ "pretrained_path": null,
37
+ "horizon": 16,
38
+ "n_action_steps": 8,
39
+ "normalization_mapping": {
40
+ "VISUAL": "MEAN_STD",
41
+ "STATE": "MIN_MAX",
42
+ "ACTION": "MIN_MAX"
43
+ },
44
+ "drop_n_last_frames": 7,
45
+ "vision_backbone": "resnet18",
46
+ "resize_shape": null,
47
+ "crop_ratio": 1.0,
48
+ "crop_shape": null,
49
+ "crop_is_random": true,
50
+ "pretrained_backbone_weights": null,
51
+ "use_group_norm": true,
52
+ "spatial_softmax_num_keypoints": 32,
53
+ "use_separate_rgb_encoder_per_camera": false,
54
+ "down_dims": [
55
+ 512,
56
+ 1024,
57
+ 2048
58
+ ],
59
+ "kernel_size": 5,
60
+ "n_groups": 8,
61
+ "diffusion_step_embed_dim": 128,
62
+ "use_film_scale_modulation": true,
63
+ "noise_scheduler_type": "DDPM",
64
+ "num_train_timesteps": 100,
65
+ "beta_schedule": "squaredcos_cap_v2",
66
+ "beta_start": 0.0001,
67
+ "beta_end": 0.02,
68
+ "prediction_type": "epsilon",
69
+ "clip_sample": true,
70
+ "clip_sample_range": 1.0,
71
+ "num_inference_steps": null,
72
+ "compile_model": false,
73
+ "compile_mode": "reduce-overhead",
74
+ "do_mask_loss_for_padding": false,
75
+ "optimizer_lr": 0.0001,
76
+ "optimizer_betas": [
77
+ 0.95,
78
+ 0.999
79
+ ],
80
+ "optimizer_eps": 1e-08,
81
+ "optimizer_weight_decay": 1e-06,
82
+ "scheduler_name": "cosine",
83
+ "scheduler_warmup_steps": 500
84
+ }
enhanced-200demos/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9daa0cb151a1fadfca2b88072e6af3538fc95d6ea7de77ef0716a435520550c5
3
+ size 1050861448
enhanced-200demos/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
+ 2
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
+ }
enhanced-200demos/policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc174a7436ff5075efebe7be7cece5e5d8430658cc9b5314fb8a652eab912d96
3
+ size 4204
enhanced-200demos/policy_preprocessor.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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.image": {
27
+ "type": "VISUAL",
28
+ "shape": [
29
+ 3,
30
+ 96,
31
+ 96
32
+ ]
33
+ },
34
+ "observation.state": {
35
+ "type": "STATE",
36
+ "shape": [
37
+ 2
38
+ ]
39
+ },
40
+ "action": {
41
+ "type": "ACTION",
42
+ "shape": [
43
+ 2
44
+ ]
45
+ }
46
+ },
47
+ "norm_map": {
48
+ "VISUAL": "MEAN_STD",
49
+ "STATE": "MIN_MAX",
50
+ "ACTION": "MIN_MAX"
51
+ }
52
+ },
53
+ "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors"
54
+ }
55
+ ]
56
+ }
enhanced-200demos/policy_preprocessor_step_3_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc174a7436ff5075efebe7be7cece5e5d8430658cc9b5314fb8a652eab912d96
3
+ size 4204
