Hugging Face's logo

HBOrtiz
/
so101_smolvla_eval3_cotrain

Safetensors
Model card Files
xet
 Community
HBOrtiz commited on
Commit
5978c4c
·
verified ·
1 Parent(s): d04cb86

checkpoint: step_020000

Browse files
Files changed (6) hide show
  1. step_020000/config.json +117 -0
  2. step_020000/model.safetensors +3 -0
  3. step_020000/policy_postprocessor.json +32 -0
  4. step_020000/policy_postprocessor_step_0_unnormalizer_processor.safetensors +3 -0
  5. step_020000/policy_preprocessor.json +103 -0
  6. step_020000/policy_preprocessor_step_5_normalizer_processor.safetensors +3 -0
step_020000/config.json ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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:0",
61
+ "use_amp": false,
62
+ "use_peft": false,
63
+ "push_to_hub": true,
64
+ "repo_id": null,
65
+ "private": null,
66
+ "tags": null,
67
+ "license": null,
68
+ "pretrained_path": null,
69
+ "chunk_size": 50,
70
+ "n_action_steps": 50,
71
+ "normalization_mapping": {
72
+ "VISUAL": "IDENTITY",
73
+ "STATE": "MEAN_STD",
74
+ "ACTION": "MEAN_STD"
75
+ },
76
+ "max_state_dim": 32,
77
+ "max_action_dim": 32,
78
+ "resize_imgs_with_padding": [
79
+ 512,
80
+ 512
81
+ ],
82
+ "empty_cameras": 2,
83
+ "adapt_to_pi_aloha": false,
84
+ "use_delta_joint_actions_aloha": false,
85
+ "tokenizer_max_length": 48,
86
+ "num_steps": 10,
87
+ "use_cache": true,
88
+ "freeze_vision_encoder": true,
89
+ "train_expert_only": false,
90
+ "train_state_proj": true,
91
+ "optimizer_lr": 0.0001,
92
+ "optimizer_betas": [
93
+ 0.9,
94
+ 0.95
95
+ ],
96
+ "optimizer_eps": 1e-08,
97
+ "optimizer_weight_decay": 1e-10,
98
+ "optimizer_grad_clip_norm": 10.0,
99
+ "scheduler_warmup_steps": 1000,
100
+ "scheduler_decay_steps": 30000,
101
+ "scheduler_decay_lr": 2.5e-06,
102
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
103
+ "load_vlm_weights": true,
104
+ "add_image_special_tokens": false,
105
+ "attention_mode": "cross_attn",
106
+ "prefix_length": 0,
107
+ "pad_language_to": "max_length",
108
+ "num_expert_layers": 0,
109
+ "num_vlm_layers": 16,
110
+ "self_attn_every_n_layers": 2,
111
+ "expert_width_multiplier": 0.75,
112
+ "min_period": 0.004,
113
+ "max_period": 4.0,
114
+ "rtc_config": null,
115
+ "compile_model": false,
116
+ "compile_mode": "max-autotune"
117
+ }
step_020000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:15ca1dd7525e96c7068e4d6d8f00f4da9112ca5b24d67ce305f6019e9deb4293
3
+ size 906712520
step_020000/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
+ }
step_020000/policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c8742a0c6c153cb7638d4d6aae2e0c3b5e9dee8687f45aac14d1656f51a2344d
3
+ size 233328
step_020000/policy_preprocessor.json ADDED
@@ -0,0 +1,103 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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.camera1": {
48
+ "type": "VISUAL",
49
+ "shape": [
50
+ 3,
51
+ 256,
52
+ 256
53
+ ]
54
+ },
55
+ "observation.images.camera2": {
56
+ "type": "VISUAL",
57
+ "shape": [
58
+ 3,
59
+ 256,
60
+ 256
61
+ ]
62
+ },
63
+ "observation.images.camera3": {
64
+ "type": "VISUAL",
65
+ "shape": [
66
+ 3,
67
+ 256,
68
+ 256
69
+ ]
70
+ },
71
+ "observation.images.empty_camera_0": {
72
+ "type": "VISUAL",
73
+ "shape": [
74
+ 3,
75
+ 480,
76
+ 640
77
+ ]
78
+ },
79
+ "observation.images.empty_camera_1": {
80
+ "type": "VISUAL",
81
+ "shape": [
82
+ 3,
83
+ 480,
84
+ 640
85
+ ]
86
+ },
87
+ "action": {
88
+ "type": "ACTION",
89
+ "shape": [
90
+ 6
91
+ ]
92
+ }
93
+ },
94
+ "norm_map": {
95
+ "VISUAL": "IDENTITY",
96
+ "STATE": "MEAN_STD",
97
+ "ACTION": "MEAN_STD"
98
+ }
99
+ },
100
+ "state_file": "policy_preprocessor_step_5_normalizer_processor.safetensors"
101
+ }
102
+ ]
103
+ }
step_020000/policy_preprocessor_step_5_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c8742a0c6c153cb7638d4d6aae2e0c3b5e9dee8687f45aac14d1656f51a2344d
3
+ size 233328