Zasha01 commited on
Commit
a01176f
·
verified ·
1 Parent(s): 244a6ab

Upload folder using huggingface_hub

Browse files
config.json ADDED
@@ -0,0 +1,101 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "pi05",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.images.context": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 360,
10
+ 640
11
+ ]
12
+ },
13
+ "observation.images.right_wrist": {
14
+ "type": "VISUAL",
15
+ "shape": [
16
+ 3,
17
+ 360,
18
+ 640
19
+ ]
20
+ },
21
+ "observation.images.left_wrist": {
22
+ "type": "VISUAL",
23
+ "shape": [
24
+ 3,
25
+ 360,
26
+ 640
27
+ ]
28
+ },
29
+ "observation.state": {
30
+ "type": "STATE",
31
+ "shape": [
32
+ 14
33
+ ]
34
+ },
35
+ "observation.cartesian_state": {
36
+ "type": "STATE",
37
+ "shape": [
38
+ 14
39
+ ]
40
+ }
41
+ },
42
+ "output_features": {
43
+ "action": {
44
+ "type": "ACTION",
45
+ "shape": [
46
+ 30
47
+ ]
48
+ }
49
+ },
50
+ "device": "cuda",
51
+ "use_amp": false,
52
+ "use_peft": false,
53
+ "push_to_hub": true,
54
+ "repo_id": "Zasha01/towel_policy_pi",
55
+ "private": null,
56
+ "tags": null,
57
+ "license": null,
58
+ "pretrained_path": "lerobot/pi05_base",
59
+ "paligemma_variant": "gemma_2b",
60
+ "action_expert_variant": "gemma_300m",
61
+ "dtype": "bfloat16",
62
+ "chunk_size": 50,
63
+ "n_action_steps": 50,
64
+ "max_state_dim": 32,
65
+ "max_action_dim": 32,
66
+ "num_inference_steps": 10,
67
+ "time_sampling_beta_alpha": 1.5,
68
+ "time_sampling_beta_beta": 1.0,
69
+ "time_sampling_scale": 0.999,
70
+ "time_sampling_offset": 0.001,
71
+ "min_period": 0.004,
72
+ "max_period": 4.0,
73
+ "rtc_config": null,
74
+ "image_resolution": [
75
+ 224,
76
+ 224
77
+ ],
78
+ "empty_cameras": 0,
79
+ "tokenizer_max_length": 200,
80
+ "normalization_mapping": {
81
+ "VISUAL": "IDENTITY",
82
+ "STATE": "QUANTILES",
83
+ "ACTION": "QUANTILES"
84
+ },
85
+ "gradient_checkpointing": true,
86
+ "compile_model": true,
87
+ "compile_mode": "max-autotune",
88
+ "freeze_vision_encoder": false,
89
+ "train_expert_only": false,
90
+ "optimizer_lr": 2.5e-05,
91
+ "optimizer_betas": [
92
+ 0.9,
93
+ 0.95
94
+ ],
95
+ "optimizer_eps": 1e-08,
96
+ "optimizer_weight_decay": 0.01,
97
+ "optimizer_grad_clip_norm": 1.0,
98
+ "scheduler_warmup_steps": 1000,
99
+ "scheduler_decay_steps": 30000,
100
+ "scheduler_decay_lr": 2.5e-06
101
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85ab9026d93fbfe6e712706f80203af6592ca96ef1e8367b9eb670493fd462bf
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
+ 30
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:c725882288398a13c3f8cfe56b1a4ba8120ae33ea0974fdc14f5e0dcc870c299
3
+ size 14492
policy_preprocessor.json ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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.images.context": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 360,
24
+ 640
25
+ ]
26
+ },
27
+ "observation.images.right_wrist": {
28
+ "type": "VISUAL",
29
+ "shape": [
30
+ 3,
31
+ 360,
32
+ 640
33
+ ]
34
+ },
35
+ "observation.images.left_wrist": {
36
+ "type": "VISUAL",
37
+ "shape": [
38
+ 3,
39
+ 360,
40
+ 640
41
+ ]
42
+ },
43
+ "observation.state": {
44
+ "type": "STATE",
45
+ "shape": [
46
+ 14
47
+ ]
48
+ },
49
+ "observation.cartesian_state": {
50
+ "type": "STATE",
51
+ "shape": [
52
+ 14
53
+ ]
54
+ },
55
+ "action": {
56
+ "type": "ACTION",
57
+ "shape": [
58
+ 30
59
+ ]
60
+ }
61
+ },
62
+ "norm_map": {
63
+ "VISUAL": "IDENTITY",
64
+ "STATE": "QUANTILES",
65
+ "ACTION": "QUANTILES"
66
+ }
67
+ },
68
+ "state_file": "policy_preprocessor_step_2_normalizer_processor.safetensors"
69
+ },
70
+ {
71
+ "registry_name": "pi05_prepare_state_tokenizer_processor_step",
72
+ "config": {}
73
+ },
74
+ {
75
+ "registry_name": "tokenizer_processor",
76
+ "config": {
77
+ "max_length": 200,
78
+ "task_key": "task",
79
+ "padding_side": "right",
80
+ "padding": "max_length",
81
+ "truncation": true,
82
+ "tokenizer_name": "google/paligemma-3b-pt-224"
83
+ }
84
+ },
85
+ {
86
+ "registry_name": "device_processor",
87
+ "config": {
88
