Hugging Face's logo

wyn0408
/
test_smolvla_2_camera

Safetensors
Model card Files
xet
Community
wyn0408 commited on
Commit
e2a9a28
·
verified ·
1 Parent(s): b8fdbb2

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +90 -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 +219 -0
config.json ADDED
@@ -0,0 +1,90 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "smolvla",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 7
9
+ ]
10
+ },
11
+ "observation.images.camera_in_hand": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 480,
16
+ 640
17
+ ]
18
+ },
19
+ "observation.images.camera_middle": {
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
+ 7
33
+ ]
34
+ }
35
+ },
36
+ "device": "cuda",
37
+ "use_amp": false,
38
+ "push_to_hub": false,
39
+ "repo_id": null,
40
+ "private": null,
41
+ "tags": null,
42
+ "license": null,
43
+ "pretrained_path": null,
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,
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": false,
79
+ "add_image_special_tokens": false,
80
+ "attention_mode": "cross_attn",
81
+ "prefix_length": -1,
82
+ "pad_language_to": "longest",
83
+ "num_expert_layers": -1,
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
+ "rtc_config": null
90
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae78c61c075ac59291cab2260666b090157b377685f143b30d802c4e564bd73f
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 sha256:5403211222eb329a72b7a8d669eeccb20ce8fdd1fb88659b77c3ac4aa9012983
3
+ size 7792
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": "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
+ 7
45
+ ]
46
+ },
47
+ "observation.images.camera_in_hand": {
48
+ "type": "VISUAL",
49
+ "shape": [
50
+ 3,
51
+ 480,
52
+ 640
53
+ ]
54
+ },
55
+ "observation.images.camera_middle": {
56
+ "type": "VISUAL",
57
+ "shape": [
58
+ 3,
59
+ 480,
60
+ 640
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 sha256:5403211222eb329a72b7a8d669eeccb20ce8fdd1fb88659b77c3ac4aa9012983
3
+ size 7792
train_config.json ADDED
@@ -0,0 +1,219 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "wyn0408/test_smolvla_pick_apple_2_camera",
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.state": {
88
+ "type": "STATE",
89
+ "shape": [
90
+ 7
91
+ ]
92
+ },
93
+ "observation.images.camera_in_hand": {
94
+ "type": "VISUAL",
95
+ "shape": [
96
+ 3,
97
+ 480,
98
+ 640
99
+ ]
100
+ },
101
+ "observation.images.camera_middle": {
102
+ "type": "VISUAL",
103
+ "shape": [
104
+ 3,
105
+ 480,
106
+ 640
107
+ ]
108
+ }
109
+ },
110
+ "output_features": {
111
+ "action": {
112
+ "type": "ACTION",
113
+ "shape": [
114
+ 7
115
+ ]
116
+ }
117
+ },
118
+ "device": "cuda",
119
+ "use_amp": false,
120
+ "push_to_hub": false,
121
+ "repo_id": null,
122
+ "private": null,
123
+ "tags": null,
124
+ "license": null,
125
+ "pretrained_path": null,
126
+ "chunk_size": 50,
127
+ "n_action_steps": 50,
128
+ "normalization_mapping": {
129
+ "VISUAL": "IDENTITY",
130
+ "STATE": "MEAN_STD",
131
+ "ACTION": "MEAN_STD"
132
+ },
133
+ "max_state_dim": 32,
134
+ "max_action_dim": 32,
135
+ "resize_imgs_with_padding": [
136
+ 512,
137
+ 512
138
+ ],
139
+ "empty_cameras": 0,
140
+ "adapt_to_pi_aloha": false,
141
+ "use_delta_joint_actions_aloha": false,
142
+ "tokenizer_max_length": 48,
143
+ "num_steps": 10,
144
+ "use_cache": true,
145
+ "freeze_vision_encoder": true,
146
+ "train_expert_only": true,
147
+ "train_state_proj": true,
148
+ "optimizer_lr": 0.0001,
149
+ "optimizer_betas": [
150
+ 0.9,
151
+ 0.95
152
+ ],
153
+ "optimizer_eps": 1e-08,
154
+ "optimizer_weight_decay": 1e-10,
155
+ "optimizer_grad_clip_norm": 10,
156
+ "scheduler_warmup_steps": 1000,
157
+ "scheduler_decay_steps": 30000,
158
+ "scheduler_decay_lr": 2.5e-06,
159
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
160
+ "load_vlm_weights": false,
161
+ "add_image_special_tokens": false,
162
+ "attention_mode": "cross_attn",
163
+ "prefix_length": -1,
164
+ "pad_language_to": "longest",
165
+ "num_expert_layers": -1,
166
+ "num_vlm_layers": 16,
167
+ "self_attn_every_n_layers": 2,
168
+ "expert_width_multiplier": 0.75,
169
+ "min_period": 0.004,
170
+ "max_period": 4.0,
171
+ "rtc_config": null
172
+ },
173
+ "output_dir": "outputs/train/my_smolvla_v2_2cam",
174
+ "job_name": "my_smolvla_training_v2_2cam",
175
+ "resume": false,
176
+ "seed": 1000,
177
+ "num_workers": 4,
178
+ "batch_size": 64,
179
+ "steps": 200000,
180
+ "eval_freq": 20000,
181
+ "log_freq": 200,
182
+ "save_checkpoint": true,
183
+ "save_freq": 10000,
184
+ "use_policy_training_preset": true,
185
+ "optimizer": {
186
+ "type": "adamw",
187
+ "lr": 0.0001,
188
+ "weight_decay": 1e-10,
189
+ "grad_clip_norm": 10,
190
+ "betas": [
191
+ 0.9,
192
+ 0.95
193
+ ],
194
+ "eps": 1e-08
195
+ },
196
+ "scheduler": {
197
+ "type": "cosine_decay_with_warmup",
198
+ "num_warmup_steps": 1000,
199
+ "num_decay_steps": 30000,
200
+ "peak_lr": 0.0001,
201
+ "decay_lr": 2.5e-06
202
+ },
203
+ "eval": {
204
+ "n_episodes": 50,
205
+ "batch_size": 50,
206
+ "use_async_envs": false
207
+ },
208
+ "wandb": {
209
+ "enable": true,
210
+ "disable_artifact": false,
211
+ "project": "lerobot",
212
+ "entity": null,
213
+ "notes": null,
214
+ "run_id": "jpguvzq9",
215
+ "mode": null
216
+ },
217
+ "checkpoint_path": null,
218
+ "rename_map": {}
219
+ }