Sendera commited on
Commit
06cb0da
·
verified ·
1 Parent(s): 83249b4

Upload folder using huggingface_hub

Browse files
config.json ADDED
@@ -0,0 +1,103 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "pi05",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 7
9
+ ]
10
+ },
11
+ "observation.images.Cam1_global": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 256,
16
+ 256
17
+ ]
18
+ },
19
+ "observation.images.Cam1_local": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 256,
24
+ 256
25
+ ]
26
+ },
27
+ "observation.images.Cam2_global": {
28
+ "type": "VISUAL",
29
+ "shape": [
30
+ 3,
31
+ 256,
32
+ 256
33
+ ]
34
+ },
35
+ "observation.images.Cam2_local": {
36
+ "type": "VISUAL",
37
+ "shape": [
38
+ 3,
39
+ 256,
40
+ 256
41
+ ]
42
+ }
43
+ },
44
+ "output_features": {
45
+ "action": {
46
+ "type": "ACTION",
47
+ "shape": [
48
+ 14
49
+ ]
50
+ }
51
+ },
52
+ "device": "cuda",
53
+ "use_amp": false,
54
+ "use_peft": false,
55
+ "push_to_hub": false,
56
+ "repo_id": null,
57
+ "private": null,
58
+ "tags": null,
59
+ "license": null,
60
+ "pretrained_path": null,
61
+ "paligemma_variant": "gemma_2b",
62
+ "action_expert_variant": "gemma_300m",
63
+ "dtype": "bfloat16",
64
+ "chunk_size": 50,
65
+ "n_action_steps": 50,
66
+ "max_state_dim": 32,
67
+ "max_action_dim": 32,
68
+ "num_inference_steps": 10,
69
+ "time_sampling_beta_alpha": 1.5,
70
+ "time_sampling_beta_beta": 1.0,
71
+ "time_sampling_scale": 0.999,
72
+ "time_sampling_offset": 0.001,
73
+ "min_period": 0.004,
74
+ "max_period": 4.0,
75
+ "rtc_config": null,
76
+ "image_resolution": [
77
+ 224,
78
+ 224
79
+ ],
80
+ "empty_cameras": 0,
81
+ "tokenizer_max_length": 200,
82
+ "normalization_mapping": {
83
+ "VISUAL": "IDENTITY",
84
+ "STATE": "QUANTILES",
85
+ "ACTION": "QUANTILES"
86
+ },
87
+ "gradient_checkpointing": true,
88
+ "compile_model": false,
89
+ "compile_mode": "max-autotune",
90
+ "freeze_vision_encoder": false,
91
+ "train_expert_only": false,
92
+ "optimizer_lr": 2.5e-05,
93
+ "optimizer_betas": [
94
+ 0.9,
95
+ 0.95
96
+ ],
97
+ "optimizer_eps": 1e-08,
98
+ "optimizer_weight_decay": 0.01,
99
+ "optimizer_grad_clip_norm": 1.0,
100
+ "scheduler_warmup_steps": 1000,
101
+ "scheduler_decay_steps": 30000,
102
+ "scheduler_decay_lr": 2.5e-06
103
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95aadd8cec8d98343988f7bceeb89a71cb85aee8387998523e3767d3d98f90a3
3
+ size 7473096344
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": "QUANTILES",
19
+ "ACTION": "QUANTILES"
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 sha256:4aa139a5c3524af4ccb19186e7e4663cbe027ac8d66dd5ea40a3e14d84cc84e1
3
+ size 10164
policy_preprocessor.json ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "normalizer_processor",
16
+ "config": {
17
+ "eps": 1e-08,
18
+ "features": {
19
+ "observation.state": {
20
+ "type": "STATE",
21
