Hugging Face's logo

YHRG
/
pi0-bi-s1-mixed-pick-place-cube-fixed-5pos

Safetensors
Model card Files
xet
 Community
wyb5210 commited on
Commit
f7ed1eb
·
verified ·
1 Parent(s): a3dd304

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +95 -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 +91 -0
  6. policy_preprocessor_step_5_normalizer_processor.safetensors +3 -0
  7. train_config.json +240 -0
config.json ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "pi0",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.images.base_0_rgb": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 224,
10
+ 224
11
+ ]
12
+ },
13
+ "observation.images.left_wrist_0_rgb": {
14
+ "type": "VISUAL",
15
+ "shape": [
16
+ 3,
17
+ 224,
18
+ 224
19
+ ]
20
+ },
21
+ "observation.images.right_wrist_0_rgb": {
22
+ "type": "VISUAL",
23
+ "shape": [
24
+ 3,
25
+ 224,
26
+ 224
27
+ ]
28
+ },
29
+ "observation.state": {
30
+ "type": "STATE",
31
+ "shape": [
32
+ 32
33
+ ]
34
+ }
35
+ },
36
+ "output_features": {
37
+ "action": {
38
+ "type": "ACTION",
39
+ "shape": [
40
+ 14
41
+ ]
42
+ }
43
+ },
44
+ "device": "cuda",
45
+ "use_amp": false,
46
+ "use_peft": false,
47
+ "push_to_hub": false,
48
+ "repo_id": null,
49
+ "private": null,
50
+ "tags": null,
51
+ "license": null,
52
+ "pretrained_path": "/mnt/sdb/model/pi0_base/snapshots/26b99b9439acb1e352439e34ee9c67af0d76efa3",
53
+ "paligemma_variant": "gemma_2b",
54
+ "action_expert_variant": "gemma_300m",
55
+ "dtype": "bfloat16",
56
+ "chunk_size": 50,
57
+ "n_action_steps": 50,
58
+ "max_state_dim": 32,
59
+ "max_action_dim": 32,
60
+ "num_inference_steps": 10,
61
+ "time_sampling_beta_alpha": 1.5,
62
+ "time_sampling_beta_beta": 1.0,
63
+ "time_sampling_scale": 0.999,
64
+ "time_sampling_offset": 0.001,
65
+ "min_period": 0.004,
66
+ "max_period": 4.0,
67
+ "rtc_config": null,
68
+ "image_resolution": [
69
+ 224,
70
+ 224
71
+ ],
72
+ "empty_cameras": 0,
73
+ "normalization_mapping": {
74
+ "VISUAL": "IDENTITY",
75
+ "STATE": "MEAN_STD",
76
+ "ACTION": "MEAN_STD"
77
+ },
78
+ "gradient_checkpointing": true,
79
+ "compile_model": false,
80
+ "compile_mode": "max-autotune",
81
+ "freeze_vision_encoder": false,
82
+ "train_expert_only": false,
83
+ "optimizer_lr": 2.5e-05,
84
+ "optimizer_betas": [
85
+ 0.9,
86
+ 0.95
87
+ ],
88
+ "optimizer_eps": 1e-08,
89
+ "optimizer_weight_decay": 0.01,
90
+ "optimizer_grad_clip_norm": 1.0,
91
+ "scheduler_warmup_steps": 1000,
92
+ "scheduler_decay_steps": 30000,
93
+ "scheduler_decay_lr": 2.5e-06,
94
+ "tokenizer_max_length": 48
95
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71319da52bbbdffec7b167c51bf4e68a730737f95c6afbf6efbb3d324586f201
3
+ size 7011548640
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
+ 14
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
+ }
policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4da7407d16cb6217e0046c66ac4a2ad644b0a45c64ccfd61a9a51d32236d1db8
3
+ size 12652
policy_preprocessor.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "policy_preprocessor",
3
+ "steps": [
4
+ {
5
