Hugging Face's logo

Long27
/
pi0-pusht-v2

Safetensors
Model card Files
xet
 Community
Long27 commited on
Commit
dc57bcc
·
verified ·
1 Parent(s): b0e19e9

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +86 -0
  2. model.safetensors +3 -0
  3. policy_postprocessor.json +38 -0
  4. policy_postprocessor_step_0_unnormalizer_processor.safetensors +3 -0
  5. policy_preprocessor.json +83 -0
  6. policy_preprocessor_step_6_normalizer_processor.safetensors +3 -0
  7. train_config.json +224 -0
config.json ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "pi0",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.image": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 96,
10
+ 96
11
+ ]
12
+ },
13
+ "observation.state": {
14
+ "type": "STATE",
15
+ "shape": [
16
+ 2
17
+ ]
18
+ }
19
+ },
20
+ "output_features": {
21
+ "action": {
22
+ "type": "ACTION",
23
+ "shape": [
24
+ 2
25
+ ]
26
+ }
27
+ },
28
+ "device": "cuda",
29
+ "use_amp": false,
30
+ "use_peft": false,
31
+ "push_to_hub": true,
32
+ "repo_id": "long-hust/pi0-final",
33
+ "private": null,
34
+ "tags": null,
35
+ "license": null,
36
+ "pretrained_path": null,
37
+ "paligemma_variant": "gemma_2b",
38
+ "action_expert_variant": "gemma_300m",
39
+ "dtype": "bfloat16",
40
+ "chunk_size": 50,
41
+ "n_action_steps": 50,
42
+ "max_state_dim": 32,
43
+ "max_action_dim": 32,
44
+ "num_inference_steps": 10,
45
+ "time_sampling_beta_alpha": 1.5,
46
+ "time_sampling_beta_beta": 1.0,
47
+ "time_sampling_scale": 0.999,
48
+ "time_sampling_offset": 0.001,
49
+ "min_period": 0.004,
50
+ "max_period": 4.0,
51
+ "use_relative_actions": false,
52
+ "relative_exclude_joints": [
53
+ "gripper"
54
+ ],
55
+ "action_feature_names": [
56
+ "motors"
57
+ ],
58
+ "rtc_config": null,
59
+ "image_resolution": [
60
+ 224,
61
+ 224
62
+ ],
63
+ "empty_cameras": 0,
64
+ "normalization_mapping": {
65
+ "VISUAL": "IDENTITY",
66
+ "STATE": "MEAN_STD",
67
+ "ACTION": "MEAN_STD"
68
+ },
69
+ "gradient_checkpointing": false,
70
+ "compile_model": false,
71
+ "compile_mode": "max-autotune",
72
+ "freeze_vision_encoder": false,
73
+ "train_expert_only": false,
74
+ "optimizer_lr": 2.5e-05,
75
+ "optimizer_betas": [
76
+ 0.9,
77
+ 0.95
78
+ ],
79
+ "optimizer_eps": 1e-08,
80
+ "optimizer_weight_decay": 0.01,
81
+ "optimizer_grad_clip_norm": 1.0,
82
+ "scheduler_warmup_steps": 1000,
83
+ "scheduler_decay_steps": 30000,
84
+ "scheduler_decay_lr": 2.5e-06,
85
+ "tokenizer_max_length": 48
86
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b70e6f69c649c2d9537a91366cd0749bf95f414fef9de8da2fcbeb15bfad340
3
+ size 8892502944
policy_postprocessor.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ 2
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": "absolute_actions_processor",
26
+ "config": {
27
+ "enabled": false
28
+ }
29
+ },
30
+ {
31
+ "registry_name": "device_processor",
32
+ "config": {
33
+ "device": "cpu",
34
+ "float_dtype": null
35
+ }
36
+ }
37
+ ]
38
+ }
policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc174a7436ff5075efebe7be7cece5e5d8430658cc9b5314fb8a652eab912d96
3
+ size 4204
policy_preprocessor.json ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "pi0_new_line_processor",
16
+ "config": {}
17
+ },
18
+ {
19
+ "registry_name": "tokenizer_processor",
20
+ "config": {
21
+ "max_length": 48,
22
+ "task_key": "task",
23
+ "padding_side": "right",
24
+ "padding": "max_length",
25
+ "truncation": true,
26
+ "tokenizer_name": "google/paligemma-3b-pt-224"
27
+ }
28
+ },
29
+ {
30
+ "registry_name": "device_processor",
31
+ "config": {
32
+ "device": "cuda",
33
+ "float_dtype": null
34
+ }
35
+ },
36
+ {
37
+ "registry_name": "delta_actions_processor",
38
+ "config": {
39
+ "enabled": false,
40
+ "exclude_joints": [
41
+ "gripper"
42
+ ],
43
+ "action_names": [
44
+ "motors"
45
+ ]
46
+ }
47
+ },
48
+ {
49
+ "registry_name": "normalizer_processor",
50
+ "config": {
51
+ "eps": 1e-08,
52
+ "features": {
53
+ "observation.image": {
54
+ "type": "VISUAL",
55
+ "shape": [
