Hugging Face's logo

astroyat
/
smolvla_pickwhiteobject

Safetensors
Model card Files
xet
Community
astroyat commited on
Commit
d369c85
·
verified ·
1 Parent(s): b775faf

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +83 -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 +71 -0
  6. policy_preprocessor_step_5_normalizer_processor.safetensors +3 -0
  7. train_config.json +219 -0
config.json ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "smolvla",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 9
9
+ ]
10
+ },
11
+ "observation.images.camera1": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 480,
16
+ 640
17
+ ]
18
+ }
19
+ },
20
+ "output_features": {
21
+ "action": {
22
+ "type": "ACTION",
23
+ "shape": [
24
+ 9
25
+ ]
26
+ }
27
+ },
28
+ "device": "cuda",
29
+ "use_amp": false,
30
+ "use_peft": false,
31
+ "push_to_hub": true,
32
+ "repo_id": "astroyat/smolvla_smolvla_pickwhiteobject",
33
+ "private": null,
34
+ "tags": null,
35
+ "license": null,
36
+ "pretrained_path": "lerobot/smolvla_base",
37
+ "chunk_size": 50,
38
+ "n_action_steps": 50,
39
+ "normalization_mapping": {
40
+ "VISUAL": "IDENTITY",
41
+ "STATE": "MEAN_STD",
42
+ "ACTION": "MEAN_STD"
43
+ },
44
+ "max_state_dim": 32,
45
+ "max_action_dim": 32,
46
+ "resize_imgs_with_padding": [
47
+ 512,
48
+ 512
49
+ ],
50
+ "empty_cameras": 0,
51
+ "adapt_to_pi_aloha": false,
52
+ "use_delta_joint_actions_aloha": false,
53
+ "tokenizer_max_length": 48,
54
+ "num_steps": 10,
55
+ "use_cache": true,
56
+ "freeze_vision_encoder": true,
57
+ "train_expert_only": true,
58
+ "train_state_proj": true,
59
+ "optimizer_lr": 0.0001,
60
+ "optimizer_betas": [
61
+ 0.9,
62
+ 0.95
63
+ ],
64
+ "optimizer_eps": 1e-08,
65
+ "optimizer_weight_decay": 1e-10,
66
+ "optimizer_grad_clip_norm": 10.0,
67
+ "scheduler_warmup_steps": 1000,
68
+ "scheduler_decay_steps": 30000,
69
+ "scheduler_decay_lr": 2.5e-06,
70
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
71
+ "load_vlm_weights": true,
72
+ "add_image_special_tokens": false,
73
+ "attention_mode": "cross_attn",
74
+ "prefix_length": 0,
75
+ "pad_language_to": "max_length",
76
+ "num_expert_layers": 0,
77
+ "num_vlm_layers": 16,
78
+ "self_attn_every_n_layers": 2,
79
+ "expert_width_multiplier": 0.75,
80
+ "min_period": 0.004,
81
+ "max_period": 4.0,
82
+ "rtc_config": null
83
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef7d49004014e1b61d459cca9597cc5f7f68d4db4cc1facdd668d5a8922b310b
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
+ 9
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 sha256:87ec83a189a1db8f954b15d0c6fc00481a93b3d0e68c71d10bfc0f97babdcf8a
3
+ size 6800
policy_preprocessor.json ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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.state": {
42
+ "type": "STATE",
43
+ "shape": [
44
+ 9
45
+ ]
46
+ },
47
+ "observation.images.camera1": {
48
+ "type": "VISUAL",
49
+ "shape": [
50
+ 3,
51
+ 480,
52
+ 640
53
+ ]
54
+ },
55
+ "action": {
56
+ "type": "ACTION",
57
+ "shape": [
58
+ 9
59
+ ]
60
+ }
61
+ },
62
+ "norm_map": {
63
+ "VISUAL": "IDENTITY",
64
+ "STATE": "MEAN_STD",
65
+ "ACTION": "MEAN_STD"
66
+ }
67
+ },
68
+ "state_file": "policy_preprocessor_step_5_normalizer_processor.safetensors"
69
+ }
70
+ ]
71
+ }
policy_preprocessor_step_5_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87ec83a189a1db8f954b15d0c6fc00481a93b3d0e68c71d10bfc0f97babdcf8a
3
+ size 6800
train_config.json ADDED
@@ -0,0 +1,219 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "astroyat/pickwhiteobject",
