Hugging Face's logo

RoboSynChallenge
/
SmolVLA_sim_sample_loading

Safetensors
Model card Files
xet
 Community
RoboSynChallenge commited on
Commit
2a4f41a
·
verified ·
1 Parent(s): 4abd5aa

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +114 -0
  2. model.safetensors +3 -0
  3. policy_postprocessor.json +32 -0
  4. policy_postprocessor_step_0_unnormalizer_processor.safetensors +3 -0
  5. policy_preprocessor.json +99 -0
  6. policy_preprocessor_step_5_normalizer_processor.safetensors +3 -0
  7. train_config.json +267 -0
config.json ADDED
@@ -0,0 +1,114 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "smolvla",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 14
9
+ ]
10
+ },
11
+ "observation.qvel": {
12
+ "type": "STATE",
13
+ "shape": [
14
+ 14
15
+ ]
16
+ },
17
+ "observation.qf": {
18
+ "type": "STATE",
19
+ "shape": [
20
+ 14
21
+ ]
22
+ },
23
+ "observation.images.cam_high": {
24
+ "type": "VISUAL",
25
+ "shape": [
26
+ 3,
27
+ 480,
28
+ 640
29
+ ]
30
+ },
31
+ "observation.images.cam_right_wrist": {
32
+ "type": "VISUAL",
33
+ "shape": [
34
+ 3,
35
+ 480,
36
+ 640
37
+ ]
38
+ },
39
+ "observation.images.cam_left_wrist": {
40
+ "type": "VISUAL",
41
+ "shape": [
42
+ 3,
43
+ 480,
44
+ 640
45
+ ]
46
+ }
47
+ },
48
+ "output_features": {
49
+ "action": {
50
+ "type": "ACTION",
51
+ "shape": [
52
+ 14
53
+ ]
54
+ }
55
+ },
56
+ "device": "cuda",
57
+ "use_amp": false,
58
+ "use_peft": false,
59
+ "push_to_hub": false,
60
+ "repo_id": null,
61
+ "private": null,
62
+ "tags": null,
63
+ "license": null,
64
+ "pretrained_path": null,
65
+ "pretrained_revision": null,
66
+ "chunk_size": 50,
67
+ "n_action_steps": 50,
68
+ "normalization_mapping": {
69
+ "VISUAL": "IDENTITY",
70
+ "STATE": "MEAN_STD",
71
+ "ACTION": "MEAN_STD"
72
+ },
73
+ "max_state_dim": 32,
74
+ "max_action_dim": 32,
75
+ "resize_imgs_with_padding": [
76
+ 512,
77
+ 512
78
+ ],
79
+ "empty_cameras": 0,
80
+ "adapt_to_pi_aloha": false,
81
+ "use_delta_joint_actions_aloha": false,
82
+ "tokenizer_max_length": 48,
83
+ "num_steps": 10,
84
+ "use_cache": true,
85
+ "freeze_vision_encoder": true,
86
+ "train_expert_only": true,
87
+ "train_state_proj": true,
88
+ "optimizer_lr": 0.0001,
89
+ "optimizer_betas": [
90
+ 0.9,
91
+ 0.95
92
+ ],
93
+ "optimizer_eps": 1e-08,
94
+ "optimizer_weight_decay": 1e-10,
95
+ "optimizer_grad_clip_norm": 10,
96
+ "scheduler_warmup_steps": 1000,
97
+ "scheduler_decay_steps": 30000,
98
+ "scheduler_decay_lr": 2.5e-06,
99
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
100
+ "load_vlm_weights": true,
101
+ "add_image_special_tokens": false,
102
+ "attention_mode": "cross_attn",
103
+ "prefix_length": -1,
104
+ "pad_language_to": "longest",
105
+ "num_expert_layers": -1,
106
+ "num_vlm_layers": 16,
107
+ "self_attn_every_n_layers": 2,
108
+ "expert_width_multiplier": 0.75,
109
+ "min_period": 0.004,
110
+ "max_period": 4.0,
111
+ "rtc_config": null,
112
+ "compile_model": false,
113
+ "compile_mode": "max-autotune"
114
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5faf98300e6d456e1c843b4a052aad540a4ca612f184471d156cbb95b511e359
