dannyh147 commited on
Commit
6340a67
·
verified ·
1 Parent(s): 886fd08

Upload folder using huggingface_hub

Browse files
config.json ADDED
@@ -0,0 +1,90 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "smolvla",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.images.image": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 256,
10
+ 256
11
+ ]
12
+ },
13
+ "observation.images.image2": {
14
+ "type": "VISUAL",
15
+ "shape": [
16
+ 3,
17
+ 256,
18
+ 256
19
+ ]
20
+ },
21
+ "observation.state": {
22
+ "type": "STATE",
23
+ "shape": [
24
+ 8
25
+ ]
26
+ }
27
+ },
28
+ "output_features": {
29
+ "action": {
30
+ "type": "ACTION",
31
+ "shape": [
32
+ 7
33
+ ]
34
+ }
35
+ },
36
+ "device": "cuda",
37
+ "use_amp": false,
38
+ "push_to_hub": true,
39
+ "repo_id": "/libero-test",
40
+ "private": null,
41
+ "tags": null,
42
+ "license": null,
43
+ "pretrained_path": null,
44
+ "chunk_size": 50,
45
+ "n_action_steps": 50,
46
+ "normalization_mapping": {
47
+ "VISUAL": "IDENTITY",
48
+ "STATE": "MEAN_STD",
49
+ "ACTION": "MEAN_STD"
50
+ },
51
+ "max_state_dim": 32,
52
+ "max_action_dim": 32,
53
+ "resize_imgs_with_padding": [
54
+ 512,
55
+ 512
56
+ ],
57
+ "empty_cameras": 0,
58
+ "adapt_to_pi_aloha": false,
59
+ "use_delta_joint_actions_aloha": false,
60
+ "tokenizer_max_length": 48,
61
+ "num_steps": 10,
62
+ "use_cache": true,
63
+ "freeze_vision_encoder": true,
64
+ "train_expert_only": true,
65
+ "train_state_proj": true,
66
+ "optimizer_lr": 0.0001,
67
+ "optimizer_betas": [
68
+ 0.9,
69
+ 0.95
70
+ ],
71
+ "optimizer_eps": 1e-08,
72
+ "optimizer_weight_decay": 1e-10,
73
+ "optimizer_grad_clip_norm": 10,
74
+ "scheduler_warmup_steps": 1000,
75
+ "scheduler_decay_steps": 30000,
76
+ "scheduler_decay_lr": 2.5e-06,
77
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
78
+ "load_vlm_weights": true,
79
+ "add_image_special_tokens": false,
80
+ "attention_mode": "cross_attn",
81
+ "prefix_length": -1,
82
+ "pad_language_to": "longest",
83
+ "num_expert_layers": -1,
84
+ "num_vlm_layers": 16,
85
+ "self_attn_every_n_layers": 2,
86
+ "expert_width_multiplier": 0.75,
87
+ "min_period": 0.004,
88
+ "max_period": 4.0,
89
+ "rtc_config": null
90
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ec34be8110579fa4374df257534d2b9fa51a1a66d60dd2e523b5206712052d3
3
+ size 906712520
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
+ 7
13
+ ]
14
+ }
15
+ },
16
+ "norm_map": {
17
+ "VISUAL": "IDENTITY",
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
+ }
policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a002c0df7f79c5b169c5a899ad151d4ea1bed246c7d82bd93ed1556558d517a9
3
+ size 7668
policy_preprocessor.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "smolvla_new_line_processor",
16
+ "config": {}
17
+ },
18
+ {
19
+ "registry_name": "tokenizer_processor",
20
+ "config": {
21
+ "max_length": 48,
22
+ "task_key": "task",
23
+ "padding_side": "right",
24
+ "padding": "longest",
25
+ "truncation": true,
26
+ "tokenizer_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct"
27
+ }
28
+ },
29
+ {
30
+ "registry_name": "device_processor",
31
+ "config": {
32
+ "device": "cuda",
33
+ "float_dtype": null
34
+ }
35
