Hugging Face's logo

SJun99
/
HD_task_smolvla_030000

Safetensors
Model card Files
xet
 Community
SJun99 commited on
Commit
514a80d
·
verified ·
1 Parent(s): deafc2f

Upload SmolVLA checkpoint at 30000 steps

Browse files
Files changed (7) hide show
  1. config.json +102 -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 +89 -0
  6. policy_preprocessor_step_5_normalizer_processor.safetensors +3 -0
  7. train_config.json +257 -0
config.json ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "smolvla",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 6
9
+ ]
10
+ },
11
+ "observation.images.camera1": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 256,
16
+ 256
17
+ ]
18
+ },
19
+ "observation.images.camera2": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 256,
24
+ 256
25
+ ]
26
+ },
27
+ "observation.images.camera3": {
28
+ "type": "VISUAL",
29
+ "shape": [
30
+ 3,
31
+ 256,
32
+ 256
33
+ ]
34
+ }
35
+ },
36
+ "output_features": {
37
+ "action": {
38
+ "type": "ACTION",
39
+ "shape": [
40
+ 6
41
+ ]
42
+ }
43
+ },
44
+ "device": "cuda",
45
+ "use_amp": true,
46
+ "use_peft": false,
47
+ "push_to_hub": true,
48
+ "repo_id": "SJun99/HD_task_smolvla",
49
+ "private": null,
50
+ "tags": null,
51
+ "license": null,
52
+ "pretrained_path": "lerobot/smolvla_base",
53
+ "pretrained_revision": null,
54
+ "chunk_size": 50,
55
+ "n_action_steps": 50,
56
+ "normalization_mapping": {
57
+ "VISUAL": "IDENTITY",
58
+ "STATE": "MEAN_STD",
59
+ "ACTION": "MEAN_STD"
60
+ },
61
+ "max_state_dim": 32,
62
+ "max_action_dim": 32,
63
+ "resize_imgs_with_padding": [
64
+ 512,
65
+ 512
66
+ ],
67
+ "empty_cameras": 0,
68
+ "adapt_to_pi_aloha": false,
69
+ "use_delta_joint_actions_aloha": false,
70
+ "tokenizer_max_length": 48,
71
+ "num_steps": 10,
72
+ "use_cache": true,
73
+ "freeze_vision_encoder": false,
74
+ "train_expert_only": true,
75
+ "train_state_proj": true,
76
+ "optimizer_lr": 0.0001,
77
+ "optimizer_betas": [
78
+ 0.9,
79
+ 0.95
80
+ ],
81
+ "optimizer_eps": 1e-08,
82
+ "optimizer_weight_decay": 1e-10,
83
+ "optimizer_grad_clip_norm": 10.0,
84
+ "scheduler_warmup_steps": 1000,
85
+ "scheduler_decay_steps": 150000,
86
+ "scheduler_decay_lr": 2.5e-06,
87
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
88
+ "load_vlm_weights": true,
89
+ "add_image_special_tokens": false,
90
+ "attention_mode": "cross_attn",
91
+ "prefix_length": 0,
92
+ "pad_language_to": "max_length",
93
+ "num_expert_layers": 0,
94
+ "num_vlm_layers": 16,
95
+ "self_attn_every_n_layers": 2,
96
+ "expert_width_multiplier": 0.75,
97
+ "min_period": 0.004,
98
+ "max_period": 4.0,
99
+ "rtc_config": null,
100
+ "compile_model": false,
101
+ "compile_mode": "max-autotune"
102
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e9cd76fe34c8c6b8c6fda056df6d2e712bd41105d167ab75a1c87be3ed919484
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
+ 6
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 sha256:02669fab8ac6c2b4197d06718b60c34e22ee37193f09752c7ba03943f3e0b321
3
+ size 6544
policy_preprocessor.json ADDED