enhanced-200demos/train_config.json ADDED
@@ -0,0 +1,462 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "lerobot/pusht",
4
+ "root": null,
5
+ "episodes": [
6
+ 0,
7
+ 1,
8
+ 2,
9
+ 3,
10
+ 4,
11
+ 5,
12
+ 6,
13
+ 7,
14
+ 8,
15
+ 9,
16
+ 10,
17
+ 11,
18
+ 12,
19
+ 13,
20
+ 14,
21
+ 15,
22
+ 16,
23
+ 17,
24
+ 18,
25
+ 19,
26
+ 20,
27
+ 21,
28
+ 22,
29
+ 23,
30
+ 24,
31
+ 25,
32
+ 26,
33
+ 27,
34
+ 28,
35
+ 29,
36
+ 30,
37
+ 31,
38
+ 32,
39
+ 33,
40
+ 34,
41
+ 35,
42
+ 36,
43
+ 37,
44
+ 38,
45
+ 39,
46
+ 40,
47
+ 41,
48
+ 42,
49
+ 43,
50
+ 44,
51
+ 45,
52
+ 46,
53
+ 47,
54
+ 48,
55
+ 49,
56
+ 50,
57
+ 51,
58
+ 52,
59
+ 53,
60
+ 54,
61
+ 55,
62
+ 56,
63
+ 57,
64
+ 58,
65
+ 59,
66
+ 60,
67
+ 61,
68
+ 62,
69
+ 63,
70
+ 64,
71
+ 65,
72
+ 66,
73
+ 67,
74
+ 68,
75
+ 69,
76
+ 70,
77
+ 71,
78
+ 72,
79
+ 73,
80
+ 74,
81
+ 75,
82
+ 76,
83
+ 77,
84
+ 78,
85
+ 79,
86
+ 80,
87
+ 81,
88
+ 82,
89
+ 83,
90
+ 84,
91
+ 85,
92
+ 86,
93
+ 87,
94
+ 88,
95
+ 89,
96
+ 90,
97
+ 91,
98
+ 92,
99
+ 93,
100
+ 94,
101
+ 95,
102
+ 96,
103
+ 97,
104
+ 98,
105
+ 99,
106
+ 100,
107
+ 101,
108
+ 102,
109
+ 103,
110
+ 104,
111
+ 105,
112
+ 106,
113
+ 107,
114
+ 108,
115
+ 109,
116
+ 110,
117
+ 111,
118
+ 112,
119
+ 113,
120
+ 114,
121
+ 115,
122
+ 116,
123
+ 117,
124
+ 118,
125
+ 119,
126
+ 120,
127
+ 121,
128
+ 122,
129
+ 123,
130
+ 124,
131
+ 125,
132
+ 126,
133
+ 127,
134
+ 128,
135
+ 129,
136
+ 130,
137
+ 131,
138
+ 132,
139
+ 133,
140
+ 134,
141
+ 135,
142
+ 136,
143
+ 137,
144
+ 138,
145
+ 139,
146
+ 140,
147
+ 141,
148
+ 142,
149
+ 143,
150
+ 144,
151
+ 145,
152
+ 146,
153
+ 147,
154
+ 148,
155
+ 149,
156
+ 150,
157
+ 151,
158
+ 152,
159
+ 153,
160
+ 154,
161
+ 155,
162
+ 156,
163
+ 157,
164
+ 158,
165
+ 159,
166
+ 160,
167
+ 161,
168
+ 162,
169
+ 163,
170
+ 164,
171
+ 165,
172
+ 166,
173
+ 167,
174
+ 168,
175
+ 169,
176
+ 170,
177
+ 171,
178
+ 172,
179
+ 173,
180
+ 174,
181
+ 175,
182
+ 176,
183
+ 177,
184
+ 178,
185
+ 179,
186
+ 180,
187
+ 181,
188
+ 182,
189
+ 183,
190
+ 184,
191
+ 185,
192
+ 186,
193
+ 187,
194
+ 188,
195
+ 189,
196
+ 190,
197
+ 191,
198
+ 192,
199
+ 193,
200
+ 194,
201
+ 195,
202
+ 196,
203
+ 197,
204
+ 198,
205
+ 199
206
+ ],
207
+ "image_transforms": {
208
+ "enable": true,
209
+ "max_num_transforms": 3,
210
+ "random_order": false,
211
+ "tfs": {
212
+ "brightness": {
213
+ "weight": 1.0,
214
+ "type": "ColorJitter",
215
+ "kwargs": {
216
+ "brightness": [
217
+ 0.8,
218
+ 1.2
219
+ ]
220
+ }
221
+ },
222
+ "contrast": {
223
+ "weight": 1.0,
224
+ "type": "ColorJitter",
225
+ "kwargs": {
226