+ "registry_name": "rename_observations_processor",
6
+ "config": {
7
+ "rename_map": {
8
+ "observation.images.left": "observation.images.left_wrist_0_rgb",
9
+ "observation.images.middle": "observation.images.base_0_rgb",
10
+ "observation.images.right": "observation.images.right_wrist_0_rgb"
11
+ }
12
+ }
13
+ },
14
+ {
15
+ "registry_name": "to_batch_processor",
16
+ "config": {}
17
+ },
18
+ {
19
+ "registry_name": "pi0_new_line_processor",
20
+ "config": {}
21
+ },
22
+ {
23
+ "registry_name": "tokenizer_processor",
24
+ "config": {
25
+ "max_length": 48,
26
+ "task_key": "task",
27
+ "padding_side": "right",
28
+ "padding": "max_length",
29
+ "truncation": true,
30
+ "tokenizer_name": "google/paligemma-3b-pt-224"
31
+ }
32
+ },
33
+ {
34
+ "registry_name": "device_processor",
35
+ "config": {
36
+ "device": "cuda",
37
+ "float_dtype": null
38
+ }
39
+ },
40
+ {
41
+ "registry_name": "normalizer_processor",
42
+ "config": {
43
+ "eps": 1e-08,
44
+ "features": {
45
+ "observation.images.base_0_rgb": {
46
+ "type": "VISUAL",
47
+ "shape": [
48
+ 3,
49
+ 224,
50
+ 224
51
+ ]
52
+ },
53
+ "observation.images.left_wrist_0_rgb": {
54
+ "type": "VISUAL",
55
+ "shape": [
56
+ 3,
57
+ 224,
58
+ 224
59
+ ]
60
+ },
61
+ "observation.images.right_wrist_0_rgb": {
62
+ "type": "VISUAL",
63
+ "shape": [
64
+ 3,
65
+ 224,
66
+ 224
67
+ ]
68
+ },
69
+ "observation.state": {
70
+ "type": "STATE",
71
+ "shape": [
72
+ 32
73
+ ]
74
+ },
75
+ "action": {
76
+ "type": "ACTION",
77
+ "shape": [
78
+ 14
79
+ ]
80
+ }
81
+ },
82
+ "norm_map": {
83
+ "VISUAL": "IDENTITY",
84
+ "STATE": "MEAN_STD",
85
+ "ACTION": "MEAN_STD"
86
+ }
87
+ },
88
+ "state_file": "policy_preprocessor_step_5_normalizer_processor.safetensors"
89
+ }
90
+ ]
91
+ }
policy_preprocessor_step_5_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4da7407d16cb6217e0046c66ac4a2ad644b0a45c64ccfd61a9a51d32236d1db8
3
+ size 12652
train_config.json ADDED
@@ -0,0 +1,240 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "/home/alg/wyb/evo-rl-workspace/evo-rl-datasets/bi_s1_mix_pick_place_cube_fix_5",
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
+ "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": "pyav",
80
+ "streaming": false
81
+ },
82
+ "env": null,
83
+ "policy": {
84
+ "type": "pi0",
85
+ "n_obs_steps": 1,
86
+ "input_features": {
87
+ "observation.images.base_0_rgb": {
88
+ "type": "VISUAL",
89
+ "shape": [
90
+ 3,
91
+ 224,
92
+ 224
93
+ ]
94
+ },
95
+ "observation.images.left_wrist_0_rgb": {
96
+ "type": "VISUAL",
97
+ "shape": [
98
+ 3,
99
+ 224,
100
+ 224
101
+ ]
102
+ },
103
+ "observation.images.right_wrist_0_rgb": {
104
+ "type": "VISUAL",
105
+ "shape": [
106
+ 3,
107
+ 224,
108
+ 224
109
+ ]
110
+ },
111
+ "observation.state": {
112
+ "type": "STATE",
113
+ "shape": [
114
+ 32
115
+ ]
116
+ }
117
+ },
118
+ "output_features": {
119
+ "action": {
120
+ "type": "ACTION",
121
+ "shape": [
122
+ 14
123
+ ]
124
+ }
125
+ },
126
+ "device": "cuda",
127
+ "use_amp": false,
128
+ "use_peft": false,
129
+ "push_to_hub": false,
130
+ "repo_id": null,
131
+ "private": null,
132
