Hugging Face's logo

k1000dai
/
rt_spatial_test_tf

Safetensors
Model card Files
xet
Community
k1000dai commited on
Commit
84c2abe
·
verified ·
1 Parent(s): 42e45d5

Upload folder using huggingface_hub

Browse files
Files changed (3) hide show
  1. config.json +60 -0
  2. model.safetensors +3 -0
  3. train_config.json +166 -0
config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "residual_transformer",
3
+ "n_obs_steps": 1,
4
+ "normalization_mapping": {
5
+ "VISUAL": "MEAN_STD",
6
+ "STATE": "MEAN_STD",
7
+ "ACTION": "MEAN_STD",
8
+ "ENV": "MEAN_STD"
9
+ },
10
+ "input_features": {
11
+ "observation.images.image": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 256,
16
+ 256
17
+ ]
18
+ },
19
+ "observation.images.wrist_image": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 256,
24
+ 256
25
+ ]
26
+ },
27
+ "observation.state": {
28
+ "type": "STATE",
29
+ "shape": [
30
+ 8
31
+ ]
32
+ }
33
+ },
34
+ "output_features": {
35
+ "action": {
36
+ "type": "ACTION",
37
+ "shape": [
38
+ 7
39
+ ]
40
+ }
41
+ },
42
+ "device": "cuda",
43
+ "use_amp": false,
44
+ "push_to_hub": true,
45
+ "repo_id": "k1000dai/residual_transformer_libero_spatial_tf",
46
+ "private": null,
47
+ "tags": null,
48
+ "license": null,
49
+ "vision_backbone": "resnet18",
50
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
51
+ "freeze_vision_backbone": false,
52
+ "dim_model": 512,
53
+ "n_heads": 8,
54
+ "n_encoder_layers": 10,
55
+ "dim_feedforward": 2048,
56
+ "dropout": 0.1,
57
+ "use_language": true,
58
+ "optimizer_lr": 1e-05,
59
+ "optimizer_weight_decay": 1e-05
60
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c4535facf6ab9fb929e42364c6ef891bfd451641972da3e32aa3a010eacdab6
3
+ size 173946684
train_config.json ADDED
@@ -0,0 +1,166 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "k1000dai/libero-spatial-smolvla",
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
+ }
62
+ },
63
+ "revision": null,
64
+ "use_imagenet_stats": true,
65
+ "video_backend": "torchcodec"
66
+ },
67
+ "env": null,
68
+ "policy": {
69
+ "type": "residual_transformer",
70
+ "n_obs_steps": 1,
71
+ "normalization_mapping": {
72
+ "VISUAL": "MEAN_STD",
73
+ "STATE": "MEAN_STD",
74
+ "ACTION": "MEAN_STD",
75
+ "ENV": "MEAN_STD"
76
+ },
77
+ "input_features": {
78
+ "observation.images.image": {
79
+ "type": "VISUAL",
80
+ "shape": [
81
+ 3,
82
+ 256,
83
+ 256
84
+ ]
85
+ },
86
+ "observation.images.wrist_image": {
87
+ "type": "VISUAL",
88
+ "shape": [
89
+ 3,
90
+ 256,
91
+ 256
92
+ ]
93
+ },
94
+ "observation.state": {
95
+ "type": "STATE",
96
+ "shape": [
97
+ 8
98
+ ]
99
+ }
100
+ },
101
+ "output_features": {
102
+ "action": {
103
+ "type": "ACTION",
104
+ "shape": [
105
+ 7
106
+ ]
107
+ }
108
+ },
109
+ "device": "cuda",
110
+ "use_amp": false,
111
+ "push_to_hub": true,
112
+ "repo_id": "k1000dai/residual_transformer_libero_spatial_tf",
113
+ "private": null,
114
+ "tags": null,
115
+ "license": null,
116
+ "vision_backbone": "resnet18",
117
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
118
+ "freeze_vision_backbone": false,
119
+ "dim_model": 512,
120
+ "n_heads": 8,
121
+ "n_encoder_layers": 10,
122
+ "dim_feedforward": 2048,
123
+ "dropout": 0.1,
124
+ "use_language": true,
125
+ "optimizer_lr": 1e-05,
126
+ "optimizer_weight_decay": 1e-05
127
+ },
128
+ "output_dir": "output_residual_transformer_spatial_tf",
129
+ "job_name": "residual_transformer_libero_spatial_tf",
130
+ "resume": false,
131
+ "seed": 1000,
132
+ "num_workers": 16,
133
+ "batch_size": 64,
134
+ "steps": 400000,
135
+ "eval_freq": 20000,
136
+ "log_freq": 200,
137
+ "save_checkpoint": true,
138
+ "save_freq": 20000,
139
+ "use_policy_training_preset": true,
140
+ "optimizer": {
141
+ "type": "adamw",
142
+ "lr": 1e-05,
143
+ "weight_decay": 1e-05,
144
+ "grad_clip_norm": 10.0,
145
+ "betas": [
146
+ 0.9,
147
+ 0.999
148
+ ],
149
+ "eps": 1e-08
150
+ },
151
+ "scheduler": null,
152
+ "eval": {
153
+ "n_episodes": 50,
154
+ "batch_size": 50,
155
+ "use_async_envs": false
156
+ },
157
+ "wandb": {
158
+ "enable": true,
159
+ "disable_artifact": false,
160
+ "project": "lerobot",
161
+ "entity": null,
162
+ "notes": null,
163
+ "run_id": "1obybpuq",
164
+ "mode": null
165
+ }
166
+ }