Hugging Face's logo

Odog16
/
block_sorting_SmolVLA_8k

Safetensors
Model card Files
xet
 Community
Odog16 commited on
Commit
73f20f6
·
verified ·
1 Parent(s): 8331f55

Upload config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +95 -96
config.json CHANGED
@@ -1,101 +1,100 @@
1
  {
2
- "type": "smolvla",
3
- "n_obs_steps": 1,
4
- "input_features": {
5
- "observation.state": {
6
- "type": "STATE",
7
- "shape": [
8
- 18
9
- ]
10
- },
11
- "observation.images.head": {
12
- "type": "VISUAL",
13
- "shape": [
14
- 3,
15
- 360,
16
- 640
17
- ]
18
- },
19
- "observation.images.left_wrist": {
20
- "type": "VISUAL",
21
- "shape": [
22
- 3,
23
- 360,
24
- 640
25
- ]
26
- },
27
- "observation.images.right_wrist": {
28
- "type": "VISUAL",
29
- "shape": [
30
- 3,
31
- 360,
32
- 640
33
- ]
34
- }
35
  },
36
- "output_features": {
37
- "action": {
38
- "type": "ACTION",
39
- "shape": [
40
- 18
41
- ]
42
- }
43
  },
44
- "device": "cuda",
45
- "use_amp": true,
46
- "use_peft": false,
47
- "push_to_hub": true,
48
- "repo_id": "Odog16/block_sorting_SmolVLA_policy",
49
- "private": null,
50
- "tags": null,
51
- "license": null,
52
- "pretrained_path": "lerobot/smolvla_base",
53
- "chunk_size": 30,
54
- "n_action_steps": 30,
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,
83
- "scheduler_warmup_steps": 500,
84
- "scheduler_decay_steps": 18000,
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": -1,
91
- "pad_language_to": "longest",
92
- "num_expert_layers": -1,
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
  }
 
1
  {
2
+ "n_obs_steps": 1,
3
+ "input_features": {
4
+ "observation.state": {
5
+ "type": "STATE",
6
+ "shape": [
7
+ 18
8
+ ]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9
  },
10
+ "observation.images.head": {
11
+ "type": "VISUAL",
12
+ "shape": [
13
+ 3,
14
+ 360,
15
+ 640
16
+ ]
17
  },
18
+ "observation.images.left_wrist": {
19
+ "type": "VISUAL",
20
+ "shape": [
21
+ 3,
22
+ 360,
23
+ 640
24
+ ]
 
 
 
 
 
 
 
 
25
  },
26
+ "observation.images.right_wrist": {
27
+ "type": "VISUAL",
28
+ "shape": [
29
+ 3,
30
+ 360,
31
+ 640
32
+ ]
33
+ }
34
+ },
35
+ "output_features": {
36
+ "action": {
37
+ "type": "ACTION",
38
+ "shape": [
39
+ 18
40
+ ]
41
+ }
42
+ },
43
+ "device": "cuda",
44
+ "use_amp": true,
45
+ "use_peft": false,
46
+ "push_to_hub": true,
47
+ "repo_id": "Odog16/block_sorting_SmolVLA_policy",
48
+ "private": null,
49
+ "tags": null,
50
+ "license": null,
51
+ "pretrained_path": "lerobot/smolvla_base",
52
+ "chunk_size": 30,
53
+ "n_action_steps": 30,
54
+ "normalization_mapping": {
55
+ "VISUAL": "IDENTITY",
56
+ "STATE": "MEAN_STD",
57
+ "ACTION": "MEAN_STD"
58
+ },
59
+ "max_state_dim": 32,
60
+ "max_action_dim": 32,
61
+ "resize_imgs_with_padding": [
62
+ 512,
63
+ 512
64
+ ],
65
+ "empty_cameras": 0,
66
+ "adapt_to_pi_aloha": false,
67
+ "use_delta_joint_actions_aloha": false,
68
+ "tokenizer_max_length": 48,
69
+ "num_steps": 10,
70
+ "use_cache": true,
71
+ "freeze_vision_encoder": true,
72
+ "train_expert_only": true,
73
+ "train_state_proj": true,
74
+ "optimizer_lr": 0.0001,
75
+ "optimizer_betas": [
76
+ 0.9,
77
+ 0.95
78
+ ],
79
+ "optimizer_eps": 1e-08,
80
+ "optimizer_weight_decay": 1e-10,
81
+ "optimizer_grad_clip_norm": 10,
82
+ "scheduler_warmup_steps": 500,
83
+ "scheduler_decay_steps": 18000,
84
+ "scheduler_decay_lr": 2.5e-06,
85
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
86
+ "load_vlm_weights": true,
87
+ "add_image_special_tokens": false,
88
+ "attention_mode": "cross_attn",
89
+ "prefix_length": -1,
90
+ "pad_language_to": "longest",
91
+ "num_expert_layers": -1,
92
+ "num_vlm_layers": 16,
93
+ "self_attn_every_n_layers": 2,
94
+ "expert_width_multiplier": 0.75,
95
+ "min_period": 0.004,
96
+ "max_period": 4.0,
97
+ "rtc_config": null,
98
+ "compile_model": false,
99
+ "compile_mode": "max-autotune"
100
  }