Hugging Face's logo

usr256864
/
pi05_quant_1000

Safetensors
Model card Files
xet
 Community
usr256864 commited on
Commit
d5f20be
·
verified ·
1 Parent(s): 8b9b02e

Initial model upload

Browse files
Files changed (7) hide show
  1. config.json +79 -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 +71 -0
  6. policy_preprocessor_step_2_normalizer_processor.safetensors +3 -0
  7. train_config.json +260 -0
config.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "pi05",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.image": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 256,
10
+ 256
11
+ ]
12
+ },
13
+ "observation.state": {
14
+ "type": "STATE",
15
+ "shape": [
16
+ 32
17
+ ]
18
+ }
19
+ },
20
+ "output_features": {
21
+ "action": {
22
+ "type": "ACTION",
23
+ "shape": [
24
+ 32
25
+ ]
26
+ }
27
+ },
28
+ "device": "cuda",
29
+ "use_amp": false,
30
+ "use_peft": false,
31
+ "push_to_hub": false,
32
+ "repo_id": "usr256864/pi0",
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
+ "rtc_config": null,
52
+ "image_resolution": [
53
+ 224,
54
+ 224
55
+ ],
56
+ "empty_cameras": 0,
57
+ "tokenizer_max_length": 200,
58
+ "normalization_mapping": {
59
+ "VISUAL": "IDENTITY",
60
+ "STATE": "QUANTILES",
61
+ "ACTION": "QUANTILES"
62
+ },
63
+ "gradient_checkpointing": true,
64
+ "compile_model": false,
65
+ "compile_mode": "max-autotune",
66
+ "freeze_vision_encoder": false,
67
+ "train_expert_only": false,
68
+ "optimizer_lr": 2.5e-05,
69
+ "optimizer_betas": [
70
+ 0.9,
71
+ 0.95
72
+ ],
73
+ "optimizer_eps": 1e-08,
74
+ "optimizer_weight_decay": 0.01,
75
+ "optimizer_grad_clip_norm": 1.0,
76
+ "scheduler_warmup_steps": 1000,
77
+ "scheduler_decay_steps": 30000,
78
+ "scheduler_decay_lr": 2.5e-06
79
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b9e88097fb348edfbf43f753f835c4a63359b3dfdc49760bc8ecab159e54d45
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
+ 32
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 sha256:9519068ba0094384fc7a499a7eb7fefdb58c5f74f3db6c1360bf34ea2b6b21e6
3
+ size 8528
policy_preprocessor.json ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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.image": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 256,
24
+ 256
25
+ ]
26
+ },
27
+ "observation.state": {
28
+ "type": "STATE",
29
+ "shape": [
30
+ 32
31
+ ]
32
+ },
33
+ "action": {
34
+ "type": "ACTION",
35
+ "shape": [
36
+ 32
37
+ ]
38
+ }
39
+ },
40
+ "norm_map": {
41
+ "VISUAL": "IDENTITY",
42
+ "STATE": "QUANTILES",
43
+ "ACTION": "QUANTILES"
44
+ }
45
+ },
46
+ "state_file": "policy_preprocessor_step_2_normalizer_processor.safetensors"
47
+ },
48
+ {
49
+ "registry_name": "pi05_prepare_state_tokenizer_processor_step",
50
+ "config": {}
51
+ },
52
+ {
53
+ "registry_name": "tokenizer_processor",
54
+ "config": {
55
+ "max_length": 200,
56
+ "task_key": "task",
57
+ "padding_side": "right",
58
+ "padding": "max_length",
59
+ "truncation": true,
60
+ "tokenizer_name": "google/paligemma-3b-pt-224"
61
+ }
62
+ },
63
+ {
64
+ "registry_name": "device_processor",
65
+ "config": {
66
+ "device": "cuda",
67
+ "float_dtype": null
68
+ }
69
+ }
70
+ ]
71
+ }
policy_preprocessor_step_2_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9519068ba0094384fc7a499a7eb7fefdb58c5f74f3db6c1360bf34ea2b6b21e6
3
+ size 8528
train_config.json ADDED
@@ -0,0 +1,260 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "usr256864/lr_ds",
4
+ "root": "/netscratch/mboukheddimi/RH5_dataset_threads/",
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": {
83
+ "type": "rh5",
84
+ "task": "walk",
85
+ "fps": 30,
86
+ "features": {
87
+ "action": {
88
+ "type": "ACTION",
89
+ "shape": [
90
+ 32
91
+ ]
92
+ },
93
+ "agent_pos": {
94
+ "type": "STATE",
95
+ "shape": [
96
+ 32
97
+ ]
98
+ },
99
+ "pixels/top": {
100
+ "type": "VISUAL",
101
+ "shape": [
102
+ 480,