+ },
36
+ {
37
+ "registry_name": "normalizer_processor",
38
+ "config": {
39
+ "eps": 1e-08,
40
+ "features": {
41
+ "observation.images.image": {
42
+ "type": "VISUAL",
43
+ "shape": [
44
+ 3,
45
+ 256,
46
+ 256
47
+ ]
48
+ },
49
+ "observation.images.image2": {
50
+ "type": "VISUAL",
51
+ "shape": [
52
+ 3,
53
+ 256,
54
+ 256
55
+ ]
56
+ },
57
+ "observation.state": {
58
+ "type": "STATE",
59
+ "shape": [
60
+ 8
61
+ ]
62
+ },
63
+ "action": {
64
+ "type": "ACTION",
65
+ "shape": [
66
+ 7
67
+ ]
68
+ }
69
+ },
70
+ "norm_map": {
71
+ "VISUAL": "IDENTITY",
72
+ "STATE": "MEAN_STD",
73
+ "ACTION": "MEAN_STD"
74
+ }
75
+ },
76
+ "state_file": "policy_preprocessor_step_5_normalizer_processor.safetensors"
77
+ }
78
+ ]
79
+ }
policy_preprocessor_step_5_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a002c0df7f79c5b169c5a899ad151d4ea1bed246c7d82bd93ed1556558d517a9
3
+ size 7668
train_config.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "HuggingFaceVLA/libero",
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
+ },
82
+ "env": {
83
+ "type": "libero",
84
+ "task": "libero_10",
85
+ "fps": 30,
86
+ "features": {
87
+ "action": {
88
+ "type": "ACTION",
89
+ "shape": [
90
+ 7
91
+ ]
92
+ },
93
+ "pixels/agentview_image": {
94
+ "type": "VISUAL",
95
+ "shape": [
96
+ 360,
97
+ 360,
98
+ 3
99
+ ]
100
+ },
101
+ "pixels/robot0_eye_in_hand_image": {
102
+ "type": "VISUAL",
103
+ "shape": [
104
+ 360,
105
+ 360,
106
+ 3
107
+ ]
108
+ },
109
+ "robot_state/eef/pos": {
110
+ "type": "STATE",
111
+ "shape": [
112
+ 3
113
+ ]
114
+ },
115
+ "robot_state/eef/quat": {
116
+ "type": "STATE",
117
+ "shape": [
118
+ 4
119
+ ]
120
+ },
121
+ "robot_state/eef/mat": {
122
+ "type": "STATE",
123
+ "shape": [
124
+ 3,
125
+ 3
126
+ ]
127
+ },
128
+ "robot_state/gripper/qpos": {
129
+ "type": "STATE",
130
+ "shape": [
131
+ 2
132
+ ]
133
+ },
134
+ "robot_state/gripper/qvel": {
135
+ "type": "STATE",
136
+ "shape": [
137
+ 2
138
+ ]
139
+ },
140
+ "robot_state/joints/pos": {
141
+ "type": "STATE",
142
+ "shape": [
143
+ 7
144
+ ]
145
+ },
146
+ "robot_state/joints/vel": {
147
+ "type": "STATE",
148
+ "shape": [
149
+ 7
150
+ ]
151
+ }
152
+ },
153
+ "features_map": {
154
+ "action": "action",
155
+ "robot_state/eef/pos": "observation.state.eef_pos",
156
+ "robot_state/eef/quat": "observation.state.eef_quat",
157
+ "robot_state/eef/mat": "observation.state.eef_mat",
158
+ "robot_state/gripper/qpos": "observation.state.gripper_qpos",
159
+ "robot_state/gripper/qvel": "observation.state.gripper_qvel",
160
+ "robot_state/joints/pos": "observation.state.joint_pos",
161
+ "robot_state/joints/vel": "observation.state.joint_vel",
162
+ "pixels/agentview_image": "observation.images.image",
163
+ "pixels/robot0_eye_in_hand_image": "observation.images.image2"
164
+ },
165
+ "max_parallel_tasks": 1,
166
+ "disable_env_checker": true,
167
+ "episode_length": 520,
168
+ "obs_type": "pixels_agent_pos",
169
+ "render_mode": "rgb_array",
170
+ "camera_name": "agentview_image,robot0_eye_in_hand_image",
171
+ "init_states": true,
172
+ "camera_name_mapping": null,
173
+ "observation_height": 360,
174
+ "observation_width": 360
175
+ },
176
+ "policy": {
177
