Hugging Face's logo

steb6
/
act-HOME_stand_wave_gc

Safetensors
Model card Files
xet
Community
steb6 commited on
Commit
8d72716
·
verified ·
1 Parent(s): 8a57ed6

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +71 -0
  2. model.safetensors +3 -0
  3. policy_postprocessor.json +32 -0
  4. policy_postprocessor_step_0_unnormalizer_processor.safetensors +3 -0
  5. policy_preprocessor.json +73 -0
  6. policy_preprocessor_step_4_normalizer_processor.safetensors +3 -0
  7. train_config.json +213 -0
config.json ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "act",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 36
9
+ ]
10
+ },
11
+ "observation.images.egocentric": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 480,
16
+ 640
17
+ ]
18
+ }
19
+ },
20
+ "output_features": {
21
+ "action": {
22
+ "type": "ACTION",
23
+ "shape": [
24
+ 36
25
+ ]
26
+ }
27
+ },
28
+ "device": "cuda",
29
+ "use_amp": false,
30
+ "use_peft": false,
31
+ "push_to_hub": true,
32
+ "repo_id": "steb6/act-HOME_stand_wave_gc",
33
+ "private": null,
34
+ "tags": null,
35
+ "license": null,
36
+ "pretrained_path": null,
37
+ "chunk_size": 20,
38
+ "n_action_steps": 20,
39
+ "normalization_mapping": {
40
+ "VISUAL": "MEAN_STD",
41
+ "STATE": "MEAN_STD",
42
+ "ACTION": "MEAN_STD"
43
+ },
44
+ "vision_backbone": "resnet18",
45
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
46
+ "replace_final_stride_with_dilation": false,
47
+ "pre_norm": false,
48
+ "dim_model": 512,
49
+ "n_heads": 8,
50
+ "dim_feedforward": 3200,
51
+ "feedforward_activation": "relu",
52
+ "n_encoder_layers": 4,
53
+ "n_decoder_layers": 1,
54
+ "use_vae": true,
55
+ "latent_dim": 32,
56
+ "n_vae_encoder_layers": 4,
57
+ "temporal_ensemble_coeff": null,
58
+ "dropout": 0.1,
59
+ "kl_weight": 10.0,
60
+ "use_language_conditioning": false,
61
+ "language_encoder_type": "clip",
62
+ "language_model_name": "openai/clip-vit-base-patch32",
63
+ "freeze_language_encoder": true,
64
+ "language_projection_dim": null,
65
+ "language_dropout": 0.1,
66
+ "max_token_length": 77,
67
+ "language_pooling": "cls",
68
+ "optimizer_lr": 1e-05,
69
+ "optimizer_weight_decay": 0.0001,
70
+ "optimizer_lr_backbone": 1e-05
71
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:efc200dd7aa9a5922b915283f62a3842932e8b9e2a7661c5f37abcc89136e844
3
+ size 206617960
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
+ 36
13
+ ]
14
+ }
15
+ },
16
+ "norm_map": {
17
+ "VISUAL": "MEAN_STD",
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
+ }
policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32bb3897b8c4f28170d4e4301e03432b1f3977ee76894dea7ad452d3879aa965
3
+ size 8944
policy_preprocessor.json ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "policy_preprocessor",
3
+ "steps": [
4
+ {
5
+ "registry_name": "image_crop_resize_processor",
6
+ "config": {
7
+ "crop_params_dict": {
8
+ "observation.images.egocentric": [
9
+ 0,
10
+ 80,
11
+ 480,
12
+ 480
13
+ ]
14
+ },
15
+ "resize_size": [
16
+ 256,
17
+ 256
18
+ ]
19
+ }
20
+ },
21
+ {
22
+ "registry_name": "rename_observations_processor",
23
+ "config": {
24
+ "rename_map": {}
25
+ }
26
+ },
27
+ {
28
+ "registry_name": "to_batch_processor",
29
+ "config": {}
30
+ },
31
+ {
32
+ "registry_name": "device_processor",
33
+ "config": {
34
+ "device": "cuda",
35
+ "float_dtype": null
36
+ }
37
+ },
38
+ {
39
+ "registry_name": "normalizer_processor",
40
+ "config": {
41
+ "eps": 1e-08,
42
+ "features": {
43
+ "observation.state": {
44
+ "type": "STATE",
45
+ "shape": [
46
+ 36
47
+ ]
48
+ },
49
+ "observation.images.egocentric": {
50
+ "type": "VISUAL",
51
+ "shape": [
52
+ 3,
53
+ 480,
54
+ 640
55
+ ]
56
+ },
57
+ "action": {
58
+ "type": "ACTION",
59
+ "shape": [
60
+ 36
61
+ ]
62
+ }
63
+ },
64
+ "norm_map": {
65
+ "VISUAL": "MEAN_STD",
66
+ "STATE": "MEAN_STD",
67
+ "ACTION": "MEAN_STD"
68
+ }
69
+ },
70
+ "state_file": "policy_preprocessor_step_4_normalizer_processor.safetensors"
71
+ }
72
+ ]
73
+ }