+ "device": "cuda",
89
+ "float_dtype": null
90
+ }
91
+ }
92
+ ]
93
+ }
policy_preprocessor_step_2_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c725882288398a13c3f8cfe56b1a4ba8120ae33ea0974fdc14f5e0dcc870c299
3
+ size 14492
train_config.json ADDED
@@ -0,0 +1,237 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "/home/ubuntu/train_pi/lerobot/towel",
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": "torchcodec",
80
+ "streaming": false
81
+ },
82
+ "env": null,
83
+ "policy": {
84
+ "type": "pi05",
85
+ "n_obs_steps": 1,
86
+ "input_features": {
87
+ "observation.images.context": {
88
+ "type": "VISUAL",
89
+ "shape": [
90
+ 3,
91
+ 360,
92
+ 640
93
+ ]
94
+ },
95
+ "observation.images.right_wrist": {
96
+ "type": "VISUAL",
97
+ "shape": [
98
+ 3,
99
+ 360,
100
+ 640
101
+ ]
102
+ },
103
+ "observation.images.left_wrist": {
104
+ "type": "VISUAL",
105
+ "shape": [
106
+ 3,
107
+ 360,
108
+ 640
109
+ ]
110
+ },
111
+ "observation.state": {
112
+ "type": "STATE",
113
+ "shape": [
114
+ 14
115
+ ]
116
+ },
117
+ "observation.cartesian_state": {
118
+ "type": "STATE",
119
+ "shape": [
120
+ 14
121
+ ]
122
+ }
123
+ },
124
+ "output_features": {
125
+ "action": {
126
+ "type": "ACTION",
127
+ "shape": [
128
+ 30
129
+ ]
130
+ }
131
+ },
132
+ "device": "cuda",
133
+ "use_amp": false,
134
+ "use_peft": false,
135
+ "push_to_hub": true,
136
+ "repo_id": "Zasha01/towel_policy_pi",
137
+ "private": null,
138
+ "tags": null,
139
+ "license": null,
140
+ "pretrained_path": "lerobot/pi05_base",
141
+ "paligemma_variant": "gemma_2b",
142
+ "action_expert_variant": "gemma_300m",
143
+ "dtype": "bfloat16",
144
+ "chunk_size": 50,
145
+ "n_action_steps": 50,
146
+ "max_state_dim": 32,
147
+ "max_action_dim": 32,
148
+ "num_inference_steps": 10,
149
+ "time_sampling_beta_alpha": 1.5,
150
+ "time_sampling_beta_beta": 1.0,
151
+ "time_sampling_scale": 0.999,
152
+ "time_sampling_offset": 0.001,
153
+ "min_period": 0.004,
154
+ "max_period": 4.0,
155
+ "rtc_config": null,
156
+ "image_resolution": [
157
+ 224,
158
+ 224
159
+ ],
160
+ "empty_cameras": 0,
161
+ "tokenizer_max_length": 200,
162
+ "normalization_mapping": {
163
+ "VISUAL": "IDENTITY",
164
+ "STATE": "QUANTILES",
165
+ "ACTION": "QUANTILES"
166
+ },
167
+ "gradient_checkpointing": true,
168
+ "compile_model": true,
169
+ "compile_mode": "max-autotune",
170
+ "freeze_vision_encoder": false,
171
+ "train_expert_only": false,
172
+ "optimizer_lr": 2.5e-05,
173
+ "optimizer_betas": [
174
+ 0.9,
175
+ 0.95
176
+ ],
177
+ "optimizer_eps": 1e-08,
178
+ "optimizer_weight_decay": 0.01,
179
+ "optimizer_grad_clip_norm": 1.0,
180
+ "scheduler_warmup_steps": 1000,
181
+ "scheduler_decay_steps": 30000,
182
+ "scheduler_decay_lr": 2.5e-06
183
+ },
184
+ "output_dir": "outputs/pi05_training_towel",
185
+ "job_name": "pi05_training_towel",
186
+ "resume": false,
187
+ "seed": 1000,
188
+ "num_workers": 4,
189
+ "batch_size": 16,
190
+ "steps": 20000,
191
+ "eval_freq": 20000,
192
+ "log_freq": 200,
193
+ "tolerance_s": 0.0001,
194
+ "save_checkpoint": true,
195
+ "save_freq": 3000,
196
+ "use_policy_training_preset": true,
197
+ "optimizer": {
198
+ "type": "adamw",
199
+ "lr": 2.5e-05,
200
+ "weight_decay": 0.01,
201
+ "grad_clip_norm": 1.0,
202
+ "betas": [
203
+ 0.9,
204
+ 0.95
205
+ ],
206
+ "eps": 1e-08
207
+ },
208
+ "scheduler": {
209
+ "type": "cosine_decay_with_warmup",
210
+ "num_warmup_steps": 1000,
211
+ "num_decay_steps": 30000,
212
+ "peak_lr": 2.5e-05,
213
+ "decay_lr": 2.5e-06
214
+ },
215
+ "eval": {
216
+ "n_episodes": 50,
217
+ "batch_size": 50,
218
+ "use_async_envs": false
219
+ },
220
+ "wandb": {
221
+ "enable": true,
222
+ "disable_artifact": false,
223
+ "project": "lerobot",
224
+ "entity": null,
225
+ "notes": null,
226
+ "run_id": "9loxypqj",
227
+ "mode": null
228
+ },
229
+ "peft": null,
230
+ "use_rabc": false,
231
+ "rabc_progress_path": null,
232
+ "rabc_kappa": 0.01,
233
+ "rabc_epsilon": 1e-06,
234
+ "rabc_head_mode": "sparse",
235
+ "rename_map": {},
236
+ "checkpoint_path": null
237
+ }