Hugging Face's logo

Natalish
/
so101-roller-brush-model

Safetensors
Model card Files
xet
 Community
Natalish commited on
Commit
01266e9
·
verified ·
1 Parent(s): 489a480

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +87 -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 +79 -0
  6. policy_preprocessor_step_2_normalizer_processor.safetensors +3 -0
  7. train_config.json +223 -0
config.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "pi05",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 6
9
+ ]
10
+ },
11
+ "observation.images.wrist": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 480,
16
+ 640
17
+ ]
18
+ },
19
+ "observation.images.front": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 480,
24
+ 640
25
+ ]
26
+ }
27
+ },
28
+ "output_features": {
29
+ "action": {
30
+ "type": "ACTION",
31
+ "shape": [
32
+ 6
33
+ ]
34
+ }
35
+ },
36
+ "device": "cuda",
37
+ "use_amp": false,
38
+ "use_peft": false,
39
+ "push_to_hub": false,
40
+ "repo_id": null,
41
+ "private": null,
42
+ "tags": null,
43
+ "license": null,
44
+ "pretrained_path": "lerobot/pi05_base",
45
+ "paligemma_variant": "gemma_2b",
46
+ "action_expert_variant": "gemma_300m",
47
+ "dtype": "bfloat16",
48
+ "chunk_size": 50,
49
+ "n_action_steps": 50,
50
+ "max_state_dim": 32,
51
+ "max_action_dim": 32,
52
+ "num_inference_steps": 10,
53
+ "time_sampling_beta_alpha": 1.5,
54
+ "time_sampling_beta_beta": 1.0,
55
+ "time_sampling_scale": 0.999,
56
+ "time_sampling_offset": 0.001,
57
+ "min_period": 0.004,
58
+ "max_period": 4.0,
59
+ "rtc_config": null,
60
+ "image_resolution": [
61
+ 224,
62
+ 224
63
+ ],
64
+ "empty_cameras": 0,
65
+ "tokenizer_max_length": 200,
66
+ "normalization_mapping": {
67
+ "VISUAL": "IDENTITY",
68
+ "STATE": "QUANTILES",
69
+ "ACTION": "QUANTILES"
70
+ },
71
+ "gradient_checkpointing": true,
72
+ "compile_model": true,
73
+ "compile_mode": "max-autotune",
74
+ "freeze_vision_encoder": false,
75
+ "train_expert_only": false,
76
+ "optimizer_lr": 2.5e-05,
77
+ "optimizer_betas": [
78
+ 0.9,
79
+ 0.95
80
+ ],
81
+ "optimizer_eps": 1e-08,
82
+ "optimizer_weight_decay": 0.01,
83
+ "optimizer_grad_clip_norm": 1.0,
84
+ "scheduler_warmup_steps": 1000,
85
+ "scheduler_decay_steps": 30000,
86
+ "scheduler_decay_lr": 2.5e-06
87
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83448f66b60afbdbc584da6fd762bad27d005e584b27a81ea6e4d5f91bc723a7
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
+ 6
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 sha256:05cce7061e29358585dde9ee7ade2f12c953c9ca4eaeb46ff391fa866afba90b
3
+ size 7552
policy_preprocessor.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ 6
23
+ ]
24
+ },
25
+ "observation.images.wrist": {
26
+ "type": "VISUAL",
27
+ "shape": [
28
+ 3,
29
+ 480,
30
+ 640
31
+ ]
32
+ },
33
+ "observation.images.front": {
34
+ "type": "VISUAL",
35
+ "shape": [
36
+ 3,
37
+ 480,
38
+ 640
39
+ ]
40
+ },
41
+ "action": {
42
+ "type": "ACTION",
43
+ "shape": [
44
+ 6
45
+ ]
46
+ }
47
+ },
48
+ "norm_map": {
49
+ "VISUAL": "IDENTITY",
50
+ "STATE": "QUANTILES",
51
+ "ACTION": "QUANTILES"
52
+ }
53
+ },
54
+ "state_file": "policy_preprocessor_step_2_normalizer_processor.safetensors"
55
+ },
56
+ {
57
+ "registry_name": "pi05_prepare_state_tokenizer_processor_step",
58
+ "config": {}
59
+ },
60
+ {
61
+ "registry_name": "tokenizer_processor",
62
+ "config": {
