Hugging Face's logo

robot-learning-group47
/
eval1_clean_12k

Safetensors
Model card Files
xet
 Community
nkim7 commited on
Commit
ec5e4c8
·
verified ·
1 Parent(s): 40a5f78

Upload Eval 1 clean SmolVLA 12k checkpoint

Browse files
Files changed (7) hide show
  1. config.json +101 -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 +187 -0
config.json ADDED
@@ -0,0 +1,101 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "robot-learning-group47/eval1_clean_256_chunk50_lightingonly_smolvla_v1",
49
+ "private": null,
50
+ "tags": null,
51
+ "license": null,
52
+ "pretrained_path": "lerobot/smolvla_base",
53
+ "chunk_size": 50,
54
+ "n_action_steps": 50,
55
+ "normalization_mapping": {
56
+ "VISUAL": "IDENTITY",
57
+ "STATE": "MEAN_STD",
58
+ "ACTION": "MEAN_STD"
59
+ },
60
+ "max_state_dim": 32,
61
+ "max_action_dim": 32,
62
+ "resize_imgs_with_padding": [
63
+ 512,
64
+ 512
65
+ ],
66
+ "empty_cameras": 0,
67
+ "adapt_to_pi_aloha": false,
68
+ "use_delta_joint_actions_aloha": false,
69
+ "tokenizer_max_length": 48,
70
+ "num_steps": 10,
71
+ "use_cache": true,
72
+ "freeze_vision_encoder": true,
73
+ "train_expert_only": true,
74
+ "train_state_proj": true,
75
+ "optimizer_lr": 0.0001,
76
+ "optimizer_betas": [
77
+ 0.9,
78
+ 0.95
79
+ ],
80
+ "optimizer_eps": 1e-08,
81
+ "optimizer_weight_decay": 1e-10,
82
+ "optimizer_grad_clip_norm": 10.0,
83
+ "scheduler_warmup_steps": 1000,
84
+ "scheduler_decay_steps": 30000,
85
+ "scheduler_decay_lr": 2.5e-06,
86
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
87
+ "load_vlm_weights": true,
88
+ "add_image_special_tokens": false,
89
+ "attention_mode": "cross_attn",
90
+ "prefix_length": 0,
91
+ "pad_language_to": "max_length",
92
+ "num_expert_layers": 0,
93
+ "num_vlm_layers": 16,
94
+ "self_attn_every_n_layers": 2,
95
+ "expert_width_multiplier": 0.75,
96
+ "min_period": 0.004,
97
+ "max_period": 4.0,
98
+ "rtc_config": null,
99
+ "compile_model": false,
100
+ "compile_mode": "max-autotune"
101
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6fe29522924130df28b61c6e07f94a4630c0e26d3abc224965f7bac8814b3f94
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:b6c6e305a0fcae663f0410b48e17c01225658a454224d086b38fe0887cda923c
3
+ size 6560
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.front": "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:b6c6e305a0fcae663f0410b48e17c01225658a454224d086b38fe0887cda923c
3
+ size 6560
train_config.json ADDED
@@ -0,0 +1,187 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "robot-learning-group47/eval1_clean",
4
+ "root": null,
5
+ "episodes": null,
6
+ "image_transforms": {
7
+ "enable": true,
8
+ "max_num_transforms": 1,
9
+ "random_order": false,
10
+ "tfs": {
11
+ "brightness": {
12
+ "weight": 1.0,
13
+ "type": "ColorJitter",
14
+ "kwargs": {
15
+ "brightness": [
16
+ 0.85,
17
+ 1.15
18
+ ]
19
+ }
20
+ }
21
+ }
22
+ },
23
+ "revision": null,
24
+ "use_imagenet_stats": true,
25
+ "video_backend": "torchcodec",
26
+ "return_uint8": false,
27
+ "streaming": false
28
+ },
29
+ "env": null,
30
+ "policy": {
31
+ "type": "smolvla",
32
+ "n_obs_steps": 1,
33
+ "input_features": {
34
+ "observation.state": {
35
+ "type": "STATE",
36
+ "shape": [
37
+ 6
38
+ ]
39
+ },
40
+ "observation.images.camera1": {
41
+ "type": "VISUAL",
42
+ "shape": [
43
+ 3,
44
+ 256,
45
+ 256
46
+ ]
47
+ },
48
+ "observation.images.camera2": {
49
+ "type": "VISUAL",
50
+ "shape": [
51
+ 3,
