Hugging Face's logo

Beeface
/
smolvla-libero-spatial

Robotics
LeRobot
Safetensors
English
smolvla
libero
manipulation
Model card Files
xet
 Community
Beeface commited on
Commit
0cbdb61
·
verified ·
1 Parent(s): 9d56fd7

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +93 -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 +79 -0
  6. policy_preprocessor_step_5_normalizer_processor.safetensors +3 -0
  7. train_config.json +231 -0
config.json ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": true,
38
+ "use_peft": false,
39
+ "push_to_hub": false,
40
+ "repo_id": null,
41
+ "private": null,
42
+ "tags": null,
43
+ "license": null,
44
+ "pretrained_path": "/content/drive/MyDrive/smolvla_libero/smolvla_base",
45
+ "chunk_size": 50,
46
+ "n_action_steps": 50,
47
+ "normalization_mapping": {
48
+ "VISUAL": "IDENTITY",
49
+ "STATE": "MEAN_STD",
50
+ "ACTION": "MEAN_STD"
51
+ },
52
+ "max_state_dim": 32,
53
+ "max_action_dim": 32,
54
+ "resize_imgs_with_padding": [
55
+ 512,
56
+ 512
57
+ ],
58
+ "empty_cameras": 0,
59
+ "adapt_to_pi_aloha": false,
60
+ "use_delta_joint_actions_aloha": false,
61
+ "tokenizer_max_length": 48,
62
+ "num_steps": 10,
63
+ "use_cache": true,
64
+ "freeze_vision_encoder": true,
65
+ "train_expert_only": true,
66
+ "train_state_proj": true,
67
+ "optimizer_lr": 0.0001,
68
+ "optimizer_betas": [
69
+ 0.9,
70
+ 0.95
71
+ ],
72
+ "optimizer_eps": 1e-08,
73
+ "optimizer_weight_decay": 1e-10,
74
+ "optimizer_grad_clip_norm": 10,
75
+ "scheduler_warmup_steps": 1000,
76
+ "scheduler_decay_steps": 30000,
77
+ "scheduler_decay_lr": 2.5e-06,
78
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
79
+ "load_vlm_weights": false,
80
+ "add_image_special_tokens": false,
81
+ "attention_mode": "cross_attn",
82
+ "prefix_length": -1,
83
+ "pad_language_to": "longest",
84
+ "num_expert_layers": -1,
85
+ "num_vlm_layers": 16,
86
+ "self_attn_every_n_layers": 2,
87
+ "expert_width_multiplier": 0.75,
88
+ "min_period": 0.004,
89
+ "max_period": 4.0,
90
+ "rtc_config": null,
91
+ "compile_model": false,
92
+ "compile_mode": "max-autotune"
93
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c83db079e01b0aefaa4020589b90122095eadba8cc5943410eb31d69910e8906
3
+ size 1197789224
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": "max_length",
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 +1,231 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": null,
83
+ "policy": {
84
+ "type": "smolvla",
85
+ "n_obs_steps": 1,
86
+ "input_features": {
87
+ "observation.images.image": {
88
+ "type": "VISUAL",
89
+ "shape": [
90
+ 3,
91
+ 256,
92
+ 256
93
+ ]
94
+ },
95
+ "observation.images.image2": {
96
+ "type": "VISUAL",
97
+ "shape": [
98
+ 3,
99
+ 256,
100
+ 256
101
+ ]
102
+ },
103
+ "observation.state": {
104
+ "type": "STATE",
105
+ "shape": [
106
+ 8
107
+ ]
108
+ }
109
+ },
110
+ "output_features": {
111
+ "action": {
112
+ "type": "ACTION",
113
+ "shape": [
114
+ 7
115
+ ]
116
+ }
117
+ },
118
+ "device": "cuda",
119
+ "use_amp": true,