63
+ "max_length": 200,
64
+ "task_key": "task",
65
+ "padding_side": "right",
66
+ "padding": "max_length",
67
+ "truncation": true,
68
+ "tokenizer_name": "google/paligemma-3b-pt-224"
69
+ }
70
+ },
71
+ {
72
+ "registry_name": "device_processor",
73
+ "config": {
74
+ "device": "cuda",
75
+ "float_dtype": null
76
+ }
77
+ }
78
+ ]
79
+ }
policy_preprocessor_step_2_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05cce7061e29358585dde9ee7ade2f12c953c9ca4eaeb46ff391fa866afba90b
3
+ size 7552
train_config.json ADDED
@@ -0,0 +1,223 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "Natalish/so101-roller-brush-demos",
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": "pi05",
85
+ "n_obs_steps": 1,
86
+ "input_features": {
87
+ "observation.state": {
88
+ "type": "STATE",
89
+ "shape": [
90
+ 6
91
+ ]
92
+ },
93
+ "observation.images.wrist": {
94
+ "type": "VISUAL",
95
+ "shape": [
96
+ 3,
97
+ 480,
98
+ 640
99
+ ]
100
+ },
101
+ "observation.images.front": {
102
+ "type": "VISUAL",
103
+ "shape": [
104
+ 3,
105
+ 480,
106
+ 640
107
+ ]
108
+ }
109
+ },
110
+ "output_features": {
111
+ "action": {
112
+ "type": "ACTION",
113
+ "shape": [
114
+ 6
115
+ ]
116
+ }
117
+ },
118
+ "device": "cuda",
119
+ "use_amp": false,
120
+ "use_peft": false,
121
+ "push_to_hub": false,
122
+ "repo_id": null,
123
+ "private": null,
124
+ "tags": null,
125
+ "license": null,
126
+ "pretrained_path": "lerobot/pi05_base",
127
+ "paligemma_variant": "gemma_2b",
128
+ "action_expert_variant": "gemma_300m",
129
+ "dtype": "bfloat16",
130
+ "chunk_size": 50,
131
+ "n_action_steps": 50,
132
+ "max_state_dim": 32,
133
+ "max_action_dim": 32,
134
+ "num_inference_steps": 10,
135
+ "time_sampling_beta_alpha": 1.5,
136
+ "time_sampling_beta_beta": 1.0,
137
+ "time_sampling_scale": 0.999,
138
+ "time_sampling_offset": 0.001,
139
+ "min_period": 0.004,
140
+ "max_period": 4.0,
141
+ "rtc_config": null,
142
+ "image_resolution": [
143
+ 224,
144
+ 224
145
+ ],
146
+ "empty_cameras": 0,
147
+ "tokenizer_max_length": 200,
148
+ "normalization_mapping": {
149
+ "VISUAL": "IDENTITY",
150
+ "STATE": "QUANTILES",
151
+ "ACTION": "QUANTILES"
152
+ },
153
+ "gradient_checkpointing": true,
154
+ "compile_model": true,
155
+ "compile_mode": "max-autotune",
156
+ "freeze_vision_encoder": false,
157
+ "train_expert_only": false,
158
+ "optimizer_lr": 2.5e-05,
159
+ "optimizer_betas": [
160
+ 0.9,
161
+ 0.95
162
+ ],
163
+ "optimizer_eps": 1e-08,
164
+ "optimizer_weight_decay": 0.01,
165
+ "optimizer_grad_clip_norm": 1.0,
166
+ "scheduler_warmup_steps": 1000,
167
+ "scheduler_decay_steps": 30000,
168
+ "scheduler_decay_lr": 2.5e-06
169
+ },
170
+ "output_dir": "outputs/pi05_roller_brush",
171
+ "job_name": "pi05_roller_brush",
172
+ "resume": false,
173
+ "seed": 1000,
174
+ "num_workers": 4,
175
+ "batch_size": 8,
176
+ "steps": 40000,
177
+ "eval_freq": 20000,
178
+ "log_freq": 200,
179
+ "tolerance_s": 0.0001,
180
+ "save_checkpoint": true,
181
+ "save_freq": 40000,
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": 50,
204
+ "use_async_envs": false
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
+ },
215
+ "peft": null,
216
+ "use_rabc": false,
217
+ "rabc_progress_path": null,
218
+ "rabc_kappa": 0.01,
219
+ "rabc_epsilon": 1e-06,
220
+ "rabc_head_mode": "sparse",
221
+ "rename_map": {},
222
+ "checkpoint_path": null
223
+ }