Hugging Face's logo

padiredd
/
dp-blockpick-2k-withoutpush

Robotics
Safetensors
LeRobot
diffusion-policy
isaac-lab
simulation
ffw-sg2
Model card Files
xet
 Community
padiredd commited on
Commit
795a30f
·
verified ·
1 Parent(s): 0d17bfb

step 20k pretrained_model

Browse files
Files changed (7) hide show
  1. step-20000/config.json +127 -0
  2. step-20000/model.safetensors +3 -0
  3. step-20000/policy_postprocessor.json +32 -0
  4. step-20000/policy_postprocessor_step_0_unnormalizer_processor.safetensors +3 -0
  5. step-20000/policy_preprocessor.json +72 -0
  6. step-20000/policy_preprocessor_step_3_normalizer_processor.safetensors +3 -0
  7. step-20000/train_config.json +326 -0
step-20000/config.json ADDED
@@ -0,0 +1,127 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "diffusion",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 22
9
+ ]
10
+ },
11
+ "observation.images.rgb.cam_left_wrist": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 240,
16
+ 240
17
+ ]
18
+ },
19
+ "observation.images.rgb.cam_right_wrist": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 240,
24
+ 240
25
+ ]
26
+ },
27
+ "observation.images.rgb.cam_left_head": {
28
+ "type": "VISUAL",
29
+ "shape": [
30
+ 3,
31
+ 240,
32
+ 240
33
+ ]
34
+ }
35
+ },
36
+ "output_features": {
37
+ "action": {
38
+ "type": "ACTION",
39
+ "shape": [
40
+ 16
41
+ ]
42
+ }
43
+ },
44
+ "device": "cuda",
45
+ "use_amp": false,
46
+ "use_peft": false,
47
+ "push_to_hub": false,
48
+ "repo_id": null,
49
+ "private": null,
50
+ "tags": null,
51
+ "license": null,
52
+ "pretrained_path": null,
53
+ "pretrained_revision": null,
54
+ "horizon": 64,
55
+ "n_action_steps": 32,
56
+ "normalization_mapping": {
57
+ "VISUAL": "MEAN_STD",
58
+ "STATE": "MIN_MAX",
59
+ "ACTION": "MIN_MAX"
60
+ },
61
+ "drop_n_last_frames": 7,
62
+ "vision_backbone": "resnet18",
63
+ "resize_shape": null,
64
+ "crop_ratio": 1.0,
65
+ "crop_shape": null,
66
+ "crop_is_random": true,
67
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
68
+ "use_group_norm": false,
69
+ "spatial_softmax_num_keypoints": 32,
70
+ "use_separate_rgb_encoder_per_camera": true,
71
+ "down_dims": [
72
+ 512,
73
+ 1024,
74
+ 2048
75
+ ],
76
+ "kernel_size": 5,
77
+ "n_groups": 8,
78
+ "diffusion_step_embed_dim": 128,
79
+ "use_film_scale_modulation": true,
80
+ "noise_scheduler_type": "DDPM",
81
+ "num_train_timesteps": 100,
82
+ "beta_schedule": "squaredcos_cap_v2",
83
+ "beta_start": 0.0001,
84
+ "beta_end": 0.02,
85
+ "prediction_type": "epsilon",
86
+ "clip_sample": true,
87
+ "clip_sample_range": 1.0,
88
+ "num_inference_steps": null,
89
+ "compile_model": false,
90
+ "compile_mode": "reduce-overhead",
91
+ "do_mask_loss_for_padding": false,
92
+ "state_noise_std": [
93
+ 0.0,
94
+ 0.0,
95
+ 0.0,
96
+ 0.0,
97
+ 0.0,
98
+ 0.0,
99
+ 0.0,
100
+ 0.0,
101