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": "pyav",
80
+ "streaming": false
81
+ },
82
+ "env": null,
83
+ "policy": {
84
+ "type": "smolvla",
85
+ "n_obs_steps": 1,
86
+ "input_features": {
87
+ "observation.state": {
88
+ "type": "STATE",
89
+ "shape": [
90
+ 9
91
+ ]
92
+ },
93
+ "observation.images.camera1": {
94
+ "type": "VISUAL",
95
+ "shape": [
96
+ 3,
97
+ 480,
98
+ 640
99
+ ]
100
+ }
101
+ },
102
+ "output_features": {
103
+ "action": {
104
+ "type": "ACTION",
105
+ "shape": [
106
+ 9
107
+ ]
108
+ }
109
+ },
110
+ "device": "cuda",
111
+ "use_amp": false,
112
+ "use_peft": false,
113
+ "push_to_hub": true,
114
+ "repo_id": "astroyat/smolvla_smolvla_pickwhiteobject",
115
+ "private": null,
116
+ "tags": null,
117
+ "license": null,
118
+ "pretrained_path": "lerobot/smolvla_base",
119
+ "chunk_size": 50,
120
+ "n_action_steps": 50,
121
+ "normalization_mapping": {
122
+ "VISUAL": "IDENTITY",
123
+ "STATE": "MEAN_STD",
124
+ "ACTION": "MEAN_STD"
125
+ },
126
+ "max_state_dim": 32,
127
+ "max_action_dim": 32,
128
+ "resize_imgs_with_padding": [
129
+ 512,
130
+ 512
131
+ ],
132
+ "empty_cameras": 0,
133
+ "adapt_to_pi_aloha": false,
134
+ "use_delta_joint_actions_aloha": false,
135
+ "tokenizer_max_length": 48,
136
+ "num_steps": 10,
137
+ "use_cache": true,
138
+ "freeze_vision_encoder": true,
139
+ "train_expert_only": true,
140
+ "train_state_proj": true,
141
+ "optimizer_lr": 0.0001,
142
+ "optimizer_betas": [
143
+ 0.9,
144
+ 0.95
145
+ ],
146
+ "optimizer_eps": 1e-08,
147
+ "optimizer_weight_decay": 1e-10,
148
+ "optimizer_grad_clip_norm": 10.0,
149
+ "scheduler_warmup_steps": 1000,
150
+ "scheduler_decay_steps": 30000,
151
+ "scheduler_decay_lr": 2.5e-06,
152
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
153
+ "load_vlm_weights": true,
154
+ "add_image_special_tokens": false,
155
+ "attention_mode": "cross_attn",
156
+ "prefix_length": 0,
157
+ "pad_language_to": "max_length",
158
+ "num_expert_layers": 0,
159
+ "num_vlm_layers": 16,
160
+ "self_attn_every_n_layers": 2,
161
+ "expert_width_multiplier": 0.75,
162
+ "min_period": 0.004,
163
+ "max_period": 4.0,
164
+ "rtc_config": null
165
+ },
166
+ "output_dir": "outputs/train/smolvla_pickwhiteobject",
167
+ "job_name": "smolvla_pickwhiteobject",
168
+ "resume": false,
169
+ "seed": 1000,
170
+ "num_workers": 4,
171
+ "batch_size": 64,
172
+ "steps": 20000,
173
+ "eval_freq": 20000,
174
+ "log_freq": 200,
175
+ "tolerance_s": 0.0001,
176
+ "save_checkpoint": true,
177
+ "save_freq": 20000,
178
+ "use_policy_training_preset": true,
179
+ "optimizer": {
180
+ "type": "adamw",
181
+ "lr": 0.0001,
182
+ "weight_decay": 1e-10,
183
+ "grad_clip_norm": 10.0,
184
+ "betas": [
185
+ 0.9,
186
+ 0.95
187
+ ],
188
+ "eps": 1e-08
189
+ },
190
+ "scheduler": {
191
+ "type": "cosine_decay_with_warmup",
192
+ "num_warmup_steps": 1000,
193
+ "num_decay_steps": 30000,
194
+ "peak_lr": 0.0001,
195
+ "decay_lr": 2.5e-06
196
+ },
197
+ "eval": {
198
+ "n_episodes": 50,
199
+ "batch_size": 50,
200
+ "use_async_envs": false
201
+ },
202
+ "wandb": {
203
+ "enable": false,
204
+ "disable_artifact": false,
205
+ "project": "lerobot",
206
+ "entity": null,
207
+ "notes": null,
208
+ "run_id": null,
209
+ "mode": null
210
+ },
211
+ "peft": null,
212
+ "use_rabc": false,
213
+ "rabc_progress_path": null,
214
+ "rabc_kappa": 0.01,
215
+ "rabc_epsilon": 1e-06,
216
+ "rabc_head_mode": "sparse",
217
+ "rename_map": {},
218
+ "checkpoint_path": null
219
+ }