Hugging Face's logo

jclinton1
/
wedgit_stack_single_protosam_40k

Safetensors
Model card Files
xet
Community
jclinton1 commited on
Commit
507a3d9
·
verified ·
1 Parent(s): 27189c1

Upload folder using huggingface_hub

Browse files
Files changed (3) hide show
  1. config.json +75 -0
  2. model.safetensors +3 -0
  3. train_config.json +182 -0
config.json ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "act",
3
+ "n_obs_steps": 1,
4
+ "normalization_mapping": {
5
+ "VISUAL": "IDENTITY",
6
+ "STATE": "MEAN_STD",
7
+ "ACTION": "MEAN_STD"
8
+ },
9
+ "input_features": {
10
+ "observation.state": {
11
+ "type": "STATE",
12
+ "shape": [
13
+ 7
14
+ ]
15
+ },
16
+ "observation.images.wrist": {
17
+ "type": "VISUAL",
18
+ "shape": [
19
+ 3,
20
+ 480,
21
+ 640
22
+ ]
23
+ },
24
+ "observation.images.webcam": {
25
+ "type": "VISUAL",
26
+ "shape": [
27
+ 3,
28
+ 480,
29
+ 640
30
+ ]
31
+ }
32
+ },
33
+ "output_features": {
34
+ "action": {
35
+ "type": "ACTION",
36
+ "shape": [
37
+ 7
38
+ ]
39
+ }
40
+ },
41
+ "device": "cuda",
42
+ "use_amp": false,
43
+ "chunk_size": 100,
44
+ "n_action_steps": 100,
45
+ "vision_backbone": "ProtoSAM",
46
+ "pretrained_backbone_weights": "weights/FastSAM-s.pt",
47
+ "proto_indices": [
48
+ 3,
49
+ 5,
50
+ 6,
51
+ 8,
52
+ 19,
53
+ 31,
54
+ 21
55
+ ],
56
+ "freeze_backbone": true,
57
+ "replace_final_stride_with_dilation": false,
58
+ "pre_norm": false,
59
+ "dim_model": 448,
60
+ "n_heads": 8,
61
+ "dim_feedforward": 3200,
62
+ "feedforward_activation": "relu",
63
+ "n_encoder_layers": 4,
64
+ "n_decoder_layers": 1,
65
+ "use_vae": true,
66
+ "latent_dim": 32,
67
+ "n_vae_encoder_layers": 4,
68
+ "use_depth_transform": false,
69
+ "temporal_ensemble_coeff": null,
70
+ "dropout": 0.1,
71
+ "kl_weight": 10.0,
72
+ "optimizer_lr": 1e-05,
73
+ "optimizer_weight_decay": 0.0001,
74
+ "optimizer_lr_backbone": 1e-05
75
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b8c49b1b83c450f12f4987d99b835b41fe7548ce8b5d2f28bbd0bd2200891247
3
+ size 184967128
train_config.json ADDED
@@ -0,0 +1,182 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "jclinton1/wedgit_stack_single_dual_cam_v3_ee",
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": "act",
70
+ "n_obs_steps": 1,
71
+ "normalization_mapping": {
72
+ "VISUAL": "IDENTITY",
73
+ "STATE": "MEAN_STD",
74
+ "ACTION": "MEAN_STD"
75
+ },
76
+ "input_features": {
77
+ "observation.state": {
78
+ "type": "STATE",
79
+ "shape": [
80
+ 7
81
+ ]
82
+ },
83
+ "observation.images.wrist": {
84
+ "type": "VISUAL",
85
+ "shape": [
86
+ 3,
87
+ 480,
88
+ 640
89
+ ]
90
+ },
91
+ "observation.images.webcam": {
92
+ "type": "VISUAL",
93
+ "shape": [
94
+ 3,
95
+ 480,
96
+ 640
97
+ ]
98
+ }
99
+ },
100
+ "output_features": {
101
+ "action": {
102
+ "type": "ACTION",
103
+ "shape": [
104
+ 7
105
+ ]
106
+ }
107
+ },
108
+ "device": "cuda",
109
+ "use_amp": false,
110
+ "chunk_size": 100,
111
+ "n_action_steps": 100,
112
+ "vision_backbone": "ProtoSAM",
113
+ "pretrained_backbone_weights": "weights/FastSAM-s.pt",
114
+ "proto_indices": [
115
+ 3,
116
+ 5,
117
+ 6,
118
+ 8,
119
+ 19,
120
+ 31,
121
+ 21
122
+ ],
123
+ "freeze_backbone": true,
124
+ "replace_final_stride_with_dilation": false,
125
+ "pre_norm": false,
126
+ "dim_model": 448,
127
+ "n_heads": 8,
128
+ "dim_feedforward": 3200,
129
+ "feedforward_activation": "relu",
130
+ "n_encoder_layers": 4,
131
+ "n_decoder_layers": 1,
132
+ "use_vae": true,
133
+ "latent_dim": 32,
134
+ "n_vae_encoder_layers": 4,
135
+ "use_depth_transform": false,
136
+ "temporal_ensemble_coeff": null,
137
+ "dropout": 0.1,
138
+ "kl_weight": 10.0,
139
+ "optimizer_lr": 1e-05,
140
+ "optimizer_weight_decay": 0.0001,
141
+ "optimizer_lr_backbone": 1e-05
142
+ },
143
+ "output_dir": "outputs/train/wedgit_stack_single_protosam",
144
+ "job_name": "protosam_conv",
145
+ "resume": false,
146
+ "seed": 1000,
147
+ "num_workers": 8,
148
+ "batch_size": 8,
149
+ "steps": 300000,
150
+ "eval_freq": 20000,
151
+ "validation_freq": 400,
152
+ "log_freq": 200,
153
+ "save_checkpoint": true,
154
+ "save_freq": 20000,
155
+ "use_policy_training_preset": true,
156
+ "optimizer": {
157
+ "type": "adamw",
158
+ "lr": 1e-05,
159
+ "weight_decay": 0.0001,
160
+ "grad_clip_norm": 10.0,
161
+ "betas": [
162
+ 0.9,
163
+ 0.999
164
+ ],
165
+ "eps": 1e-08
166
+ },
167
+ "scheduler": null,
168
+ "eval": {
169
+ "n_episodes": 50,
170
+ "batch_size": 50,
171
+ "use_async_envs": false
172
+ },
173
+ "wandb": {
174
+ "enable": true,
175
+ "disable_artifact": false,
176
+ "project": "wedgit_stack_single",
177
+ "entity": null,
178
+ "notes": null,
179
+ "run_id": null,
180
+ "mode": null
181
+ }
182
+ }