+ "contrast": [
227
+ 0.8,
228
+ 1.2
229
+ ]
230
+ }
231
+ },
232
+ "saturation": {
233
+ "weight": 1.0,
234
+ "type": "ColorJitter",
235
+ "kwargs": {
236
+ "saturation": [
237
+ 0.5,
238
+ 1.5
239
+ ]
240
+ }
241
+ },
242
+ "hue": {
243
+ "weight": 1.0,
244
+ "type": "ColorJitter",
245
+ "kwargs": {
246
+ "hue": [
247
+ -0.05,
248
+ 0.05
249
+ ]
250
+ }
251
+ },
252
+ "sharpness": {
253
+ "weight": 1.0,
254
+ "type": "SharpnessJitter",
255
+ "kwargs": {
256
+ "sharpness": [
257
+ 0.5,
258
+ 1.5
259
+ ]
260
+ }
261
+ },
262
+ "affine": {
263
+ "weight": 1.0,
264
+ "type": "RandomAffine",
265
+ "kwargs": {
266
+ "degrees": [
267
+ -5.0,
268
+ 5.0
269
+ ],
270
+ "translate": [
271
+ 0.05,
272
+ 0.05
273
+ ]
274
+ }
275
+ }
276
+ }
277
+ },
278
+ "revision": null,
279
+ "use_imagenet_stats": true,
280
+ "video_backend": "pyav",
281
+ "streaming": false
282
+ },
283
+ "env": {
284
+ "type": "pusht",
285
+ "task": "PushT-v0",
286
+ "fps": 10,
287
+ "features": {
288
+ "action": {
289
+ "type": "ACTION",
290
+ "shape": [
291
+ 2
292
+ ]
293
+ },
294
+ "agent_pos": {
295
+ "type": "STATE",
296
+ "shape": [
297
+ 2
298
+ ]
299
+ },
300
+ "pixels": {
301
+ "type": "VISUAL",
302
+ "shape": [
303
+ 384,
304
+ 384,
305
+ 3
306
+ ]
307
+ }
308
+ },
309
+ "features_map": {
310
+ "action": "action",
311
+ "agent_pos": "observation.state",
312
+ "environment_state": "observation.environment_state",
313
+ "pixels": "observation.image"
314
+ },
315
+ "max_parallel_tasks": 1,
316
+ "disable_env_checker": true,
317
+ "episode_length": 300,
318
+ "obs_type": "pixels_agent_pos",
319
+ "render_mode": "rgb_array",
320
+ "visualization_width": 384,
321
+ "visualization_height": 384,
322
+ "observation_height": 384,
323
+ "observation_width": 384
324
+ },
325
+ "policy": {
326
+ "type": "diffusion",
327
+ "n_obs_steps": 2,
328
+ "input_features": {
329
+ "observation.image": {
330
+ "type": "VISUAL",
331
+ "shape": [
332
+ 3,
333
+ 96,
334
+ 96
335
+ ]
336
+ },
337
+ "observation.state": {
338
+ "type": "STATE",
339
+ "shape": [
340
+ 2
341
+ ]
342
+ }
343
+ },
344
+ "output_features": {
345
+ "action": {
346
+ "type": "ACTION",
347
+ "shape": [
348
+ 2
349
+ ]
350
+ }
351
+ },
352
+ "device": "cuda",
353
+ "use_amp": false,
354
+ "use_peft": false,
355
+ "push_to_hub": false,
356
+ "repo_id": null,
357
+ "private": null,
358
+ "tags": null,
359
+ "license": null,
360
+ "pretrained_path": null,
361
+ "horizon": 16,
362
+ "n_action_steps": 8,
363
+ "normalization_mapping": {
364
+ "VISUAL": "MEAN_STD",
365
+ "STATE": "MIN_MAX",
366