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
+ 14
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 sha256:9ed9a6ba46bf0cc2cceb9dbd3faf26f1325190bed77ad6f99e0b67d526d38108
3
+ size 13728
policy_preprocessor.json ADDED
@@ -0,0 +1,99 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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.state": {
42
+ "type": "STATE",
43
+ "shape": [
44
+ 14
45
+ ]
46
+ },
47
+ "observation.qvel": {
48
+ "type": "STATE",
49
+ "shape": [
50
+ 14
51
+ ]
52
+ },
53
+ "observation.qf": {
54
+ "type": "STATE",
55
+ "shape": [
56
+ 14
57
+ ]
58
+ },
59
+ "observation.images.cam_high": {
60
+ "type": "VISUAL",
61
+ "shape": [
62
+ 3,
63
+ 480,
64
+ 640
65
+ ]
66
+ },
67
+ "observation.images.cam_right_wrist": {
68
+ "type": "VISUAL",
69
+ "shape": [
70
+ 3,
71
+ 480,
72
+ 640
73
+ ]
74
+ },
75
+ "observation.images.cam_left_wrist": {
76
+ "type": "VISUAL",
77
+ "shape": [
78
+ 3,
79
+ 480,
80
+ 640
81
+ ]
82
+ },
83
+ "action": {
84
+ "type": "ACTION",
85
+ "shape": [
86
+ 14
87
+ ]
88
+ }
89
+ },
90
+ "norm_map": {
91
+ "VISUAL": "IDENTITY",
92
+ "STATE": "MEAN_STD",
93
+ "ACTION": "MEAN_STD"
94
+ }
95
+ },
96
+ "state_file": "policy_preprocessor_step_5_normalizer_processor.safetensors"
97
+ }
98
+ ]
99
+ }
policy_preprocessor_step_5_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79d587e4f925bbf0760aac4c445da46a51144c0c0e97425dbf08a54b779fabe7
3
+ size 13736
train_config.json ADDED
@@ -0,0 +1,267 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "RoboSynChallenge/cobotmagic_Sim_sample_loading",
4
+ "root": "/data1/jinruixing/datasets/cobotmagic_Sim_sample_loading",
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
+ "return_uint8": false,
81
+ "depth_output_unit": "mm",
82
+ "streaming": false,
83
+ "eval_split": 0.0
84
+ },
85
+ "env": null,
86
+ "policy": {
87
+ "type": "smolvla",
88
+ "n_obs_steps": 1,
89
+ "input_features": {
90
+ "observation.state": {
91
+ "type": "STATE",
92
+ "shape": [
93
+ 14
94
+ ]
95
+ },
96
+ "observation.qvel": {
97
+ "type": "STATE",
98
+ "shape": [
99
+ 14
100
+ ]
101
+ },
102
+ "observation.qf": {
103
+ "type": "STATE",
104
+ "shape": [
105
+ 14
106
+ ]
107
+ },
108
+ "observation.images.cam_high": {
109
+ "type": "VISUAL",
110
+ "shape": [
111
+ 3,
112
+ 480,
113
+ 640
114
+ ]
115
+ },
116
+ "observation.images.cam_right_wrist": {
117
+ "type": "VISUAL",
118
+ "shape": [
119
+ 3,
120
+ 480,
121
+ 640
122
+ ]
123
+ },
124
+ "observation.images.cam_left_wrist": {
125
+ "type": "VISUAL",
126
+ "shape": [
127
+ 3,
128
+ 480,
129
+ 640
130
+ ]
131
+ }
132
+ },
133
+ "output_features": {
134
+ "action": {
135
+ "type": "ACTION",
136
+ "shape": [
137
+ 14
138
+ ]
139
+ }
140
+ },
141
+ "device": "cuda",
142
+ "use_amp": false,
143
+ "use_peft": false,
144
+ "push_to_hub": false,
145
+ "repo_id": null,
146
+ "private": null,
147
+ "tags": null,
148
+ "license": null,
149
+ "pretrained_path": null,
150
+ "pretrained_revision": null,
151
+ "chunk_size": 50,
152
+ "n_action_steps": 50,
153
