Hugging Face's logo

chomeed
/
peg_insertion_full_rabc_delta2_using_rewind

Robotics
LeRobot
Safetensors
rabc_diffusion
Model card Files
xet
 Community
chomeed commited on
Commit
08a4af3
·
verified ·
1 Parent(s): 4e59557

Upload policy weights, train config and readme

Browse files
Files changed (4) hide show
  1. README.md +62 -0
  2. config.json +91 -0
  3. model.safetensors +3 -0
  4. train_config.json +391 -0
README.md ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ datasets: chomeed/peg_insertion_full_rabc_delta2_using_rewind
3
+ library_name: lerobot
4
+ license: apache-2.0
5
+ model_name: rabc_diffusion
6
+ pipeline_tag: robotics
7
+ tags:
8
+ - lerobot
9
+ - robotics
10
+ - rabc_diffusion
11
+ ---
12
+
13
+ # Model Card for rabc_diffusion
14
+
15
+ <!-- Provide a quick summary of what the model is/does. -->
16
+
17
+
18
+ _Model type not recognized — please update this template._
19
+
20
+
21
+ This policy has been trained and pushed to the Hub using [LeRobot](https://github.com/huggingface/lerobot).
22
+ See the full documentation at [LeRobot Docs](https://huggingface.co/docs/lerobot/index).
23
+
24
+ ---
25
+
26
+ ## How to Get Started with the Model
27
+
28
+ For a complete walkthrough, see the [training guide](https://huggingface.co/docs/lerobot/il_robots#train-a-policy).
29
+ Below is the short version on how to train and run inference/eval:
30
+
31
+ ### Train from scratch
32
+
33
+ ```bash
34
+ lerobot-train \
35
+ --dataset.repo_id=${HF_USER}/<dataset> \
36
+ --policy.type=act \
37
+ --output_dir=outputs/train/<desired_policy_repo_id> \
38
+ --job_name=lerobot_training \
39
+ --policy.device=cuda \
40
+ --policy.repo_id=${HF_USER}/<desired_policy_repo_id>
41
+ --wandb.enable=true
42
+ ```
43
+
44
+ _Writes checkpoints to `outputs/train/<desired_policy_repo_id>/checkpoints/`._
45
+
46
+ ### Evaluate the policy/run inference
47
+
48
+ ```bash
49
+ lerobot-record \
50
+ --robot.type=so100_follower \
51
+ --dataset.repo_id=<hf_user>/eval_<dataset> \
52
+ --policy.path=<hf_user>/<desired_policy_repo_id> \
53
+ --episodes=10
54
+ ```
55
+
56
+ Prefix the dataset repo with **eval\_** and supply `--policy.path` pointing to a local or hub checkpoint.
57
+
58
+ ---
59
+
60
+ ## Model Details
61
+
62
+ - **License:** apache-2.0
config.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "rabc_diffusion",
3
+ "n_obs_steps": 2,
4
+ "input_features": {
5
+ "observation.images.front_rgb": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 256,
10
+ 256
11
+ ]
12
+ },
13
+ "observation.images.wrist_rgb": {
14
+ "type": "VISUAL",
15
+ "shape": [
16
+ 3,
17
+ 256,
18
+ 256
19
+ ]
20
+ },
21
+ "observation.state": {
22
+ "type": "STATE",
23
+ "shape": [
24
+ 19
25
+ ]
26
+ }
27
+ },
28
+ "output_features": {
29
+ "action": {
30
+ "type": "ACTION",
31
+ "shape": [
32
+ 6
33
+ ]
34
+ }
35
+ },
36
+ "device": "cuda",
37
+ "use_amp": false,
38
+ "use_peft": false,
39
+ "push_to_hub": true,
40
+ "repo_id": "chomeed/peg_insertion_full_rabc_delta2_using_rewind",
41
+ "private": null,
42
+ "tags": null,
43