+ "shape": [
22
+ 7
23
+ ]
24
+ },
25
+ "observation.images.Cam1_global": {
26
+ "type": "VISUAL",
27
+ "shape": [
28
+ 3,
29
+ 256,
30
+ 256
31
+ ]
32
+ },
33
+ "observation.images.Cam1_local": {
34
+ "type": "VISUAL",
35
+ "shape": [
36
+ 3,
37
+ 256,
38
+ 256
39
+ ]
40
+ },
41
+ "observation.images.Cam2_global": {
42
+ "type": "VISUAL",
43
+ "shape": [
44
+ 3,
45
+ 256,
46
+ 256
47
+ ]
48
+ },
49
+ "observation.images.Cam2_local": {
50
+ "type": "VISUAL",
51
+ "shape": [
52
+ 3,
53
+ 256,
54
+ 256
55
+ ]
56
+ },
57
+ "action": {
58
+ "type": "ACTION",
59
+ "shape": [
60
+ 14
61
+ ]
62
+ }
63
+ },
64
+ "norm_map": {
65
+ "VISUAL": "IDENTITY",
66
+ "STATE": "QUANTILES",
67
+ "ACTION": "QUANTILES"
68
+ }
69
+ },
70
+ "state_file": "policy_preprocessor_step_2_normalizer_processor.safetensors"
71
+ },
72
+ {
73
+ "registry_name": "pi05_prepare_state_tokenizer_processor_step",
74
+ "config": {}
75
+ },
76
+ {
77
+ "registry_name": "tokenizer_processor",
78
+ "config": {
79
+ "max_length": 200,
80
+ "task_key": "task",
81
+ "padding_side": "right",
82
+ "padding": "max_length",
83
+ "truncation": true,
84
+ "tokenizer_name": "google/paligemma-3b-pt-224"
85
+ }
86
+ },
87
+ {
88
+ "registry_name": "device_processor",
89
+ "config": {
90
+ "device": "cuda",
91
+ "float_dtype": null
92
+ }
93
+ }
94
+ ]
95
+ }
policy_preprocessor_step_2_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4aa139a5c3524af4ccb19186e7e4663cbe027ac8d66dd5ea40a3e14d84cc84e1
3
+ size 10164
train_config.json ADDED
@@ -0,0 +1,255 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "Sendera/pick2",
4
+ "root": "/srv/scratch/z5649701/datasets/Sendera/pick2",
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
+ "resize": {
76
+ "weight": 0.0,
77
+ "type": "Resize",
78
+ "kwargs": {
79
+ "size": 512,
80
+ "interpolation": 2,
81
+ "antialias": true
82
+ }
83
+ },
84
+ "center_crop": {
85
+ "weight": 0.0,
86
+ "type": "CenterCrop",
87
+ "kwargs": {
88
+ "size": 512
89
+ }
90
+ }
91
+ }
92
+ },
93
+ "revision": null,
94
+ "use_imagenet_stats": true,
95
+ "video_backend": "pyav",
96
+ "streaming": false
97
+ },
98
+ "env": null,
99
+ "policy": {
100
+ "type": "pi05",
101
+ "n_obs_steps": 1,
102
+ "input_features": {
103
+ "observation.state": {
104
+ "type": "STATE",
105
+ "shape": [
106
+ 7
107
+ ]
108
+ },
109
+ "observation.images.Cam1_global": {
110
+ "type": "VISUAL",
111
+ "shape": [
112
+ 3,
113
+ 256,
114
+ 256
115
+ ]
116
+ },
117
+ "observation.images.Cam1_local": {
118
+ "type": "VISUAL",
119
+ "shape": [
120
+ 3,
121
+ 256,
122
+ 256
123
+ ]
124
+ },
125
+ "observation.images.Cam2_global": {
126
+ "type": "VISUAL",
127
+ "shape": [
128
+ 3,
129
+ 256,
130
+ 256
131
+ ]
132
+ },
133
+ "observation.images.Cam2_local": {
134
+ "type": "VISUAL",
135
+ "shape": [
136
+ 3,
137
+ 256,
138
+ 256
139
+ ]
140
+ }
141
+ },
142
+ "output_features": {