@@ -0,0 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "policy_preprocessor",
3
+ "steps": [
4
+ {
5
+ "registry_name": "rename_observations_processor",
6
+ "config": {
7
+ "rename_map": {
8
+ "observation.images.top": "observation.images.camera1"
9
+ }
10
+ }
11
+ },
12
+ {
13
+ "registry_name": "to_batch_processor",
14
+ "config": {}
15
+ },
16
+ {
17
+ "registry_name": "smolvla_new_line_processor",
18
+ "config": {}
19
+ },
20
+ {
21
+ "registry_name": "tokenizer_processor",
22
+ "config": {
23
+ "max_length": 48,
24
+ "task_key": "task",
25
+ "padding_side": "right",
26
+ "padding": "max_length",
27
+ "truncation": true,
28
+ "tokenizer_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct"
29
+ }
30
+ },
31
+ {
32
+ "registry_name": "device_processor",
33
+ "config": {
34
+ "device": "cuda",
35
+ "float_dtype": null
36
+ }
37
+ },
38
+ {
39
+ "registry_name": "normalizer_processor",
40
+ "config": {
41
+ "eps": 1e-08,
42
+ "features": {
43
+ "observation.state": {
44
+ "type": "STATE",
45
+ "shape": [
46
+ 6
47
+ ]
48
+ },
49
+ "observation.images.camera1": {
50
+ "type": "VISUAL",
51
+ "shape": [
52
+ 3,
53
+ 256,
54
+ 256
55
+ ]
56
+ },
57
+ "observation.images.camera2": {
58
+ "type": "VISUAL",
59
+ "shape": [
60
+ 3,
61
+ 256,
62
+ 256
63
+ ]
64
+ },
65
+ "observation.images.camera3": {
66
+ "type": "VISUAL",
67
+ "shape": [
68
+ 3,
69
+ 256,
70
+ 256
71
+ ]
72
+ },
73
+ "action": {
74
+ "type": "ACTION",
75
+ "shape": [
76
+ 6
77
+ ]
78
+ }
79
+ },
80
+ "norm_map": {
81
+ "VISUAL": "IDENTITY",
82
+ "STATE": "MEAN_STD",
83
+ "ACTION": "MEAN_STD"
84
+ }
85
+ },
86
+ "state_file": "policy_preprocessor_step_5_normalizer_processor.safetensors"
87
+ }
88
+ ]
89
+ }
policy_preprocessor_step_5_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02669fab8ac6c2b4197d06718b60c34e22ee37193f09752c7ba03943f3e0b321
3
+ size 6544
train_config.json ADDED
@@ -0,0 +1,257 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "SJun99/HD_task",
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
+ "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
+ 6
94
+ ]
95
+ },
96
+ "observation.images.camera1": {
97
+ "type": "VISUAL",
98
+ "shape": [
99
+ 3,
100
+ 256,
101
+ 256
102
+ ]
103
+ },
104
+ "observation.images.camera2": {
105
+ "type": "VISUAL",
106
+ "shape": [
107
+ 3,
108
+ 256,
109
+ 256
110
+ ]
111
+ },
112
+ "observation.images.camera3": {
113
+ "type": "VISUAL",
114
+ "shape": [
115
+ 3,
116
+ 256,
117
+ 256
118
+ ]
119
+ }
120
+ },
121
+ "output_features": {
122
+ "action": {
123
+ "type": "ACTION",
124
+ "shape": [
125
+ 6
126
+ ]
127
+ }
128
+ },
129
+ "device": "cuda",
130
+ "use_amp": true,
131
+ "use_peft": false,
132
+ "push_to_hub": true,
133
+ "repo_id": "SJun99/HD_task_smolvla",
134
+ "private": null,
135
+ "tags": null,
136
+ "license": null,
137
+ "pretrained_path": "lerobot/smolvla_base",
138
+ "pretrained_revision": null,
139
+ "chunk_size": 50,
140
+ "n_action_steps": 50,
141
+ "normalization_mapping": {
142
+ "VISUAL": "IDENTITY",
143
+ "STATE": "MEAN_STD",
