Hugging Face's logo

brucekimrok
/
push-button-from-base-015000

Safetensors
Model card Files
xet
Community
brucekimrok commited on
Commit
d7e5d32
·
verified ·
1 Parent(s): 9661358

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +89 -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 +202 -0
config.json ADDED
@@ -0,0 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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.top": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 480,
16
+ 640
17
+ ]
18
+ },
19
+ "observation.images.wrist": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 480,
24
+ 640
25
+ ]
26
+ }
27
+ },
28
+ "output_features": {
29
+ "action": {
30
+ "type": "ACTION",
31
+ "shape": [
32
+ 6
33
+ ]
34
+ }
35
+ },
36
+ "device": "cuda",
37
+ "use_amp": false,
38
+ "push_to_hub": true,
39
+ "repo_id": "brucekimrok/push-button_FROM_base",
40
+ "private": null,
41
+ "tags": null,
42
+ "license": null,
43
+ "pretrained_path": "lerobot/smolvla_base",
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": 10.0,
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": 0,
82
+ "pad_language_to": "max_length",
83
+ "num_expert_layers": 0,
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
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4c98a98d0ef126ce444f92b01732eaf8a0acf2cb17556d015a8f487803fcc45
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 sha256:7de8f0e80561b3f3835f4484ee3cd7e695eb5391d0f34c004b5e9a8334bd1be4
3
+ size 7536
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.state": {
42
+ "type": "STATE",
43
+ "shape": [
44
+ 6
45
+ ]
46
+ },
47
+ "observation.images.top": {
48
+ "type": "VISUAL",
49
+ "shape": [
50
+ 3,
51
+ 480,
52
+ 640
53
+ ]
54
+ },
55
+ "observation.images.wrist": {
56
+ "type": "VISUAL",
57
+ "shape": [
58
+ 3,
59
+ 480,
60
+ 640
61
+ ]
62
+ },
63
+ "action": {
64
+ "type": "ACTION",
65
+ "shape": [
66
+ 6
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 sha256:7de8f0e80561b3f3835f4484ee3cd7e695eb5391d0f34c004b5e9a8334bd1be4
3
+ size 7536
train_config.json ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "jskim/record-push-button-merged-01-05",
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
+ }
62
+ },
63
+ "revision": null,
64
+ "use_imagenet_stats": true,
65
+ "video_backend": "torchcodec",
66
+ "streaming": false
67
+ },
68
+ "env": null,
69
+ "policy": {
70
+ "type": "smolvla",
71
+ "n_obs_steps": 1,
72
+ "input_features": {
73
+ "observation.state": {
74
+ "type": "STATE",
75
+ "shape": [
76
+ 6
77
+ ]
78
+ },
79
+ "observation.images.top": {
80
+ "type": "VISUAL",
81
+ "shape": [
82
+ 3,
83
+ 480,
84
+ 640
85
+ ]
86
+ },
87
+ "observation.images.wrist": {
88
+ "type": "VISUAL",
89
+ "shape": [
90
+ 3,
91
+ 480,
92
+ 640
93
+ ]
94
+ }
95
+ },
96
+ "output_features": {
97
+ "action": {
98
+ "type": "ACTION",
99
+ "shape": [
100
+ 6
101
+ ]
102
+ }
103
+ },
104
+ "device": "cuda",
105
+ "use_amp": false,
106
+ "push_to_hub": true,
107
+ "repo_id": "brucekimrok/push-button_FROM_base",
108
+ "private": null,
109
+ "tags": null,
110
+ "license": null,
111
+ "pretrained_path": "lerobot/smolvla_base",
112
+ "chunk_size": 50,
113
+ "n_action_steps": 50,
114
+ "normalization_mapping": {
115
+ "VISUAL": "IDENTITY",
116
+ "STATE": "MEAN_STD",
117
+ "ACTION": "MEAN_STD"
118
+ },
119
+ "max_state_dim": 32,
120
+ "max_action_dim": 32,
121
+ "resize_imgs_with_padding": [
122
+ 512,
123
+ 512
124
+ ],
125
+ "empty_cameras": 0,
126
+ "adapt_to_pi_aloha": false,
127
+ "use_delta_joint_actions_aloha": false,
128
+ "tokenizer_max_length": 48,
129
+ "num_steps": 10,
130
+ "use_cache": true,
131
+ "freeze_vision_encoder": true,
132
+ "train_expert_only": true,
133
+ "train_state_proj": true,
134
+ "optimizer_lr": 0.0001,
135
+ "optimizer_betas": [
136
+ 0.9,
137
+ 0.95
138
+ ],
139
+ "optimizer_eps": 1e-08,
140
+ "optimizer_weight_decay": 1e-10,
141
+ "optimizer_grad_clip_norm": 10.0,
142
+ "scheduler_warmup_steps": 1000,
143
+ "scheduler_decay_steps": 30000,
144
+ "scheduler_decay_lr": 2.5e-06,
145
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
146
+ "load_vlm_weights": true,
147
+ "add_image_special_tokens": false,
148
+ "attention_mode": "cross_attn",
149
+ "prefix_length": 0,
150
+ "pad_language_to": "max_length",
151
+ "num_expert_layers": 0,
152
+ "num_vlm_layers": 16,
153
+ "self_attn_every_n_layers": 2,
154
+ "expert_width_multiplier": 0.75,
155
+ "min_period": 0.004,
156
+ "max_period": 4.0
157
+ },
158
+ "output_dir": "outputs/train/push-button-from-base",
159
+ "job_name": "push-button-from-smolvla",
160
+ "resume": false,
161
+ "seed": 1000,
162
+ "num_workers": 4,
163
+ "batch_size": 16,
164
+ "steps": 20000,
165
+ "eval_freq": 20000,
166
+ "log_freq": 200,
167
+ "save_checkpoint": true,
168
+ "save_freq": 1000,
169
+ "use_policy_training_preset": true,
170
+ "optimizer": {
171
+ "type": "adamw",
172
+ "lr": 0.0001,
173
+ "weight_decay": 1e-10,
174
+ "grad_clip_norm": 10.0,
175
+ "betas": [
176
+ 0.9,
177
+ 0.95
178
+ ],
179
+ "eps": 1e-08
180
+ },
181
+ "scheduler": {
182
+ "type": "cosine_decay_with_warmup",
183
+ "num_warmup_steps": 1000,
184
+ "num_decay_steps": 30000,
185
+ "peak_lr": 0.0001,
186
+ "decay_lr": 2.5e-06
187
+ },
188
+ "eval": {
189
+ "n_episodes": 50,
190
+ "batch_size": 50,
191
+ "use_async_envs": false
192
+ },
193
+ "wandb": {
194
+ "enable": true,
195
+ "disable_artifact": false,
196
+ "project": "lerobot",
197
+ "entity": null,
198
+ "notes": null,
199
+ "run_id": "8o1fdcoo",
200
+ "mode": null
201
+ }
202
+ }