143
+ "action": {
144
+ "type": "ACTION",
145
+ "shape": [
146
+ 14
147
+ ]
148
+ }
149
+ },
150
+ "device": "cuda",
151
+ "use_amp": false,
152
+ "use_peft": false,
153
+ "push_to_hub": false,
154
+ "repo_id": null,
155
+ "private": null,
156
+ "tags": null,
157
+ "license": null,
158
+ "pretrained_path": null,
159
+ "paligemma_variant": "gemma_2b",
160
+ "action_expert_variant": "gemma_300m",
161
+ "dtype": "bfloat16",
162
+ "chunk_size": 50,
163
+ "n_action_steps": 50,
164
+ "max_state_dim": 32,
165
+ "max_action_dim": 32,
166
+ "num_inference_steps": 10,
167
+ "time_sampling_beta_alpha": 1.5,
168
+ "time_sampling_beta_beta": 1.0,
169
+ "time_sampling_scale": 0.999,
170
+ "time_sampling_offset": 0.001,
171
+ "min_period": 0.004,
172
+ "max_period": 4.0,
173
+ "rtc_config": null,
174
+ "image_resolution": [
175
+ 224,
176
+ 224
177
+ ],
178
+ "empty_cameras": 0,
179
+ "tokenizer_max_length": 200,
180
+ "normalization_mapping": {
181
+ "VISUAL": "IDENTITY",
182
+ "STATE": "QUANTILES",
183
+ "ACTION": "QUANTILES"
184
+ },
185
+ "gradient_checkpointing": true,
186
+ "compile_model": false,
187
+ "compile_mode": "max-autotune",
188
+ "freeze_vision_encoder": false,
189
+ "train_expert_only": false,
190
+ "optimizer_lr": 2.5e-05,
191
+ "optimizer_betas": [
192
+ 0.9,
193
+ 0.95
194
+ ],
195
+ "optimizer_eps": 1e-08,
196
+ "optimizer_weight_decay": 0.01,
197
+ "optimizer_grad_clip_norm": 1.0,
198
+ "scheduler_warmup_steps": 1000,
199
+ "scheduler_decay_steps": 30000,
200
+ "scheduler_decay_lr": 2.5e-06
201
+ },
202
+ "output_dir": "outputs/train/2026-02-06/21-20-56_pi05_pick2_h200_full",
203
+ "job_name": "pi05_pick2_h200_full",
204
+ "resume": false,
205
+ "seed": 1000,
206
+ "num_workers": 8,
207
+ "batch_size": 8,
208
+ "steps": 250000,
209
+ "eval_freq": 20000,
210
+ "log_freq": 200,
211
+ "tolerance_s": 0.0001,
212
+ "save_checkpoint": true,
213
+ "save_freq": 5000,
214
+ "use_policy_training_preset": true,
215
+ "optimizer": {
216
+ "type": "adamw",
217
+ "lr": 2.5e-05,
218
+ "weight_decay": 0.01,
219
+ "grad_clip_norm": 1.0,
220
+ "betas": [
221
+ 0.9,
222
+ 0.95
223
+ ],
224
+ "eps": 1e-08
225
+ },
226
+ "scheduler": {
227
+ "type": "cosine_decay_with_warmup",
228
+ "num_warmup_steps": 1000,
229
+ "num_decay_steps": 30000,
230
+ "peak_lr": 2.5e-05,
231
+ "decay_lr": 2.5e-06
232
+ },
233
+ "eval": {
234
+ "n_episodes": 50,
235
+ "batch_size": 50,
236
+ "use_async_envs": false
237
+ },
238
+ "wandb": {
239
+ "enable": true,
240
+ "disable_artifact": false,
241
+ "project": "lerobot_leo",
242
+ "entity": null,
243
+ "notes": null,
244
+ "run_id": "gd6vmkw4",
245
+ "mode": null
246
+ },
247
+ "peft": null,
248
+ "use_rabc": false,
249
+ "rabc_progress_path": null,
250
+ "rabc_kappa": 0.01,
251
+ "rabc_epsilon": 1e-06,
252
+ "rabc_head_mode": "sparse",
253
+ "rename_map": {},
254
+ "checkpoint_path": null
255
+ }