+ "tags": null,
133
+ "license": null,
134
+ "pretrained_path": "/mnt/sdb/model/pi0_base/snapshots/26b99b9439acb1e352439e34ee9c67af0d76efa3",
135
+ "paligemma_variant": "gemma_2b",
136
+ "action_expert_variant": "gemma_300m",
137
+ "dtype": "bfloat16",
138
+ "chunk_size": 50,
139
+ "n_action_steps": 50,
140
+ "max_state_dim": 32,
141
+ "max_action_dim": 32,
142
+ "num_inference_steps": 10,
143
+ "time_sampling_beta_alpha": 1.5,
144
+ "time_sampling_beta_beta": 1.0,
145
+ "time_sampling_scale": 0.999,
146
+ "time_sampling_offset": 0.001,
147
+ "min_period": 0.004,
148
+ "max_period": 4.0,
149
+ "rtc_config": null,
150
+ "image_resolution": [
151
+ 224,
152
+ 224
153
+ ],
154
+ "empty_cameras": 0,
155
+ "normalization_mapping": {
156
+ "VISUAL": "IDENTITY",
157
+ "STATE": "MEAN_STD",
158
+ "ACTION": "MEAN_STD"
159
+ },
160
+ "gradient_checkpointing": true,
161
+ "compile_model": false,
162
+ "compile_mode": "max-autotune",
163
+ "freeze_vision_encoder": false,
164
+ "train_expert_only": false,
165
+ "optimizer_lr": 2.5e-05,
166
+ "optimizer_betas": [
167
+ 0.9,
168
+ 0.95
169
+ ],
170
+ "optimizer_eps": 1e-08,
171
+ "optimizer_weight_decay": 0.01,
172
+ "optimizer_grad_clip_norm": 1.0,
173
+ "scheduler_warmup_steps": 1000,
174
+ "scheduler_decay_steps": 30000,
175
+ "scheduler_decay_lr": 2.5e-06,
176
+ "tokenizer_max_length": 48
177
+ },
178
+ "output_dir": "outputs/policy_train/bi_s1_mix_pick_place_cube_fix_5_200episode_pi0",
179
+ "job_name": "bi_s1_mix_pick_place_cube_fix_5_200episode_pi0_train",
180
+ "resume": false,
181
+ "seed": 1000,
182
+ "num_workers": 4,
183
+ "batch_size": 32,
184
+ "steps": 200000,
185
+ "eval_freq": 20000,
186
+ "log_freq": 200,
187
+ "tolerance_s": 0.0001,
188
+ "save_checkpoint": true,
189
+ "save_freq": 5000,
190
+ "use_policy_training_preset": true,
191
+ "optimizer": {
192
+ "type": "adamw",
193
+ "lr": 2.5e-05,
194
+ "weight_decay": 0.01,
195
+ "grad_clip_norm": 1.0,
196
+ "betas": [
197
+ 0.9,
198
+ 0.95
199
+ ],
200
+ "eps": 1e-08
201
+ },
202
+ "scheduler": {
203
+ "type": "cosine_decay_with_warmup",
204
+ "num_warmup_steps": 1000,
205
+ "num_decay_steps": 30000,
206
+ "peak_lr": 2.5e-05,
207
+ "decay_lr": 2.5e-06
208
+ },
209
+ "eval": {
210
+ "n_episodes": 50,
211
+ "batch_size": 50,
212
+ "use_async_envs": false
213
+ },
214
+ "wandb": {
215
+ "enable": false,
216
+ "disable_artifact": false,
217
+ "project": "lerobot",
218
+ "entity": null,
219
+ "notes": null,
220
+ "run_id": null,
221
+ "mode": null
222
+ },
223
+ "peft": null,
224
+ "acp": {
225
+ "enable": false,
226
+ "indicator_field": "complementary_info.acp_indicator",
227
+ "indicator_dropout_prob": 0.0
228
+ },
229
+ "use_rabc": false,
230
+ "rabc_progress_path": null,
231
+ "rabc_kappa": 0.01,
232
+ "rabc_epsilon": 1e-06,
233
+ "rabc_head_mode": "sparse",
234
+ "rename_map": {
235
+ "observation.images.left": "observation.images.left_wrist_0_rgb",
236
+ "observation.images.middle": "observation.images.base_0_rgb",
237
+ "observation.images.right": "observation.images.right_wrist_0_rgb"
238
+ },
239
+ "checkpoint_path": null
240
+ }