Hugging Face's logo

ThiennNguyen
/
smolvla_candy

Robotics
LeRobot
Safetensors
smolvla
Model card Files
xet
 Community
ThiennNguyen commited on
Commit
591defe
·
verified ·
1 Parent(s): c5ec3ec

Upload policy

Browse files
Files changed (2) hide show
  1. config.json +118 -0
  2. model.safetensors +3 -0
config.json ADDED
@@ -0,0 +1,118 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "observation.images.empty_camera_0": {
36
+ "type": "VISUAL",
37
+ "shape": [
38
+ 3,
39
+ 480,
40
+ 640
41
+ ]
42
+ },
43
+ "observation.images.empty_camera_1": {
44
+ "type": "VISUAL",
45
+ "shape": [
46
+ 3,
47
+ 480,
48
+ 640
49
+ ]
50
+ }
51
+ },
52
+ "output_features": {
53
+ "action": {
54
+ "type": "ACTION",
55
+ "shape": [
56
+ 6
57
+ ]
58
+ }
59
+ },
60
+ "device": "cuda",
61
+ "use_amp": false,
62
+ "use_peft": false,
63
+ "push_to_hub": true,
64
+ "repo_id": "ThiennNguyen/smolvla_candy",
65
+ "private": null,
66
+ "tags": null,
67
+ "license": null,
68
+ "pretrained_path": "lerobot/smolvla_base",
69
+ "pretrained_revision": null,
70
+ "chunk_size": 50,
71
+ "n_action_steps": 50,
72
+ "normalization_mapping": {
73
+ "VISUAL": "IDENTITY",
74
+ "STATE": "MEAN_STD",
75
+ "ACTION": "MEAN_STD"
76
+ },
77
+ "max_state_dim": 32,
78
+ "max_action_dim": 32,
79
+ "resize_imgs_with_padding": [
80
+ 512,
81
+ 512
82
+ ],
83
+ "empty_cameras": 2,
84
+ "adapt_to_pi_aloha": false,
85
+ "use_delta_joint_actions_aloha": false,
86
+ "tokenizer_max_length": 48,
87
+ "num_steps": 10,
88
+ "use_cache": true,
89
+ "freeze_vision_encoder": true,
90
+ "train_expert_only": true,
91
+ "train_state_proj": true,
92
+ "optimizer_lr": 0.0001,
93
+ "optimizer_betas": [
94
+ 0.9,
95
+ 0.95
96
+ ],
97
+ "optimizer_eps": 1e-08,
98
+ "optimizer_weight_decay": 1e-10,
99
+ "optimizer_grad_clip_norm": 10.0,
100
+ "scheduler_warmup_steps": 1000,
101
+ "scheduler_decay_steps": 30000,
102
+ "scheduler_decay_lr": 2.5e-06,
103
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
104
+ "load_vlm_weights": true,
105
+ "add_image_special_tokens": false,
106
+ "attention_mode": "cross_attn",
107
+ "prefix_length": 0,
108
+ "pad_language_to": "max_length",
109
+ "num_expert_layers": 0,
110
+ "num_vlm_layers": 16,
111
+ "self_attn_every_n_layers": 2,
112
+ "expert_width_multiplier": 0.75,
113
+ "min_period": 0.004,
114
+ "max_period": 4.0,
115
+ "rtc_config": null,
116
+ "compile_model": false,
117
+ "compile_mode": "max-autotune"
118
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:75ed0cba05b44d63000083d8e76240ac3f2f1d1b27e4c9333c791fb106cc14a4
3
+ size 906712552