Hugging Face's logo

usr256864
/
act

Safetensors
Model card Files
xet
 Community
usr256864 commited on
Commit
4376ac2
·
verified ·
1 Parent(s): 7fb85d9

Initial model upload

Browse files
Files changed (7) hide show
  1. config.json +63 -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 +56 -0
  6. policy_preprocessor_step_3_normalizer_processor.safetensors +3 -0
  7. train_config.json +238 -0
config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "act",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.image": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 256,
10
+ 256
11
+ ]
12
+ },
13
+ "observation.state": {
14
+ "type": "STATE",
15
+ "shape": [
16
+ 32
17
+ ]
18
+ }
19
+ },
20
+ "output_features": {
21
+ "action": {
22
+ "type": "ACTION",
23
+ "shape": [
24
+ 32
25
+ ]
26
+ }
27
+ },
28
+ "device": "cuda",
29
+ "use_amp": false,
30
+ "use_peft": false,
31
+ "push_to_hub": false,
32
+ "repo_id": "usr256864/pi0",
33
+ "private": null,
34
+ "tags": null,
35
+ "license": null,
36
+ "pretrained_path": null,
37
+ "chunk_size": 100,
38
+ "n_action_steps": 100,
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
+ "optimizer_lr": 1e-05,
61
+ "optimizer_weight_decay": 0.0001,
62
+ "optimizer_lr_backbone": 1e-05
63
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:220cfa2d3f9312c7ba297c5d7e44b76164129a7fa93782079cc7b02d089860fb
3
+ size 206912856
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
+ 32
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 sha256:9519068ba0094384fc7a499a7eb7fefdb58c5f74f3db6c1360bf34ea2b6b21e6
3
+ size 8528
policy_preprocessor.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "device_processor",
16
+ "config": {
17
+ "device": "cuda",
18
+ "float_dtype": null
19
+ }
20
+ },
21
+ {
22
+ "registry_name": "normalizer_processor",
23
+ "config": {
24
+ "eps": 1e-08,
25
+ "features": {
26
+ "observation.image": {
27
+ "type": "VISUAL",
28
+ "shape": [
29
+ 3,
30
+ 256,
31
+ 256
32
+ ]
33
+ },
34
+ "observation.state": {
35
+ "type": "STATE",
36
+ "shape": [
37
+ 32
38
+ ]
39
+ },
40
+ "action": {
41
+ "type": "ACTION",
42
+ "shape": [
43
+ 32
44
+ ]
45
+ }
46
+ },
47
+ "norm_map": {
48
+ "VISUAL": "MEAN_STD",
49
+ "STATE": "MEAN_STD",
50
+ "ACTION": "MEAN_STD"
51
+ }
52
+ },
53
+ "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors"
54
+ }
55
+ ]
56
+ }
policy_preprocessor_step_3_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9519068ba0094384fc7a499a7eb7fefdb58c5f74f3db6c1360bf34ea2b6b21e6
3
+ size 8528
train_config.json ADDED
@@ -0,0 +1,238 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "usr256864/lr_ds",
4
+ "root": "/netscratch/mboukheddimi/RH5_dataset_threads/",
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
+ "affine": {
62
+ "weight": 1.0,
63
+ "type": "RandomAffine",
64
+ "kwargs": {
65
+ "degrees": [
66
+ -5.0,
67
+ 5.0
68
+ ],
69
+ "translate": [
70
+ 0.05,
71
+ 0.05
72
+ ]
73
+ }
74
+ }
75
+ }
76
+ },
77
+ "revision": null,
78
+ "use_imagenet_stats": true,
79
+ "video_backend": "torchcodec",
80
+ "streaming": false
81
+ },
82
+ "env": {
83
+ "type": "rh5",
84
+ "task": "walk",
85
+ "fps": 30,
86
+ "features": {
87
+ "action": {
88
+ "type": "ACTION",
89
+ "shape": [
90
+ 32
91
+ ]
92
+ },
93
+ "agent_pos": {
94
+ "type": "STATE",
95
+ "shape": [
96
+ 32
97
+ ]
98
+ },
99
+ "pixels/top": {