+ "normalization_mapping": {
154
+ "VISUAL": "IDENTITY",
155
+ "STATE": "MEAN_STD",
156
+ "ACTION": "MEAN_STD"
157
+ },
158
+ "max_state_dim": 32,
159
+ "max_action_dim": 32,
160
+ "resize_imgs_with_padding": [
161
+ 512,
162
+ 512
163
+ ],
164
+ "empty_cameras": 0,
165
+ "adapt_to_pi_aloha": false,
166
+ "use_delta_joint_actions_aloha": false,
167
+ "tokenizer_max_length": 48,
168
+ "num_steps": 10,
169
+ "use_cache": true,
170
+ "freeze_vision_encoder": true,
171
+ "train_expert_only": true,
172
+ "train_state_proj": true,
173
+ "optimizer_lr": 0.0001,
174
+ "optimizer_betas": [
175
+ 0.9,
176
+ 0.95
177
+ ],
178
+ "optimizer_eps": 1e-08,
179
+ "optimizer_weight_decay": 1e-10,
180
+ "optimizer_grad_clip_norm": 10,
181
+ "scheduler_warmup_steps": 1000,
182
+ "scheduler_decay_steps": 30000,
183
+ "scheduler_decay_lr": 2.5e-06,
184
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
185
+ "load_vlm_weights": true,
186
+ "add_image_special_tokens": false,
187
+ "attention_mode": "cross_attn",
188
+ "prefix_length": -1,
189
+ "pad_language_to": "longest",
190
+ "num_expert_layers": -1,
191
+ "num_vlm_layers": 16,
192
+ "self_attn_every_n_layers": 2,
193
+ "expert_width_multiplier": 0.75,
194
+ "min_period": 0.004,
195
+ "max_period": 4.0,
196
+ "rtc_config": null,
197
+ "compile_model": false,
198
+ "compile_mode": "max-autotune"
199
+ },
200
+ "reward_model": null,
201
+ "output_dir": "outputs/train/cobotmagic_smolvla_sample_loading_run1",
202
+ "job_name": "cobotmagic_smolvla_sample_loading_run1",
203
+ "resume": false,
204
+ "seed": 1000,
205
+ "cudnn_deterministic": false,
206
+ "num_workers": 8,
207
+ "batch_size": 32,
208
+ "prefetch_factor": 4,
209
+ "persistent_workers": true,
210
+ "steps": 50000,
211
+ "env_eval_freq": 20000,
212
+ "log_freq": 200,
213
+ "eval_steps": 0,
214
+ "max_eval_samples": 0,
215
+ "tolerance_s": 0.0001,
216
+ "save_checkpoint": true,
217
+ "save_freq": 20000,
218
+ "use_policy_training_preset": true,
219
+ "optimizer": {
220
+ "type": "adamw",
221
+ "lr": 0.0001,
222
+ "weight_decay": 1e-10,
223
+ "grad_clip_norm": 10,
224
+ "betas": [
225
+ 0.9,
226
+ 0.95
227
+ ],
228
+ "eps": 1e-08
229
+ },
230
+ "scheduler": {
231
+ "type": "cosine_decay_with_warmup",
232
+ "num_warmup_steps": 1000,
233
+ "num_decay_steps": 30000,
234
+ "peak_lr": 0.0001,
235
+ "decay_lr": 2.5e-06
236
+ },
237
+ "eval": {
238
+ "n_episodes": 50,
239
+ "batch_size": 50,
240
+ "use_async_envs": true,
241
+ "recording": false,
242
+ "recording_repo_id": null,
243
+ "recording_private": false
244
+ },
245
+ "wandb": {
246
+ "enable": false,
247
+ "disable_artifact": false,
248
+ "project": "lerobot",
249
+ "entity": null,
250
+ "notes": null,
251
+ "run_id": null,
252
+ "mode": null,
253
+ "add_tags": true
254
+ },
255
+ "peft": null,
256
+ "job": {
257
+ "target": null,
258
+ "image": "huggingface/lerobot-gpu:latest",
259
+ "timeout": "2d",
260
+ "detach": false,
261
+ "tags": []
262
+ },
263
+ "save_checkpoint_to_hub": false,
264
+ "sample_weighting": null,
265
+ "rename_map": {},
266
+ "checkpoint_path": null
267
+ }