56
+ 3,
57
+ 96,
58
+ 96
59
+ ]
60
+ },
61
+ "observation.state": {
62
+ "type": "STATE",
63
+ "shape": [
64
+ 2
65
+ ]
66
+ },
67
+ "action": {
68
+ "type": "ACTION",
69
+ "shape": [
70
+ 2
71
+ ]
72
+ }
73
+ },
74
+ "norm_map": {
75
+ "VISUAL": "IDENTITY",
76
+ "STATE": "MEAN_STD",
77
+ "ACTION": "MEAN_STD"
78
+ }
79
+ },
80
+ "state_file": "policy_preprocessor_step_6_normalizer_processor.safetensors"
81
+ }
82
+ ]
83
+ }
policy_preprocessor_step_6_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc174a7436ff5075efebe7be7cece5e5d8430658cc9b5314fb8a652eab912d96
3
+ size 4204
train_config.json ADDED
@@ -0,0 +1,224 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "lerobot/pusht",
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.image": {
88
+ "type": "VISUAL",
89
+ "shape": [
90
+ 3,
91
+ 96,
92
+ 96
93
+ ]
94
+ },
95
+ "observation.state": {
96
+ "type": "STATE",
97
+ "shape": [
98
+ 2
99
+ ]
100
+ }
101
+ },
102
+ "output_features": {
103
+ "action": {
104
+ "type": "ACTION",
105
+ "shape": [
106
+ 2
107
+ ]
108
+ }
109
+ },
110
+ "device": "cuda",
111
+ "use_amp": false,
112
+ "use_peft": false,
113
+ "push_to_hub": true,
114
+ "repo_id": "long-hust/pi0-final",
115
+ "private": null,
116
+ "tags": null,
117
+ "license": null,
118
+ "pretrained_path": null,
119
+ "paligemma_variant": "gemma_2b",
120
+ "action_expert_variant": "gemma_300m",
121
+ "dtype": "bfloat16",
122
+ "chunk_size": 50,
123
+ "n_action_steps": 50,
124
+ "max_state_dim": 32,
125
+ "max_action_dim": 32,
126
+ "num_inference_steps": 10,
127
+ "time_sampling_beta_alpha": 1.5,
128
+ "time_sampling_beta_beta": 1.0,
129
+ "time_sampling_scale": 0.999,
130
+ "time_sampling_offset": 0.001,
131
+ "min_period": 0.004,
132
+ "max_period": 4.0,
133
+ "use_relative_actions": false,
134
+ "relative_exclude_joints": [
135
+ "gripper"
136
+ ],
137
+ "action_feature_names": [
138
+ "motors"
139
+ ],
140
+ "rtc_config": null,
141
+ "image_resolution": [
142
+ 224,
143
+ 224
144
+ ],
145
+ "empty_cameras": 0,
146
+ "normalization_mapping": {
147
+ "VISUAL": "IDENTITY",
148
+ "STATE": "MEAN_STD",
149
+ "ACTION": "MEAN_STD"
150
+ },
151
+ "gradient_checkpointing": false,
152
+ "compile_model": false,
153
+ "compile_mode": "max-autotune",
154
+ "freeze_vision_encoder": false,
155
+ "train_expert_only": false,
156
+ "optimizer_lr": 2.5e-05,
157
+ "optimizer_betas": [
158
+ 0.9,
159
+ 0.95
160
+ ],
161
+ "optimizer_eps": 1e-08,
162
+ "optimizer_weight_decay": 0.01,
163
+ "optimizer_grad_clip_norm": 1.0,
164
+ "scheduler_warmup_steps": 1000,
165
+ "scheduler_decay_steps": 30000,
166
+ "scheduler_decay_lr": 2.5e-06,
167
+ "tokenizer_max_length": 48
168
+ },
169
+ "output_dir": "outputs/pi0_pusht",
170
+ "job_name": "pi0",
171
+ "resume": false,
172
+ "seed": 1000,
173
+ "cudnn_deterministic": false,
174
+ "num_workers": 4,
175
+ "batch_size": 1,
176
+ "steps": 3000,
177
+ "eval_freq": 20000,
178
+ "log_freq": 200,
179
+ "tolerance_s": 0.0001,
180
+ "save_checkpoint": true,
181
+ "save_freq": 20000,
182
+ "use_policy_training_preset": true,
183
+ "optimizer": {
184
+ "type": "adamw",
185
+ "lr": 2.5e-05,
186
+ "weight_decay": 0.01,
187
+ "grad_clip_norm": 1.0,
188
+ "betas": [
189
+ 0.9,
190
+ 0.95
191
+ ],
192
+ "eps": 1e-08
193
+ },
194
+ "scheduler": {
195
+ "type": "cosine_decay_with_warmup",
196
+ "num_warmup_steps": 1000,
197
+ "num_decay_steps": 30000,
198
+ "peak_lr": 2.5e-05,
199
+ "decay_lr": 2.5e-06
200
+ },
201
+ "eval": {
202
+ "n_episodes": 50,
203
+ "batch_size": 8,
204
+ "use_async_envs": true
205
+ },
206
+ "wandb": {
207
+ "enable": false,
208
+ "disable_artifact": false,
209
+ "project": "lerobot",
210
+ "entity": null,
211
+ "notes": null,
212
+ "run_id": null,
213
+ "mode": null,
214
+ "add_tags": true
215
+ },
216
+ "peft": null,
217
+ "use_rabc": false,
218
+ "rabc_progress_path": null,
219
+ "rabc_kappa": 0.01,
220
+ "rabc_epsilon": 1e-06,
221
+ "rabc_head_mode": "sparse",
222
+ "rename_map": {},
223
+ "checkpoint_path": null
224
+ }