Hugging Face's logo

K-vr
/
cube_stack3_pi0_abs_full_batch8_dagger

Robotics
LeRobot
Safetensors
pi0
Model card Files
xet
 Community
K-vr commited on
Commit
6ab99a3
·
verified ·
1 Parent(s): 028bfc3

Upload trained policy checkpoint

Browse files
Files changed (7) hide show
  1. config.json +116 -0
  2. model.safetensors +3 -0
  3. policy_postprocessor.json +39 -0
  4. policy_postprocessor_step_0_unnormalizer_processor.safetensors +3 -0
  5. policy_preprocessor.json +102 -0
  6. policy_preprocessor_step_6_normalizer_processor.safetensors +3 -0
  7. train_config.json +273 -0
config.json ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": true,
48
+ "repo_id": "K-vr/cube_stack3_Pi0_abs_full_batch8_dagger",
49
+ "private": false,
50
+ "tags": null,
51
+ "license": null,
52
+ "pretrained_path": "K-vr/cube_stack3_Pi0_abs_full_batch8",
53
+ "pretrained_revision": null,
54
+ "paligemma_variant": "gemma_2b",
55
+ "action_expert_variant": "gemma_300m",
56
+ "dtype": "bfloat16",
57
+ "chunk_size": 50,
58
+ "n_action_steps": 50,
59
+ "max_state_dim": 32,
60
+ "max_action_dim": 32,
61
+ "num_inference_steps": 10,
62
+ "time_sampling_beta_alpha": 1.5,
63
+ "time_sampling_beta_beta": 1.0,
64
+ "time_sampling_scale": 0.999,
65
+ "time_sampling_offset": 0.001,
66
+ "min_period": 0.004,
67
+ "max_period": 4.0,
68
+ "use_relative_actions": false,
69
+ "relative_exclude_joints": [
70
+ "gripper"
71
+ ],
72
+ "action_feature_names": [
73
+ "left_shoulder_pan.pos",
74
+ "left_shoulder_lift.pos",
75
+ "left_elbow_flex.pos",
76
+ "left_forearm_roll.pos",
77
+ "left_wrist_flex.pos",
78
+ "left_wrist_roll.pos",
79
+ "left_gripper.pos",
80
+ "right_shoulder_pan.pos",
81
+ "right_shoulder_lift.pos",
82
+ "right_elbow_flex.pos",
83
+ "right_forearm_roll.pos",
84
+ "right_wrist_flex.pos",
85
+ "right_wrist_roll.pos",
86
+ "right_gripper.pos"
87
+ ],
88
+ "rtc_config": null,
89
+ "image_resolution": [
90
+ 224,
91
+ 224
92
+ ],
93
+ "empty_cameras": 0,
94
+ "normalization_mapping": {
95
+ "VISUAL": "IDENTITY",
96
+ "STATE": "MEAN_STD",
97
+ "ACTION": "MEAN_STD"
98
+ },
99
+ "gradient_checkpointing": true,
100
+ "compile_model": false,
101
+ "compile_mode": "max-autotune",
102
+ "freeze_vision_encoder": false,
103
+ "train_expert_only": false,
104
+ "optimizer_lr": 2.5e-05,
105
+ "optimizer_betas": [
106
+ 0.9,
107
+ 0.95
108
+ ],
109
+ "optimizer_eps": 1e-08,
110
+ "optimizer_weight_decay": 0.01,
111
+ "optimizer_grad_clip_norm": 1.0,
112
+ "scheduler_warmup_steps": 1000,
113
+ "scheduler_decay_steps": 30000,
114
+ "scheduler_decay_lr": 2.5e-06,
115
+ "tokenizer_max_length": 48
116
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:668cf03d5484aef06616b53cd6da3a6047376fdca2cea7a02cc688461d7f0573
3
+ size 8892502944
policy_postprocessor.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "absolute_actions_processor",
26
+ "config": {
27
+ "enabled": false,
28
+ "execution_steps": 1
29
+ }
30
+ },
31
+ {
32
+ "registry_name": "device_processor",
33
+ "config": {
34
+ "device": "cpu",
35
+ "float_dtype": null
36
+ }
37
+ }
38
+ ]
39
+ }
policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f484d1003413c7d19fea72c507ae9802c046466b0854aa16a6933103fcfdc65a
3
+ size 11344
policy_preprocessor.json ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "policy_preprocessor",
3
+ "steps": [
4
+ {
5