policy_preprocessor_step_4_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32bb3897b8c4f28170d4e4301e03432b1f3977ee76894dea7ad452d3879aa965
3
+ size 8944
train_config.json ADDED
@@ -0,0 +1,213 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "steb6/HOME_stand_wave",
4
+ "root": null,
5
+ "episodes": null,
6
+ "image_transforms": {
7
+ "enable": true,
8
+ "max_num_transforms": 3,
9
+ "random_order": true,
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
+ "green_screen_replace": {
62
+ "weight": -1.0,
63
+ "type": "GreenScreenReplace",
64
+ "kwargs": {
65
+ "pool_dir": "./gs_image_pool",
66
+ "green_ratio": 0.001,
67
+ "min_green": 0.001,
68
+ "spill": 0.001,
69
+ "hue_min": 0.1,
70
+ "hue_max": 0.5,
71
+ "min_s": 0.1,
72
+ "min_v": 0.1
73
+ }
74
+ }
75
+ }
76
+ },
77
+ "revision": null,
78
+ "use_imagenet_stats": true,
79
+ "video_backend": "torchcodec",
80
+ "streaming": false
81
+ },
82
+ "env": null,
83
+ "policy": {
84
+ "type": "act",
85
+ "n_obs_steps": 1,
86
+ "input_features": {
87
+ "observation.state": {
88
+ "type": "STATE",
89
+ "shape": [
90
+ 36
91
+ ]
92
+ },
93
+ "observation.images.egocentric": {
94
+ "type": "VISUAL",
95
+ "shape": [
96
+ 3,
97
+ 480,
98
+ 640
99
+ ]
100
+ }
101
+ },
102
+ "output_features": {
103
+ "action": {
104
+ "type": "ACTION",
105
+ "shape": [
106
+ 36
107
+ ]
108
+ }
109
+ },
110
+ "device": "cuda",
111
+ "use_amp": false,
112
+ "use_peft": false,
113
+ "push_to_hub": true,
114
+ "repo_id": "steb6/act-HOME_stand_wave_gc",
115
+ "private": null,
116
+ "tags": null,
117
+ "license": null,
118
+ "pretrained_path": null,
119
+ "chunk_size": 20,
120
+ "n_action_steps": 20,
121
+ "normalization_mapping": {
122
+ "VISUAL": "MEAN_STD",
123
+ "STATE": "MEAN_STD",
124
+ "ACTION": "MEAN_STD"
125
+ },
126
+ "vision_backbone": "resnet18",
127
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
128
+ "replace_final_stride_with_dilation": false,
129
+ "pre_norm": false,
130
+ "dim_model": 512,
131
+ "n_heads": 8,
132
+ "dim_feedforward": 3200,
133
+ "feedforward_activation": "relu",
134
+ "n_encoder_layers": 4,
135
+ "n_decoder_layers": 1,
136
+ "use_vae": true,
137
+ "latent_dim": 32,
138
+ "n_vae_encoder_layers": 4,
139
+ "temporal_ensemble_coeff": null,
140
+ "dropout": 0.1,
141
+ "kl_weight": 10.0,
142
+ "use_language_conditioning": false,
143
+ "language_encoder_type": "clip",
144
+ "language_model_name": "openai/clip-vit-base-patch32",
145
+ "freeze_language_encoder": true,
146
+ "language_projection_dim": null,
147
+ "language_dropout": 0.1,
148
+ "max_token_length": 77,
149
+ "language_pooling": "cls",
150
+ "optimizer_lr": 1e-05,
151
+ "optimizer_weight_decay": 0.0001,
152
+ "optimizer_lr_backbone": 1e-05
153
+ },
154
+ "output_dir": "checkpoints/act_HOME_stand_wave_gc",
155
+ "job_name": "act",
156
+ "resume": false,
157
+ "seed": 1000,
158
+ "num_workers": 4,
159
+ "batch_size": 8,
160
+ "steps": 100000,
161
+ "eval_freq": 20000,
162
+ "log_freq": 200,
163
+ "tolerance_s": 0.0001,
164
+ "save_checkpoint": true,
165
+ "save_freq": 20000,
166
+ "use_policy_training_preset": true,
167
+ "optimizer": {
168
+ "type": "adamw",
169
+ "lr": 1e-05,
170
+ "weight_decay": 0.0001,
171
+ "grad_clip_norm": 10.0,
172
+ "betas": [
173
+ 0.9,
174
+ 0.999
175
+ ],
176
+ "eps": 1e-08
177
+ },
178
+ "scheduler": null,
179
+ "eval": {
180
+ "n_episodes": 50,
181
+ "batch_size": 50,
182
+ "use_async_envs": false
183
+ },
184
+ "wandb": {
185
+ "enable": false,
186
+ "disable_artifact": false,
187
+ "project": "lerobot",
188
+ "entity": null,
189
+ "notes": null,
190
+ "run_id": null,
191
+ "mode": null
192
+ },
193
+ "peft": null,
194
+ "use_rabc": false,
195
+ "rabc_progress_path": null,
196
+ "rabc_kappa": 0.01,
197
+ "rabc_epsilon": 1e-06,
198
+ "rabc_head_mode": "sparse",
199
+ "rename_map": {},
200
+ "image_resize_size": [
201
+ 256,
202
+ 256
203
+ ],
204
+ "image_crop_params": {
205
+ "observation.images.egocentric": [
206
+ 0,
207
+ 80,
208
+ 480,
209
+ 480
210
+ ]
211
+ },
212
+ "checkpoint_path": null
213
+ }