52
+ 256,
53
+ 256
54
+ ]
55
+ },
56
+ "observation.images.camera3": {
57
+ "type": "VISUAL",
58
+ "shape": [
59
+ 3,
60
+ 256,
61
+ 256
62
+ ]
63
+ }
64
+ },
65
+ "output_features": {
66
+ "action": {
67
+ "type": "ACTION",
68
+ "shape": [
69
+ 6
70
+ ]
71
+ }
72
+ },
73
+ "device": "cuda",
74
+ "use_amp": true,
75
+ "use_peft": false,
76
+ "push_to_hub": true,
77
+ "repo_id": "robot-learning-group47/eval1_clean_256_chunk50_lightingonly_smolvla_v1",
78
+ "private": null,
79
+ "tags": null,
80
+ "license": null,
81
+ "pretrained_path": "lerobot/smolvla_base",
82
+ "chunk_size": 50,
83
+ "n_action_steps": 50,
84
+ "normalization_mapping": {
85
+ "VISUAL": "IDENTITY",
86
+ "STATE": "MEAN_STD",
87
+ "ACTION": "MEAN_STD"
88
+ },
89
+ "max_state_dim": 32,
90
+ "max_action_dim": 32,
91
+ "resize_imgs_with_padding": [
92
+ 512,
93
+ 512
94
+ ],
95
+ "empty_cameras": 0,
96
+ "adapt_to_pi_aloha": false,
97
+ "use_delta_joint_actions_aloha": false,
98
+ "tokenizer_max_length": 48,
99
+ "num_steps": 10,
100
+ "use_cache": true,
101
+ "freeze_vision_encoder": true,
102
+ "train_expert_only": true,
103
+ "train_state_proj": true,
104
+ "optimizer_lr": 0.0001,
105
+ "optimizer_betas": [
106
+ 0.9,
107
+ 0.95
108
+ ],
109
+ "optimizer_eps": 1e-08,
110
+ "optimizer_weight_decay": 1e-10,
111
+ "optimizer_grad_clip_norm": 10.0,
112
+ "scheduler_warmup_steps": 1000,
113
+ "scheduler_decay_steps": 30000,
114
+ "scheduler_decay_lr": 2.5e-06,
115
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
116
+ "load_vlm_weights": true,
117
+ "add_image_special_tokens": false,
118
+ "attention_mode": "cross_attn",
119
+ "prefix_length": 0,
120
+ "pad_language_to": "max_length",
121
+ "num_expert_layers": 0,
122
+ "num_vlm_layers": 16,
123
+ "self_attn_every_n_layers": 2,
124
+ "expert_width_multiplier": 0.75,
125
+ "min_period": 0.004,
126
+ "max_period": 4.0,
127
+ "rtc_config": null,
128
+ "compile_model": false,
129
+ "compile_mode": "max-autotune"
130
+ },
131
+ "reward_model": null,
132
+ "output_dir": "outputs/train/eval1_clean_256_chunk50_lightingonly_smolvla_v1",
133
+ "job_name": "eval1_clean_256_chunk50_lightingonly_smolvla_v1",
134
+ "resume": false,
135
+ "seed": 1000,
136
+ "cudnn_deterministic": false,
137
+ "num_workers": 4,
138
+ "batch_size": 64,
139
+ "prefetch_factor": 4,
140
+ "persistent_workers": true,
141
+ "steps": 20000,
142
+ "eval_freq": 2000,
143
+ "log_freq": 200,
144
+ "tolerance_s": 0.0001,
145
+ "save_checkpoint": true,
146
+ "save_freq": 2000,
147
+ "use_policy_training_preset": true,
148
+ "optimizer": {
149
+ "type": "adamw",
150
+ "lr": 0.0001,
151
+ "weight_decay": 1e-10,
152
+ "grad_clip_norm": 10.0,
153
+ "betas": [
154
+ 0.9,
155
+ 0.95
156
+ ],
157
+ "eps": 1e-08
158
+ },
159
+ "scheduler": {
160
+ "type": "cosine_decay_with_warmup",
161
+ "num_warmup_steps": 1000,
162
+ "num_decay_steps": 30000,
163
+ "peak_lr": 0.0001,
164
+ "decay_lr": 2.5e-06
165
+ },
166
+ "eval": {
167
+ "n_episodes": 50,
168
+ "batch_size": 8,
169
+ "use_async_envs": true
170
+ },
171
+ "wandb": {
172
+ "enable": true,
173
+ "disable_artifact": false,
174
+ "project": "lerobot",
175
+ "entity": null,
176
+ "notes": null,
177
+ "run_id": "4dc1p4jz",
178
+ "mode": null,
179
+ "add_tags": true
180
+ },
181
+ "peft": null,
182
+ "sample_weighting": null,
183
+ "rename_map": {
184
+ "observation.images.front": "observation.images.camera1"
185
+ },
186
+ "checkpoint_path": null
187
+ }