+ 0.0,
102
+ 0.0,
103
+ 0.0,
104
+ 0.0,
105
+ 0.0,
106
+ 0.0,
107
+ 0.0,
108
+ 0.0,
109
+ 0.0,
110
+ 0.0,
111
+ 0.0,
112
+ 0.0,
113
+ 0.0,
114
+ 0.0
115
+ ],
116
+ "state_dropout_prob": 0.0,
117
+ "state_dropout_dims": [],
118
+ "optimizer_lr": 0.0001,
119
+ "optimizer_betas": [
120
+ 0.95,
121
+ 0.999
122
+ ],
123
+ "optimizer_eps": 1e-08,
124
+ "optimizer_weight_decay": 1e-06,
125
+ "scheduler_name": "cosine",
126
+ "scheduler_warmup_steps": 500
127
+ }
step-20000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b0f5f60d8536c90c21172f9ff9b60fd08cdb053e3772b1c2a4af2431a6f32bf4
3
+ size 1150197416
step-20000/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
+ 16
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
+ }
step-20000/policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d98829bd0af5df598f1baed888ca09ecde45e668103548611c512df0ef8d08f0
3
+ size 9960
step-20000/policy_preprocessor.json ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ 22
30
+ ]
31
+ },
32
+ "observation.images.rgb.cam_left_wrist": {
33
+ "type": "VISUAL",
34
+ "shape": [
35
+ 3,
36
+ 240,
37
+ 240
38
+ ]
39
+ },
40
+ "observation.images.rgb.cam_right_wrist": {
41
+ "type": "VISUAL",
42
+ "shape": [
43
+ 3,
44
+ 240,
45
+ 240
46
+ ]
47
+ },
48
+ "observation.images.rgb.cam_left_head": {
49
+ "type": "VISUAL",
50
+ "shape": [
51
+ 3,
52
+ 240,
53
+ 240
54
+ ]
55
+ },
56
+ "action": {
57
+ "type": "ACTION",
58
+ "shape": [
59
+ 16
60
+ ]
61
+ }
62
+ },
63
+ "norm_map": {
64
+ "VISUAL": "MEAN_STD",
65
+ "STATE": "MIN_MAX",
66
+ "ACTION": "MIN_MAX"
67
+ }
68
+ },
69
+ "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors"
70
+ }
71
+ ]
72
+ }
step-20000/policy_preprocessor_step_3_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7a21008650ac51b8b96e5dbd81b80236f4ed3880d389fd1293e4d68fd9f4540
3
+ size 9968
step-20000/train_config.json ADDED
@@ -0,0 +1,326 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "local/pick-cube-2k-withoutpush-yaw45-full-dr2-3cam",
4
+ "root": "data_local/lerobot/pick-cube-2k-withoutpush-yaw45-full-dr2-3cam",
5
+ "episodes": null,
6
+ "image_transforms": {
7
+ "enable": true,
8
+ "max_num_transforms": 5,
9
+ "random_order": true,
10
+ "tfs": {
11
+ "brightness": {
12
+ "weight": 1.0,
13
+ "type": "ColorJitter",
14
+ "kwargs": {
15
+ "brightness": [
16
+ 0.6,
17
+ 1.4
18
+ ]
19
+ }
20
+ },
21
+ "contrast": {
22
+ "weight": 1.0,
23
+ "type": "ColorJitter",
24
+ "kwargs": {
25
+ "contrast": [
26
+ 0.6,
27
+ 1.4
28
+ ]
29
+ }
30
+ },
31
+ "saturation": {
32
+ "weight": 1.0,
33
+ "type": "ColorJitter",
34
+ "kwargs": {
35
+ "saturation": [
36
+ 0.3,
37
+ 1.7
38
+ ]
39
+ }
40
+ },
41
+ "hue": {
42
+ "weight": 1.0,
43
+ "type": "ColorJitter",
44
+ "kwargs": {
45
+ "hue": [
46
+ -0.1,
47
+ 0.1
48
+ ]
49
+ }
50
+ },
51
+ "sharpness": {
52
+ "weight": 1.0,
53