+ "license": null,
44
+ "pretrained_path": null,
45
+ "horizon": 2,
46
+ "n_action_steps": 2,
47
+ "normalization_mapping": {
48
+ "VISUAL": "MEAN_STD",
49
+ "STATE": "MIN_MAX",
50
+ "ACTION": "MIN_MAX"
51
+ },
52
+ "drop_n_last_frames": 7,
53
+ "vision_backbone": "resnet18",
54
+ "crop_shape": [
55
+ 256,
56
+ 256
57
+ ],
58
+ "crop_is_random": true,
59
+ "pretrained_backbone_weights": null,
60
+ "use_group_norm": true,
61
+ "spatial_softmax_num_keypoints": 32,
62
+ "use_separate_rgb_encoder_per_camera": false,
63
+ "down_dims": [
64
+ 512
65
+ ],
66
+ "kernel_size": 5,
67
+ "n_groups": 8,
68
+ "diffusion_step_embed_dim": 128,
69
+ "use_film_scale_modulation": true,
70
+ "noise_scheduler_type": "DDPM",
71
+ "num_train_timesteps": 100,
72
+ "beta_schedule": "squaredcos_cap_v2",
73
+ "beta_start": 0.0001,
74
+ "beta_end": 0.02,
75
+ "prediction_type": "epsilon",
76
+ "clip_sample": true,
77
+ "clip_sample_range": 1.0,
78
+ "num_inference_steps": null,
79
+ "do_mask_loss_for_padding": false,
80
+ "optimizer_lr": 0.0001,
81
+ "optimizer_betas": [
82
+ 0.95,
83
+ 0.999
84
+ ],
85
+ "optimizer_eps": 1e-08,
86
+ "optimizer_weight_decay": 1e-06,
87
+ "scheduler_name": "cosine",
88
+ "scheduler_warmup_steps": 500,
89
+ "use_rabc_weighting": true,
90
+ "weight_key": "rabc.weight"
91
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71385d861b7e1950635cd34fa33416b83f50eb29c7f0bec1b5f82b6bb212591a
3
+ size 94346048
train_config.json ADDED
@@ -0,0 +1,391 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "chomeed/peg_insertion_full_rabc_delta2_using_rewind",
4
+ "root": null,
5
+ "episodes": null,
6
+ "image_transforms": {
7
+ "enable": false,
8
+ "max_num_transforms": 3,
9
+ "random_order": false,
10
+ "tfs": {
11
+ "brightness": {
12
+ "weight": 1.0,
13
+ "type": "ColorJitter",
14
+ "kwargs": {
15
+ "brightness": [
16
+ 0.8,
17
+ 1.2
18
+ ]
19
+ }
20
+ },
21
+ "contrast": {
22
+ "weight": 1.0,
23
+ "type": "ColorJitter",
24
+ "kwargs": {
25
+ "contrast": [
26
+ 0.8,
27
+ 1.2
28
+ ]
29
+ }
30
+ },
31
+ "saturation": {
32
+ "weight": 1.0,
33
+ "type": "ColorJitter",
34
+ "kwargs": {
35
+ "saturation": [
36
+ 0.5,
37
+ 1.5
38
+ ]
39
+ }
40
+ },
41
+ "hue": {
42
+ "weight": 1.0,
43
+ "type": "ColorJitter",
44
+ "kwargs": {
45
+ "hue": [
46
+ -0.05,
47
+ 0.05
48
+ ]
49
+ }
50
+ },
51
+ "sharpness": {
52
+ "weight": 1.0,
53
+ "type": "SharpnessJitter",
54
+ "kwargs": {
55
+ "sharpness": [
56
+ 0.5,
57
+ 1.5
58
+ ]
59
+ }
60
+ },
61
+ "affine": {
62
+ "weight": 1.0,
63
+ "type": "RandomAffine",
64
+ "kwargs": {
65
+ "degrees": [
66
+ -5.0,
67
+ 5.0
68
+ ],
69
+ "translate": [
70
+ 0.05,
71
+ 0.05
72
+ ]
73
+ }
74
+ }
75
+ }
76
+ },
77
+ "revision": null,
78
+ "use_imagenet_stats": true,
79
+ "video_backend": "torchcodec",
80
+ "streaming": false,
81
+ "cf_aug": false
82
+ },
83
+ "env": null,
84
+ "policy": {
85
+ "type": "rabc_diffusion",
86
+ "n_obs_steps": 2,
87
+ "input_features": {
88
+ "observation.images.front_rgb": {
89
+ "type": "VISUAL",
90
+ "shape": [
91
+ 3,
92
+ 256,
93
+ 256
94
+ ]
95
+ },
96
+ "observation.images.wrist_rgb": {