+ "ACTION": "MIN_MAX"
367
+ },
368
+ "drop_n_last_frames": 7,
369
+ "vision_backbone": "resnet18",
370
+ "resize_shape": null,
371
+ "crop_ratio": 1.0,
372
+ "crop_shape": null,
373
+ "crop_is_random": true,
374
+ "pretrained_backbone_weights": null,
375
+ "use_group_norm": true,
376
+ "spatial_softmax_num_keypoints": 32,
377
+ "use_separate_rgb_encoder_per_camera": false,
378
+ "down_dims": [
379
+ 512,
380
+ 1024,
381
+ 2048
382
+ ],
383
+ "kernel_size": 5,
384
+ "n_groups": 8,
385
+ "diffusion_step_embed_dim": 128,
386
+ "use_film_scale_modulation": true,
387
+ "noise_scheduler_type": "DDPM",
388
+ "num_train_timesteps": 100,
389
+ "beta_schedule": "squaredcos_cap_v2",
390
+ "beta_start": 0.0001,
391
+ "beta_end": 0.02,
392
+ "prediction_type": "epsilon",
393
+ "clip_sample": true,
394
+ "clip_sample_range": 1.0,
395
+ "num_inference_steps": null,
396
+ "compile_model": false,
397
+ "compile_mode": "reduce-overhead",
398
+ "do_mask_loss_for_padding": false,
399
+ "optimizer_lr": 0.0001,
400
+ "optimizer_betas": [
401
+ 0.95,
402
+ 0.999
403
+ ],
404
+ "optimizer_eps": 1e-08,
405
+ "optimizer_weight_decay": 1e-06,
406
+ "scheduler_name": "cosine",
407
+ "scheduler_warmup_steps": 500
408
+ },
409
+ "output_dir": "/home/jovyan/sample-efficient-imitation/runs/de_diffusion_enhanced_n200_s0",
410
+ "job_name": "de_diffusion_enhanced_n200_s0",
411
+ "resume": false,
412
+ "seed": 0,
413
+ "cudnn_deterministic": false,
414
+ "num_workers": 4,
415
+ "batch_size": 8,
416
+ "steps": 150000,
417
+ "eval_freq": 150000,
418
+ "log_freq": 200,
419
+ "tolerance_s": 0.0001,
420
+ "save_checkpoint": true,
421
+ "save_freq": 20000,
422
+ "use_policy_training_preset": true,
423
+ "optimizer": {
424
+ "type": "adam",
425
+ "lr": 0.0001,
426
+ "weight_decay": 1e-06,
427
+ "grad_clip_norm": 10.0,
428
+ "betas": [
429
+ 0.95,
430
+ 0.999
431
+ ],
432
+ "eps": 1e-08
433
+ },
434
+ "scheduler": {
435
+ "type": "diffuser",
436
+ "num_warmup_steps": 500,
437
+ "name": "cosine"
438
+ },
439
+ "eval": {
440
+ "n_episodes": 100,
441
+ "batch_size": 10,
442
+ "use_async_envs": false
443
+ },
444
+ "wandb": {
445
+ "enable": false,
446
+ "disable_artifact": false,
447
+ "project": "lerobot",
448
+ "entity": null,
449
+ "notes": null,
450
+ "run_id": null,
451
+ "mode": null,
452
+ "add_tags": true
453
+ },
454
+ "peft": null,
455
+ "use_rabc": false,
456
+ "rabc_progress_path": null,
457
+ "rabc_kappa": 0.01,
458
+ "rabc_epsilon": 1e-06,
459
+ "rabc_head_mode": "sparse",
460
+ "rename_map": {},
461
+ "checkpoint_path": null
462
+ }