+ "registry_name": "rename_observations_processor",
6
+ "config": {
7
+ "rename_map": {
8
+ "observation.images.left_cam_scene": "observation.images.base_0_rgb",
9
+ "observation.images.left_cam_left": "observation.images.left_wrist_0_rgb",
10
+ "observation.images.right_cam_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": "relative_actions_processor",
42
+ "config": {
43
+ "enabled": false,
44
+ "exclude_joints": [
45
+ "gripper"
46
+ ],
47
+ "action_names": null,
48
+ "state_index": null
49
+ }
50
+ },
51
+ {
52
+ "registry_name": "normalizer_processor",
53
+ "config": {
54
+ "eps": 1e-08,
55
+ "features": {
56
+ "observation.images.base_0_rgb": {
57
+ "type": "VISUAL",
58
+ "shape": [
59
+ 3,
60
+ 224,
61
+ 224
62
+ ]
63
+ },
64
+ "observation.images.left_wrist_0_rgb": {
65
+ "type": "VISUAL",
66
+ "shape": [
67
+ 3,
68
+ 224,
69
+ 224
70
+ ]
71
+ },
72
+ "observation.images.right_wrist_0_rgb": {
73
+ "type": "VISUAL",
74
+ "shape": [
75
+ 3,
76
+ 224,
77
+ 224
78
+ ]
79
+ },
80
+ "observation.state": {
81
+ "type": "STATE",
82
+ "shape": [
83
+ 32
84
+ ]
85
+ },
86
+ "action": {
87
+ "type": "ACTION",
88
+ "shape": [
89
+ 14
90
+ ]
91
+ }
92
+ },
93
+ "norm_map": {
94
+ "VISUAL": "IDENTITY",
95
+ "STATE": "MEAN_STD",
96
+ "ACTION": "MEAN_STD"
97
+ }
98
+ },
99
+ "state_file": "policy_preprocessor_step_6_normalizer_processor.safetensors"
100
+ }
101
+ ]
102
+ }
policy_preprocessor_step_6_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f484d1003413c7d19fea72c507ae9802c046466b0854aa16a6933103fcfdc65a
3
+ size 11344
train_config.json ADDED
@@ -0,0 +1,273 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "K-vr/cube_stack3_Pi0_abs_batch8_combined",
4
+ "root": "/cube_stack3_Pi0_abs_batch8_combined",
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
+ "return_uint8": false,
81
+ "depth_output_unit": "mm",
82
+ "streaming": false,
83
+ "eval_split": 0.0
84
+ },
85
+ "env": null,
86
+ "policy": {
87
+ "type": "pi0",
88
+ "n_obs_steps": 1,
89
+ "input_features": {
90
+ "observation.images.base_0_rgb": {
91
+ "type": "VISUAL",
92
+ "shape": [
93
+ 3,
94
+ 224,
95
+ 224
96
+ ]
97
+ },
98
+ "observation.images.left_wrist_0_rgb": {
99
+ "type": "VISUAL",
100
+ "shape": [
101
+ 3,
102
+ 224,
103
+ 224
104
+ ]
105
+ },
106
+ "observation.images.right_wrist_0_rgb": {
107
+ "type": "VISUAL",
108
+ "shape": [
109
+ 3,
110
+ 224,
111
+ 224
112
+ ]
113
+ },
114
+ "observation.state": {
115
+ "type": "STATE",
116
+ "shape": [
117
+ 32
118
+ ]
119
+ }
120
+ },
121
+ "output_features": {
122
+ "action": {
123
+ "type": "ACTION",
124
+ "shape": [
125
+ 14
126
+ ]
127
+ }
128
+ },
129
+ "device": "cuda",
130
+ "use_amp": false,
131
+ "use_peft": false,
132
+ "push_to_hub": true,
133
+ "repo_id": "K-vr/cube_stack3_Pi0_abs_full_batch8_dagger",
134
+ "private": false,
135
+ "tags": null,
136
+ "license": null,
137
+ "pretrained_path": "K-vr/cube_stack3_Pi0_abs_full_batch8",
138
+ "pretrained_revision": null,
139
+ "paligemma_variant": "gemma_2b",
140
+ "action_expert_variant": "gemma_300m",
141
+ "dtype": "bfloat16",
142
+ "chunk_size": 50,
143
+ "n_action_steps": 50,
144
+ "max_state_dim": 32,
145
+ "max_action_dim": 32,
146
+ "num_inference_steps": 10,
147
+ "time_sampling_beta_alpha": 1.5,
148
+ "time_sampling_beta_beta": 1.0,
149
+ "time_sampling_scale": 0.999,
150