97
+ "type": "VISUAL",
98
+ "shape": [
99
+ 3,
100
+ 256,
101
+ 256
102
+ ]
103
+ },
104
+ "observation.state": {
105
+ "type": "STATE",
106
+ "shape": [
107
+ 19
108
+ ]
109
+ }
110
+ },
111
+ "output_features": {
112
+ "action": {
113
+ "type": "ACTION",
114
+ "shape": [
115
+ 6
116
+ ]
117
+ }
118
+ },
119
+ "device": "cuda",
120
+ "use_amp": false,
121
+ "use_peft": false,
122
+ "push_to_hub": true,
123
+ "repo_id": "chomeed/peg_insertion_full_rabc_delta2_using_rewind",
124
+ "private": null,
125
+ "tags": null,
126
+ "license": null,
127
+ "pretrained_path": null,
128
+ "horizon": 2,
129
+ "n_action_steps": 2,
130
+ "normalization_mapping": {
131
+ "VISUAL": "MEAN_STD",
132
+ "STATE": "MIN_MAX",
133
+ "ACTION": "MIN_MAX"
134
+ },
135
+ "drop_n_last_frames": 7,
136
+ "vision_backbone": "resnet18",
137
+ "crop_shape": [
138
+ 256,
139
+ 256
140
+ ],
141
+ "crop_is_random": true,
142
+ "pretrained_backbone_weights": null,
143
+ "use_group_norm": true,
144
+ "spatial_softmax_num_keypoints": 32,
145
+ "use_separate_rgb_encoder_per_camera": false,
146
+ "down_dims": [
147
+ 512
148
+ ],
149
+ "kernel_size": 5,
150
+ "n_groups": 8,
151
+ "diffusion_step_embed_dim": 128,
152
+ "use_film_scale_modulation": true,
153
+ "noise_scheduler_type": "DDPM",
154
+ "num_train_timesteps": 100,
155
+ "beta_schedule": "squaredcos_cap_v2",
156
+ "beta_start": 0.0001,
157
+ "beta_end": 0.02,
158
+ "prediction_type": "epsilon",
159
+ "clip_sample": true,
160
+ "clip_sample_range": 1.0,
161
+ "num_inference_steps": null,
162
+ "do_mask_loss_for_padding": false,
163
+ "optimizer_lr": 0.0001,
164
+ "optimizer_betas": [
165
+ 0.95,
166
+ 0.999
167
+ ],
168
+ "optimizer_eps": 1e-08,
169
+ "optimizer_weight_decay": 1e-06,
170
+ "scheduler_name": "cosine",
171
+ "scheduler_warmup_steps": 500,
172
+ "use_rabc_weighting": true,
173
+ "weight_key": "rabc.weight"
174
+ },
175
+ "output_dir": "/scratch2/chomeed/outputs/train/peg_insertion_dp_rabc_delta2_rewind_256",
176
+ "job_name": "peg_insertion_full_rabc_delta2_using_rewind",
177
+ "resume": false,
178
+ "seed": 1000,
179
+ "num_workers": 8,
180
+ "batch_size": 64,
181
+ "steps": 200000,
182
+ "eval_freq": 0,
183
+ "log_freq": 200,
184
+ "tolerance_s": 0.0001,
185
+ "save_checkpoint": true,
186
+ "save_freq": 10000,
187
+ "use_policy_training_preset": true,
188
+ "optimizer": {
189
+ "type": "adam",
190
+ "lr": 0.0001,
191
+ "weight_decay": 1e-06,
192
+ "grad_clip_norm": 10.0,
193
+ "betas": [
194
+ 0.95,
195
+ 0.999
196
+ ],
197
+ "eps": 1e-08
198
+ },
199
+ "scheduler": {
200
+ "type": "diffuser",
201
+ "num_warmup_steps": 500,
202
+ "name": "cosine"
203
+ },
204
+ "val": {
205
+ "repo_id": [
206
+ "chomeed/anubis_pick_carrot_out_of_pot_randomized_val_single",
207
+ "chomeed/anubis_stack_tuna_single",
208
+ "chomeed/aloha_static_cups_open_single",
209
+ "chomeed/aloha_static_candy_single",
210
+ "jellyho/aloha_box_into_pot_single",
211
+ "jellyho/aloha_dish_drainer_single"
212
+ ],
213
+ "root": null,
214
+ "episodes": null,
215
+ "image_transforms": {
216
+ "enable": false,