120
+ "use_peft": false,
121
+ "push_to_hub": false,
122
+ "repo_id": null,
123
+ "private": null,
124
+ "tags": null,
125
+ "license": null,
126
+ "pretrained_path": "/content/drive/MyDrive/smolvla_libero/smolvla_base",
127
+ "chunk_size": 50,
128
+ "n_action_steps": 50,
129
+ "normalization_mapping": {
130
+ "VISUAL": "IDENTITY",
131
+ "STATE": "MEAN_STD",
132
+ "ACTION": "MEAN_STD"
133
+ },
134
+ "max_state_dim": 32,
135
+ "max_action_dim": 32,
136
+ "resize_imgs_with_padding": [
137
+ 512,
138
+ 512
139
+ ],
140
+ "empty_cameras": 0,
141
+ "adapt_to_pi_aloha": false,
142
+ "use_delta_joint_actions_aloha": false,
143
+ "tokenizer_max_length": 48,
144
+ "num_steps": 10,
145
+ "use_cache": true,
146
+ "freeze_vision_encoder": true,
147
+ "train_expert_only": true,
148
+ "train_state_proj": true,
149
+ "optimizer_lr": 0.0001,
150
+ "optimizer_betas": [
151
+ 0.9,
152
+ 0.95
153
+ ],
154
+ "optimizer_eps": 1e-08,
155
+ "optimizer_weight_decay": 1e-10,
156
+ "optimizer_grad_clip_norm": 10,
157
+ "scheduler_warmup_steps": 1000,
158
+ "scheduler_decay_steps": 30000,
159
+ "scheduler_decay_lr": 2.5e-06,
160
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
161
+ "load_vlm_weights": false,
162
+ "add_image_special_tokens": false,
163
+ "attention_mode": "cross_attn",
164
+ "prefix_length": -1,
165
+ "pad_language_to": "longest",
166
+ "num_expert_layers": -1,
167
+ "num_vlm_layers": 16,
168
+ "self_attn_every_n_layers": 2,
169
+ "expert_width_multiplier": 0.75,
170
+ "min_period": 0.004,
171
+ "max_period": 4.0,
172
+ "rtc_config": null,
173
+ "compile_model": false,
174
+ "compile_mode": "max-autotune"
175
+ },
176
+ "output_dir": "/content/drive/MyDrive/smolvla_libero/train_main",
177
+ "job_name": "smolvla",
178
+ "resume": false,
179
+ "seed": 42,
180
+ "cudnn_deterministic": false,
181
+ "num_workers": 4,
182
+ "batch_size": 8,
183
+ "steps": 20000,
184
+ "eval_freq": 0,
185
+ "log_freq": 200,
186
+ "tolerance_s": 0.0001,
187
+ "save_checkpoint": true,
188
+ "save_freq": 2000,
189
+ "use_policy_training_preset": true,
190
+ "optimizer": {
191
+ "type": "adamw",
192
+ "lr": 0.0001,
193
+ "weight_decay": 1e-10,
194
+ "grad_clip_norm": 10,
195
+ "betas": [
196
+ 0.9,
197
+ 0.95
198
+ ],
199
+ "eps": 1e-08
200
+ },
201
+ "scheduler": {
202
+ "type": "cosine_decay_with_warmup",
203
+ "num_warmup_steps": 1000,
204
+ "num_decay_steps": 30000,
205
+ "peak_lr": 0.0001,
206
+ "decay_lr": 2.5e-06
207
+ },
208
+ "eval": {
209
+ "n_episodes": 50,
210
+ "batch_size": 50,
211
+ "use_async_envs": false
212
+ },
213
+ "wandb": {
214
+ "enable": true,
215
+ "disable_artifact": false,
216
+ "project": "VLA-Assessment",
217
+ "entity": null,
218
+ "notes": null,
219
+ "run_id": "4drqxcf5",
220
+ "mode": null,
221
+ "add_tags": true
222
+ },
223
+ "peft": null,
224
+ "use_rabc": false,
225
+ "rabc_progress_path": null,
226
+ "rabc_kappa": 0.01,
227
+ "rabc_epsilon": 1e-06,
228
+ "rabc_head_mode": "sparse",
229
+ "rename_map": {},
230
+ "checkpoint_path": null
231
+ }