+ "type": "SharpnessJitter",
54
+ "kwargs": {
55
+ "sharpness": [
56
+ 0.3,
57
+ 2.0
58
+ ]
59
+ }
60
+ },
61
+ "affine": {
62
+ "weight": 1.0,
63
+ "type": "RandomAffine",
64
+ "kwargs": {
65
+ "degrees": [
66
+ -10.0,
67
+ 10.0
68
+ ],
69
+ "translate": [
70
+ 0.1,
71
+ 0.1
72
+ ],
73
+ "scale": [
74
+ 0.9,
75
+ 1.1
76
+ ]
77
+ }
78
+ },
79
+ "perspective": {
80
+ "weight": 1.0,
81
+ "type": "RandomPerspective",
82
+ "kwargs": {
83
+ "distortion_scale": 0.2,
84
+ "p": 1.0
85
+ }
86
+ },
87
+ "blur": {
88
+ "weight": 1.0,
89
+ "type": "GaussianBlur",
90
+ "kwargs": {
91
+ "kernel_size": 5,
92
+ "sigma": [
93
+ 0.1,
94
+ 2.0
95
+ ]
96
+ }
97
+ },
98
+ "erasing": {
99
+ "weight": 1.0,
100
+ "type": "RandomErasing",
101
+ "kwargs": {
102
+ "p": 1.0,
103
+ "scale": [
104
+ 0.02,
105
+ 0.15
106
+ ],
107
+ "ratio": [
108
+ 0.3,
109
+ 3.3
110
+ ],
111
+ "value": 0
112
+ }
113
+ },
114
+ "grayscale": {
115
+ "weight": 0.5,
116
+ "type": "RandomGrayscale",
117
+ "kwargs": {
118
+ "p": 1.0
119
+ }
120
+ }
121
+ }
122
+ },
123
+ "revision": null,
124
+ "use_imagenet_stats": true,
125
+ "video_backend": "torchcodec",
126
+ "return_uint8": false,
127
+ "depth_output_unit": "mm",
128
+ "streaming": false,
129
+ "eval_split": 0.0,
130
+ "cotrain_alpha": null,
131
+ "cotrain_split_episode": null
132
+ },
133
+ "env": null,
134
+ "policy": {
135
+ "type": "diffusion",
136
+ "n_obs_steps": 1,
137
+ "input_features": {
138
+ "observation.state": {
139
+ "type": "STATE",
140
+ "shape": [
141
+ 22
142
+ ]
143
+ },
144
+ "observation.images.rgb.cam_left_wrist": {
145
+ "type": "VISUAL",
146
+ "shape": [
147
+ 3,
148
+ 240,
149
+ 240
150
+ ]
151
+ },
152
+ "observation.images.rgb.cam_right_wrist": {
153
+ "type": "VISUAL",
154
+ "shape": [
155
+ 3,
156
+ 240,
157
+ 240
158
+ ]
159
+ },
160
+ "observation.images.rgb.cam_left_head": {
161
+ "type": "VISUAL",
162
+ "shape": [
163
+ 3,
164
+ 240,
165
+ 240
166
+ ]
167
+ }
168
+ },
169
+ "output_features": {
170
+ "action": {
171
+ "type": "ACTION",
172
+ "shape": [
173
+ 16
174
+ ]
175
+ }
176
+ },
177
+ "device": "cuda",
178
+ "use_amp": false,
179
+ "use_peft": false,
180
+ "push_to_hub": false,
181
+ "repo_id": null,
182
+ "private": null,
183
+ "tags": null,
184
+ "license": null,
185
+ "pretrained_path": null,
186
+ "pretrained_revision": null,
187
+ "horizon": 64,
188
+ "n_action_steps": 32,
189
+ "normalization_mapping": {
190
+ "VISUAL": "MEAN_STD",
191
+ "STATE": "MIN_MAX",
192
+ "ACTION": "MIN_MAX"
193
+ },
194
+ "drop_n_last_frames": 7,
195
+ "vision_backbone": "resnet18",
196
+ "resize_shape": null,
197
+ "crop_ratio": 1.0,
198
+ "crop_shape": null,
199
+ "crop_is_random": true,
200
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
201
+ "use_group_norm": false,
202
+ "spatial_softmax_num_keypoints": 32,
203
+ "use_separate_rgb_encoder_per_camera": true,