217
+ "max_num_transforms": 3,
218
+ "random_order": false,
219
+ "tfs": {
220
+ "brightness": {
221
+ "weight": 1.0,
222
+ "type": "ColorJitter",
223
+ "kwargs": {
224
+ "brightness": [
225
+ 0.8,
226
+ 1.2
227
+ ]
228
+ }
229
+ },
230
+ "contrast": {
231
+ "weight": 1.0,
232
+ "type": "ColorJitter",
233
+ "kwargs": {
234
+ "contrast": [
235
+ 0.8,
236
+ 1.2
237
+ ]
238
+ }
239
+ },
240
+ "saturation": {
241
+ "weight": 1.0,
242
+ "type": "ColorJitter",
243
+ "kwargs": {
244
+ "saturation": [
245
+ 0.5,
246
+ 1.5
247
+ ]
248
+ }
249
+ },
250
+ "hue": {
251
+ "weight": 1.0,
252
+ "type": "ColorJitter",
253
+ "kwargs": {
254
+ "hue": [
255
+ -0.05,
256
+ 0.05
257
+ ]
258
+ }
259
+ },
260
+ "sharpness": {
261
+ "weight": 1.0,
262
+ "type": "SharpnessJitter",
263
+ "kwargs": {
264
+ "sharpness": [
265
+ 0.5,
266
+ 1.5
267
+ ]
268
+ }
269
+ },
270
+ "affine": {
271
+ "weight": 1.0,
272
+ "type": "RandomAffine",
273
+ "kwargs": {
274
+ "degrees": [
275
+ -5.0,
276
+ 5.0
277
+ ],
278
+ "translate": [
279
+ 0.05,
280
+ 0.05
281
+ ]
282
+ }
283
+ }
284
+ }
285
+ },
286
+ "revision": null,
287
+ "use_imagenet_stats": true,
288
+ "video_backend": "torchcodec",
289
+ "streaming": false
290
+ },
291
+ "test": {
292
+ "repo_id": [
293
+ "chomeed/aloha_static_coffee_single",
294
+ "jellyho/aloha_handover_box_single"
295
+ ],
296
+ "root": null,
297
+ "episodes": null,
298
+ "image_transforms": {
299
+ "enable": false,
300
+ "max_num_transforms": 3,
301
+ "random_order": false,
302
+ "tfs": {
303
+ "brightness": {
304
+ "weight": 1.0,
305
+ "type": "ColorJitter",
306
+ "kwargs": {
307
+ "brightness": [
308
+ 0.8,
309
+ 1.2
310
+ ]
311
+ }
312
+ },
313
+ "contrast": {
314
+ "weight": 1.0,
315
+ "type": "ColorJitter",
316
+ "kwargs": {
317
+ "contrast": [
318
+ 0.8,
319
+ 1.2
320
+ ]
321
+ }
322
+ },
323
+ "saturation": {
324
+ "weight": 1.0,
325
+ "type": "ColorJitter",
326
+ "kwargs": {
327
+ "saturation": [
328
+ 0.5,
329
+ 1.5
330
+ ]
331
+ }
332
+ },
333
+ "hue": {
334
+ "weight": 1.0,
335
+ "type": "ColorJitter",
336
+ "kwargs": {
337
+ "hue": [
338
+ -0.05,
339
+ 0.05
340
+ ]
341
+ }
342
+ },
343
+ "sharpness": {
344
+ "weight": 1.0,
345
+ "type": "SharpnessJitter",
346
+ "kwargs": {
347
+ "sharpness": [
348
+ 0.5,
349
+ 1.5
350
+ ]
351
+ }
352
+ },
353
+ "affine": {
354
+ "weight": 1.0,
355
+ "type": "RandomAffine",
356
+ "kwargs": {
357
+ "degrees": [
358
+ -5.0,
359
+ 5.0
360
+ ],
361
+ "translate": [
362
+ 0.05,
363
+ 0.05
364
+ ]
365
+ }
366
+ }
367
+ }
368
+ },
369
+ "revision": null,
370
+ "use_imagenet_stats": true,
371
+ "video_backend": "torchcodec",
372
+ "streaming": false
373
+ },
374
+ "wandb": {
375
+ "enable": true,
376
+ "disable_artifact": true,
377
+ "project": "peg_insertion_dp",
378
+ "entity": null,
379
+ "notes": null,
380
+ "run_id": "p5a9qp6a",
381
+ "mode": null
382
+ },
383
+ "peft": null,
384
+ "use_rabc": false,
385
+ "rabc_progress_path": null,
386
+ "rabc_kappa": 0.01,
387
+ "rabc_epsilon": 1e-06,
388
+ "rabc_head_mode": "sparse",
389
+ "rename_map": {},
390
+ "checkpoint_path": null
391
+ }