+ "type": "smolvla",
178
+ "n_obs_steps": 1,
179
+ "input_features": {
180
+ "observation.images.image": {
181
+ "type": "VISUAL",
182
+ "shape": [
183
+ 3,
184
+ 256,
185
+ 256
186
+ ]
187
+ },
188
+ "observation.images.image2": {
189
+ "type": "VISUAL",
190
+ "shape": [
191
+ 3,
192
+ 256,
193
+ 256
194
+ ]
195
+ },
196
+ "observation.state": {
197
+ "type": "STATE",
198
+ "shape": [
199
+ 8
200
+ ]
201
+ }
202
+ },
203
+ "output_features": {
204
+ "action": {
205
+ "type": "ACTION",
206
+ "shape": [
207
+ 7
208
+ ]
209
+ }
210
+ },
211
+ "device": "cuda",
212
+ "use_amp": false,
213
+ "push_to_hub": true,
214
+ "repo_id": "/libero-test",
215
+ "private": null,
216
+ "tags": null,
217
+ "license": null,
218
+ "pretrained_path": null,
219
+ "chunk_size": 50,
220
+ "n_action_steps": 50,
221
+ "normalization_mapping": {
222
+ "VISUAL": "IDENTITY",
223
+ "STATE": "MEAN_STD",
224
+ "ACTION": "MEAN_STD"
225
+ },
226
+ "max_state_dim": 32,
227
+ "max_action_dim": 32,
228
+ "resize_imgs_with_padding": [
229
+ 512,
230
+ 512
231
+ ],
232
+ "empty_cameras": 0,
233
+ "adapt_to_pi_aloha": false,
234
+ "use_delta_joint_actions_aloha": false,
235
+ "tokenizer_max_length": 48,
236
+ "num_steps": 10,
237
+ "use_cache": true,
238
+ "freeze_vision_encoder": true,
239
+ "train_expert_only": true,
240
+ "train_state_proj": true,
241
+ "optimizer_lr": 0.0001,
242
+ "optimizer_betas": [
243
+ 0.9,
244
+ 0.95
245
+ ],
246
+ "optimizer_eps": 1e-08,
247
+ "optimizer_weight_decay": 1e-10,
248
+ "optimizer_grad_clip_norm": 10,
249
+ "scheduler_warmup_steps": 1000,
250
+ "scheduler_decay_steps": 30000,
251
+ "scheduler_decay_lr": 2.5e-06,
252
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
253
+ "load_vlm_weights": true,
254
+ "add_image_special_tokens": false,
255
+ "attention_mode": "cross_attn",
256
+ "prefix_length": -1,
257
+ "pad_language_to": "longest",
258
+ "num_expert_layers": -1,
259
+ "num_vlm_layers": 16,
260
+ "self_attn_every_n_layers": 2,
261
+ "expert_width_multiplier": 0.75,
262
+ "min_period": 0.004,
263
+ "max_period": 4.0,
264
+ "rtc_config": null
265
+ },
266
+ "output_dir": "outputs/smolvla_libero_run1",
267
+ "job_name": "libero_smolvla",
268
+ "resume": false,
269
+ "seed": 1000,
270
+ "num_workers": 4,
271
+ "batch_size": 4,
272
+ "steps": 100000,
273
+ "eval_freq": 1000,
274
+ "log_freq": 200,
275
+ "save_checkpoint": true,
276
+ "save_freq": 20000,
277
+ "use_policy_training_preset": true,
278
+ "optimizer": {
279
+ "type": "adamw",
280
+ "lr": 0.0001,
281
+ "weight_decay": 1e-10,
282
+ "grad_clip_norm": 10,
283
+ "betas": [
284
+ 0.9,
285
+ 0.95
286
+ ],
287
+ "eps": 1e-08
288
+ },
289
+ "scheduler": {
290
+ "type": "cosine_decay_with_warmup",
291
+ "num_warmup_steps": 1000,
292
+ "num_decay_steps": 30000,
293
+ "peak_lr": 0.0001,
294
+ "decay_lr": 2.5e-06
295
+ },
296
+ "eval": {
297
+ "n_episodes": 1,
298
+ "batch_size": 1,
299
+ "use_async_envs": false
300
+ },
301
+ "wandb": {
302
+ "enable": false,
303
+ "disable_artifact": false,
304
+ "project": "lerobot",
305
+ "entity": null,
306
+ "notes": null,
307
+ "run_id": null,
308
+ "mode": null
309
+ },
310
+ "checkpoint_path": null,
311
+ "rename_map": {}
312
+ }