100
+ "type": "VISUAL",
101
+ "shape": [
102
+ 480,
103
+ 640,
104
+ 3
105
+ ]
106
+ }
107
+ },
108
+ "features_map": {
109
+ "action": "action",
110
+ "agent_pos": "observation.state",
111
+ "top": "observation.image",
112
+ "pixels/top": "observation.image"
113
+ },
114
+ "max_parallel_tasks": 1,
115
+ "disable_env_checker": true,
116
+ "config": {
117
+ "eval_ds": "/netscratch/mboukheddimi/RH5_dataset_threads/",
118
+ "episode_idx": 806,
119
+ "absolute": false
120
+ },
121
+ "episode_length": 800,
122
+ "obs_type": "pixels_agent_pos",
123
+ "render_mode": "rgb_array"
124
+ },
125
+ "policy": {
126
+ "type": "act",
127
+ "n_obs_steps": 1,
128
+ "input_features": {
129
+ "observation.image": {
130
+ "type": "VISUAL",
131
+ "shape": [
132
+ 3,
133
+ 256,
134
+ 256
135
+ ]
136
+ },
137
+ "observation.state": {
138
+ "type": "STATE",
139
+ "shape": [
140
+ 32
141
+ ]
142
+ }
143
+ },
144
+ "output_features": {
145
+ "action": {
146
+ "type": "ACTION",
147
+ "shape": [
148
+ 32
149
+ ]
150
+ }
151
+ },
152
+ "device": "cuda",
153
+ "use_amp": false,
154
+ "use_peft": false,
155
+ "push_to_hub": false,
156
+ "repo_id": "usr256864/pi0",
157
+ "private": null,
158
+ "tags": null,
159
+ "license": null,
160
+ "pretrained_path": null,
161
+ "chunk_size": 100,
162
+ "n_action_steps": 100,
163
+ "normalization_mapping": {
164
+ "VISUAL": "MEAN_STD",
165
+ "STATE": "MEAN_STD",
166
+ "ACTION": "MEAN_STD"
167
+ },
168
+ "vision_backbone": "resnet18",
169
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
170
+ "replace_final_stride_with_dilation": false,
171
+ "pre_norm": false,
172
+ "dim_model": 512,
173
+ "n_heads": 8,
174
+ "dim_feedforward": 3200,
175
+ "feedforward_activation": "relu",
176
+ "n_encoder_layers": 4,
177
+ "n_decoder_layers": 1,
178
+ "use_vae": true,
179
+ "latent_dim": 32,
180
+ "n_vae_encoder_layers": 4,
181
+ "temporal_ensemble_coeff": null,
182
+ "dropout": 0.1,
183
+ "kl_weight": 10.0,
184
+ "optimizer_lr": 1e-05,
185
+ "optimizer_weight_decay": 0.0001,
186
+ "optimizer_lr_backbone": 1e-05
187
+ },
188
+ "output_dir": "outputs/train/2026-01-28/13-59-53_act_training_delta",
189
+ "job_name": "act_training_delta",
190
+ "resume": false,
191
+ "seed": 1000,
192
+ "num_workers": 4,
193
+ "batch_size": 32,
194
+ "steps": 100000,
195
+ "eval_freq": 5000,
196
+ "log_freq": 400,
197
+ "tolerance_s": 0.0001,
198
+ "save_checkpoint": true,
199
+ "save_freq": 10000,
200
+ "use_policy_training_preset": true,
201
+ "optimizer": {
202
+ "type": "adamw",
203
+ "lr": 1e-05,
204
+ "weight_decay": 0.0001,
205
+ "grad_clip_norm": 10.0,
206
+ "betas": [
207
+ 0.9,
208
+ 0.999
209
+ ],
210
+ "eps": 1e-08
211
+ },
212
+ "scheduler": null,
213
+ "eval": {
214
+ "n_episodes": 1,
215
+ "batch_size": 1,
216
+ "use_async_envs": false
217
+ },
218
+ "wandb": {
219
+ "enable": false,
220
+ "disable_artifact": true,
221
+ "project": "lerobot",
222
+ "entity": null,
223
+ "notes": null,
224
+ "run_id": null,
225
+ "mode": null
226
+ },
227
+ "peft": null,
228
+ "use_rabc": false,
229
+ "rabc_progress_path": null,
230
+ "rabc_kappa": 0.01,
231
+ "rabc_epsilon": 1e-06,
232
+ "rabc_head_mode": "sparse",
233
+ "rename_map": {},
234
+ "checkpoint_path": null,
235
+ "eval_ds": "/netscratch/mboukheddimi/RH5_dataset_threads/",
236
+ "episode_idx": 806,
237
+ "absolute": false
238
+ }