204
+ "down_dims": [
205
+ 512,
206
+ 1024,
207
+ 2048
208
+ ],
209
+ "kernel_size": 5,
210
+ "n_groups": 8,
211
+ "diffusion_step_embed_dim": 128,
212
+ "use_film_scale_modulation": true,
213
+ "noise_scheduler_type": "DDPM",
214
+ "num_train_timesteps": 100,
215
+ "beta_schedule": "squaredcos_cap_v2",
216
+ "beta_start": 0.0001,
217
+ "beta_end": 0.02,
218
+ "prediction_type": "epsilon",
219
+ "clip_sample": true,
220
+ "clip_sample_range": 1.0,
221
+ "num_inference_steps": null,
222
+ "compile_model": false,
223
+ "compile_mode": "reduce-overhead",
224
+ "do_mask_loss_for_padding": false,
225
+ "state_noise_std": [
226
+ 0.0,
227
+ 0.0,
228
+ 0.0,
229
+ 0.0,
230
+ 0.0,
231
+ 0.0,
232
+ 0.0,
233
+ 0.0,
234
+ 0.0,
235
+ 0.0,
236
+ 0.0,
237
+ 0.0,
238
+ 0.0,
239
+ 0.0,
240
+ 0.0,
241
+ 0.0,
242
+ 0.0,
243
+ 0.0,
244
+ 0.0,
245
+ 0.0,
246
+ 0.0,
247
+ 0.0
248
+ ],
249
+ "state_dropout_prob": 0.0,
250
+ "state_dropout_dims": [],
251
+ "optimizer_lr": 0.0001,
252
+ "optimizer_betas": [
253
+ 0.95,
254
+ 0.999
255
+ ],
256
+ "optimizer_eps": 1e-08,
257
+ "optimizer_weight_decay": 1e-06,
258
+ "scheduler_name": "cosine",
259
+ "scheduler_warmup_steps": 500
260
+ },
261
+ "reward_model": null,
262
+ "output_dir": "checkpoints_train/dp-blockpick-2k-withoutpush",
263
+ "job_name": "dp_blockpick_2k_withoutpush",
264
+ "resume": false,
265
+ "seed": 1000,
266
+ "cudnn_deterministic": false,
267
+ "num_workers": 16,
268
+ "batch_size": 32,
269
+ "prefetch_factor": 4,
270
+ "persistent_workers": true,
271
+ "steps": 100000,
272
+ "env_eval_freq": 20000,
273
+ "log_freq": 200,
274
+ "eval_steps": 0,
275
+ "max_eval_samples": 0,
276
+ "tolerance_s": 0.0001,
277
+ "save_checkpoint": true,
278
+ "save_freq": 10000,
279
+ "use_policy_training_preset": true,
280
+ "optimizer": {
281
+ "type": "adam",
282
+ "lr": 0.0001,
283
+ "weight_decay": 1e-06,
284
+ "grad_clip_norm": 10.0,
285
+ "betas": [
286
+ 0.95,
287
+ 0.999
288
+ ],
289
+ "eps": 1e-08
290
+ },
291
+ "scheduler": {
292
+ "type": "diffuser",
293
+ "num_warmup_steps": 500,
294
+ "name": "cosine"
295
+ },
296
+ "eval": {
297
+ "n_episodes": 50,
298
+ "batch_size": 33,
299
+ "use_async_envs": true,
300
+ "recording": false,
301
+ "recording_repo_id": null,
302
+ "recording_private": false
303
+ },
304
+ "wandb": {
305
+ "enable": false,
306
+ "disable_artifact": false,
307
+ "project": "lerobot",
308
+ "entity": null,
309
+ "notes": null,
310
+ "run_id": null,
311
+ "mode": null,
312
+ "add_tags": true
313
+ },
314
+ "peft": null,
315
+ "job": {
316
+ "target": null,
317
+ "image": "huggingface/lerobot-gpu:latest",
318
+ "timeout": "2d",
319
+ "detach": false,
320
+ "tags": []
321
+ },
322
+ "save_checkpoint_to_hub": false,
323
+ "sample_weighting": null,
324
+ "rename_map": {},
325
+ "checkpoint_path": null
326
+ }