103
+ 640,
104
+ 3
105
+ ]
106
+ }
107
+ },
108
+ "features_map": {
109
+ "action": "action",
110
+ "agent_pos": "observation.state",
111
+ "top": "observation.image",
112
+ "pixels/top": "observation.image"
113
+ },
114
+ "max_parallel_tasks": 1,
115
+ "disable_env_checker": true,
116
+ "config": {
117
+ "eval_ds": "/netscratch/mboukheddimi/RH5_dataset_threads/",
118
+ "episode_idx": 806,
119
+ "absolute": false
120
+ },
121
+ "episode_length": 800,
122
+ "obs_type": "pixels_agent_pos",
123
+ "render_mode": "rgb_array"
124
+ },
125
+ "policy": {
126
+ "type": "pi05",
127
+ "n_obs_steps": 1,
128
+ "input_features": {
129
+ "observation.image": {
130
+ "type": "VISUAL",
131
+ "shape": [
132
+ 3,
133
+ 256,
134
+ 256
135
+ ]
136
+ },
137
+ "observation.state": {
138
+ "type": "STATE",
139
+ "shape": [
140
+ 32
141
+ ]
142
+ }
143
+ },
144
+ "output_features": {
145
+ "action": {
146
+ "type": "ACTION",
147
+ "shape": [
148
+ 32
149
+ ]
150
+ }
151
+ },
152
+ "device": "cuda",
153
+ "use_amp": false,
154
+ "use_peft": false,
155
+ "push_to_hub": false,
156
+ "repo_id": "usr256864/pi0",
157
+ "private": null,
158
+ "tags": null,
159
+ "license": null,
160
+ "pretrained_path": null,
161
+ "paligemma_variant": "gemma_2b",
162
+ "action_expert_variant": "gemma_300m",
163
+ "dtype": "bfloat16",
164
+ "chunk_size": 50,
165
+ "n_action_steps": 50,
166
+ "max_state_dim": 32,
167
+ "max_action_dim": 32,
168
+ "num_inference_steps": 10,
169
+ "time_sampling_beta_alpha": 1.5,
170
+ "time_sampling_beta_beta": 1.0,
171
+ "time_sampling_scale": 0.999,
172
+ "time_sampling_offset": 0.001,
173
+ "min_period": 0.004,
174
+ "max_period": 4.0,
175
+ "rtc_config": null,
176
+ "image_resolution": [
177
+ 224,
178
+ 224
179
+ ],
180
+ "empty_cameras": 0,
181
+ "tokenizer_max_length": 200,
182
+ "normalization_mapping": {
183
+ "VISUAL": "IDENTITY",
184
+ "STATE": "QUANTILES",
185
+ "ACTION": "QUANTILES"
186
+ },
187
+ "gradient_checkpointing": true,
188
+ "compile_model": false,
189
+ "compile_mode": "max-autotune",
190
+ "freeze_vision_encoder": false,
191
+ "train_expert_only": false,
192
+ "optimizer_lr": 2.5e-05,
193
+ "optimizer_betas": [
194
+ 0.9,
195
+ 0.95
196
+ ],
197
+ "optimizer_eps": 1e-08,
198
+ "optimizer_weight_decay": 0.01,
199
+ "optimizer_grad_clip_norm": 1.0,
200
+ "scheduler_warmup_steps": 1000,
201
+ "scheduler_decay_steps": 30000,
202
+ "scheduler_decay_lr": 2.5e-06
203
+ },
204
+ "output_dir": "outputs/train/2026-01-28/17-05-05_pi05_training_delta_quantille",
205
+ "job_name": "pi05_training_delta_quantille",
206
+ "resume": false,
207
+ "seed": 1000,
208
+ "num_workers": 4,
209
+ "batch_size": 32,
210
+ "steps": 100000,
211
+ "eval_freq": 5000,
212
+ "log_freq": 400,
213
+ "tolerance_s": 0.0001,
214
+ "save_checkpoint": true,
215
+ "save_freq": 10000,
216
+ "use_policy_training_preset": true,
217
+ "optimizer": {
218
+ "type": "adamw",
219
+ "lr": 2.5e-05,
220
+ "weight_decay": 0.01,
221
+ "grad_clip_norm": 1.0,
222
+ "betas": [
223
+ 0.9,
224
+ 0.95
225
+ ],
226
+ "eps": 1e-08
227
+ },
228
+ "scheduler": {
229
+ "type": "cosine_decay_with_warmup",
230
+ "num_warmup_steps": 1000,
231
+ "num_decay_steps": 30000,
232
+ "peak_lr": 2.5e-05,
233
+ "decay_lr": 2.5e-06
234
+ },
235
+ "eval": {
236
+ "n_episodes": 1,
237
+ "batch_size": 1,
238
+ "use_async_envs": false
239
+ },
240
+ "wandb": {
241
+ "enable": false,
242
+ "disable_artifact": true,
243
+ "project": "lerobot",
244
+ "entity": null,
245
+ "notes": null,
246
+ "run_id": null,
247
+ "mode": null
248
+ },
249
+ "peft": null,
250
+ "use_rabc": false,
251
+ "rabc_progress_path": null,
252
+ "rabc_kappa": 0.01,
253
+ "rabc_epsilon": 1e-06,
254
+ "rabc_head_mode": "sparse",
255
+ "rename_map": {},
256
+ "checkpoint_path": null,
257
+ "eval_ds": "/netscratch/mboukheddimi/RH5_dataset_threads/",
258
+ "episode_idx": 806,
259
+ "absolute": false
260
+ }