+ "time_sampling_offset": 0.001,
151
+ "min_period": 0.004,
152
+ "max_period": 4.0,
153
+ "use_relative_actions": false,
154
+ "relative_exclude_joints": [
155
+ "gripper"
156
+ ],
157
+ "action_feature_names": [
158
+ "left_shoulder_pan.pos",
159
+ "left_shoulder_lift.pos",
160
+ "left_elbow_flex.pos",
161
+ "left_forearm_roll.pos",
162
+ "left_wrist_flex.pos",
163
+ "left_wrist_roll.pos",
164
+ "left_gripper.pos",
165
+ "right_shoulder_pan.pos",
166
+ "right_shoulder_lift.pos",
167
+ "right_elbow_flex.pos",
168
+ "right_forearm_roll.pos",
169
+ "right_wrist_flex.pos",
170
+ "right_wrist_roll.pos",
171
+ "right_gripper.pos"
172
+ ],
173
+ "rtc_config": null,
174
+ "image_resolution": [
175
+ 224,
176
+ 224
177
+ ],
178
+ "empty_cameras": 0,
179
+ "normalization_mapping": {
180
+ "VISUAL": "IDENTITY",
181
+ "STATE": "MEAN_STD",
182
+ "ACTION": "MEAN_STD"
183
+ },
184
+ "gradient_checkpointing": true,
185
+ "compile_model": false,
186
+ "compile_mode": "max-autotune",
187
+ "freeze_vision_encoder": false,
188
+ "train_expert_only": false,
189
+ "optimizer_lr": 2.5e-05,
190
+ "optimizer_betas": [
191
+ 0.9,
192
+ 0.95
193
+ ],
194
+ "optimizer_eps": 1e-08,
195
+ "optimizer_weight_decay": 0.01,
196
+ "optimizer_grad_clip_norm": 1.0,
197
+ "scheduler_warmup_steps": 1000,
198
+ "scheduler_decay_steps": 30000,
199
+ "scheduler_decay_lr": 2.5e-06,
200
+ "tokenizer_max_length": 48
201
+ },
202
+ "reward_model": null,
203
+ "output_dir": "/workspace/output/pi0-training-9409629",
204
+ "job_name": "pi0-training-9409629",
205
+ "resume": false,
206
+ "seed": 1000,
207
+ "cudnn_deterministic": false,
208
+ "num_workers": 2,
209
+ "batch_size": 8,
210
+ "prefetch_factor": 4,
211
+ "persistent_workers": true,
212
+ "steps": 20000,
213
+ "env_eval_freq": 20000,
214
+ "log_freq": 500,
215
+ "eval_steps": 0,
216
+ "max_eval_samples": 0,
217
+ "tolerance_s": 0.0001,
218
+ "save_checkpoint": true,
219
+ "save_freq": 4000,
220
+ "use_policy_training_preset": true,
221
+ "optimizer": {
222
+ "type": "adamw",
223
+ "lr": 2.5e-05,
224
+ "weight_decay": 0.01,
225
+ "grad_clip_norm": 1.0,
226
+ "betas": [
227
+ 0.9,
228
+ 0.95
229
+ ],
230
+ "eps": 1e-08
231
+ },
232
+ "scheduler": {
233
+ "type": "cosine_decay_with_warmup",
234
+ "num_warmup_steps": 1000,
235
+ "num_decay_steps": 30000,
236
+ "peak_lr": 2.5e-05,
237
+ "decay_lr": 2.5e-06
238
+ },
239
+ "eval": {
240
+ "n_episodes": 50,
241
+ "batch_size": 50,
242
+ "use_async_envs": true,
243
+ "recording": false,
244
+ "recording_repo_id": null,
245
+ "recording_private": true
246
+ },
247
+ "wandb": {
248
+ "enable": false,
249
+ "disable_artifact": false,
250
+ "project": "lerobot",
251
+ "entity": null,
252
+ "notes": null,
253
+ "run_id": null,
254
+ "mode": null,
255
+ "add_tags": true
256
+ },
257
+ "peft": null,
258
+ "job": {
259
+ "target": null,
260
+ "image": "huggingface/lerobot-gpu:latest",
261
+ "timeout": "2d",
262
+ "detach": false,
263
+ "tags": []
264
+ },
265
+ "save_checkpoint_to_hub": false,
266
+ "sample_weighting": null,
267
+ "rename_map": {
268
+ "observation.images.left_cam_scene": "observation.images.base_0_rgb",
269
+ "observation.images.left_cam_left": "observation.images.left_wrist_0_rgb",
270
+ "observation.images.right_cam_right": "observation.images.right_wrist_0_rgb"
271
+ },
272
+ "checkpoint_path": null
273
+ }