144
+ "ACTION": "MEAN_STD"
145
+ },
146
+ "max_state_dim": 32,
147
+ "max_action_dim": 32,
148
+ "resize_imgs_with_padding": [
149
+ 512,
150
+ 512
151
+ ],
152
+ "empty_cameras": 0,
153
+ "adapt_to_pi_aloha": false,
154
+ "use_delta_joint_actions_aloha": false,
155
+ "tokenizer_max_length": 48,
156
+ "num_steps": 10,
157
+ "use_cache": true,
158
+ "freeze_vision_encoder": false,
159
+ "train_expert_only": true,
160
+ "train_state_proj": true,
161
+ "optimizer_lr": 0.0001,
162
+ "optimizer_betas": [
163
+ 0.9,
164
+ 0.95
165
+ ],
166
+ "optimizer_eps": 1e-08,
167
+ "optimizer_weight_decay": 1e-10,
168
+ "optimizer_grad_clip_norm": 10.0,
169
+ "scheduler_warmup_steps": 1000,
170
+ "scheduler_decay_steps": 150000,
171
+ "scheduler_decay_lr": 2.5e-06,
172
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
173
+ "load_vlm_weights": true,
174
+ "add_image_special_tokens": false,
175
+ "attention_mode": "cross_attn",
176
+ "prefix_length": 0,
177
+ "pad_language_to": "max_length",
178
+ "num_expert_layers": 0,
179
+ "num_vlm_layers": 16,
180
+ "self_attn_every_n_layers": 2,
181
+ "expert_width_multiplier": 0.75,
182
+ "min_period": 0.004,
183
+ "max_period": 4.0,
184
+ "rtc_config": null,
185
+ "compile_model": false,
186
+ "compile_mode": "max-autotune"
187
+ },
188
+ "reward_model": null,
189
+ "output_dir": "outputs/train/smolvla_so101/HD_task_smolvla",
190
+ "job_name": "HD_task_smolvla",
191
+ "resume": false,
192
+ "seed": 1000,
193
+ "cudnn_deterministic": false,
194
+ "num_workers": 8,
195
+ "batch_size": 16,
196
+ "prefetch_factor": 4,
197
+ "persistent_workers": true,
198
+ "steps": 150000,
199
+ "env_eval_freq": 20000,
200
+ "log_freq": 200,
201
+ "eval_steps": 0,
202
+ "max_eval_samples": 0,
203
+ "tolerance_s": 0.05,
204
+ "save_checkpoint": true,
205
+ "save_freq": 30000,
206
+ "use_policy_training_preset": true,
207
+ "optimizer": {
208
+ "type": "adamw",
209
+ "lr": 0.0001,
210
+ "weight_decay": 1e-10,
211
+ "grad_clip_norm": 10.0,
212
+ "betas": [
213
+ 0.9,
214
+ 0.95
215
+ ],
216
+ "eps": 1e-08
217
+ },
218
+ "scheduler": {
219
+ "type": "cosine_decay_with_warmup",
220
+ "num_warmup_steps": 1000,
221
+ "num_decay_steps": 150000,
222
+ "peak_lr": 0.0001,
223
+ "decay_lr": 2.5e-06
224
+ },
225
+ "eval": {
226
+ "n_episodes": 50,
227
+ "batch_size": 16,
228
+ "use_async_envs": true,
229
+ "recording": false,
230
+ "recording_repo_id": null,
231
+ "recording_private": false
232
+ },
233
+ "wandb": {
234
+ "enable": false,
235
+ "disable_artifact": false,
236
+ "project": "lerobot",
237
+ "entity": null,
238
+ "notes": null,
239
+ "run_id": null,
240
+ "mode": null,
241
+ "add_tags": true
242
+ },
243
+ "peft": null,
244
+ "job": {
245
+ "target": null,
246
+ "image": "huggingface/lerobot-gpu:latest",
247
+ "timeout": "2d",
248
+ "detach": false,
249
+ "tags": []
250
+ },
251
+ "save_checkpoint_to_hub": false,
252
+ "sample_weighting": null,
253
+ "rename_map": {
254
+ "observation.images.top": "observation.images.camera1"
255
+ },
256
+ "checkpoint_path": null
257
+ }