Hugging Face's logo

jadechoghari
/
pi-folding-20k

Safetensors
Model card Files
xet
 Community
jadechoghari commited on
Commit
1126fe6
·
verified ·
1 Parent(s): 59fb952

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +93 -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 +99 -0
  6. policy_preprocessor_step_2_normalizer_processor.safetensors +3 -0
  7. train_config.json +232 -0
config.json ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "pi0_fast",
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
+ 16
41
+ ]
42
+ }
43
+ },
44
+ "device": "cuda",
45
+ "use_amp": false,
46
+ "push_to_hub": true,
47
+ "repo_id": "jadechoghari/pi0fast-folding",
48
+ "private": null,
49
+ "tags": null,
50
+ "license": null,
51
+ "pretrained_path": "/fsx/jade_choghari/models/pi0fast-base",
52
+ "paligemma_variant": "gemma_2b",
53
+ "action_expert_variant": "gemma_300m",
54
+ "dtype": "bfloat16",
55
+ "chunk_size": 10,
56
+ "n_action_steps": 10,
57
+ "max_state_dim": 32,
58
+ "max_action_dim": 32,
59
+ "max_action_tokens": 256,
60
+ "rtc_config": null,
61
+ "image_resolution": [
62
+ 224,
63
+ 224
64
+ ],
65
+ "empty_cameras": 0,
66
+ "tokenizer_max_length": 200,
67
+ "text_tokenizer_name": "google/paligemma-3b-pt-224",
68
+ "action_tokenizer_name": "jadechoghari/fast-folding-tokenizer",
69
+ "temperature": 0.0,
70
+ "max_decoding_steps": 256,
71
+ "fast_skip_tokens": 128,
72
+ "validate_action_token_prefix": true,
73
+ "use_kv_cache": true,
74
+ "normalization_mapping": {
75
+ "VISUAL": "IDENTITY",
76
+ "STATE": "MEAN_STD",
77
+ "ACTION": "MEAN_STD"
78
+ },
79
+ "gradient_checkpointing": true,
80
+ "compile_model": false,
81
+ "compile_mode": "max-autotune",
82
+ "optimizer_lr": 2.5e-05,
83
+ "optimizer_betas": [
84
+ 0.9,
85
+ 0.95
86
+ ],
87
+ "optimizer_eps": 1e-08,
88
+ "optimizer_weight_decay": 0.01,
89
+ "optimizer_grad_clip_norm": 1.0,
90
+ "scheduler_warmup_steps": 4000,
91
+ "scheduler_decay_steps": 100000,
92
+ "scheduler_decay_lr": 1e-05
93
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:36fedf6cb9424ce792429161bda8792f5cedd2c0a5b1333e917c8a5bbb0b7b83
3
+ size 5848876592
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
+ 16
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": "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 sha256:000d96b15bc3d22c1cfe7ca25419b2981cf281147cbc00bcedfc64e292253700
3
+ size 9456
policy_preprocessor.json ADDED
@@ -0,0 +1,99 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "policy_preprocessor",
3
+ "steps": [
4
+ {
5
+ "registry_name": "rename_observations_processor",
6
+ "config": {
7
+ "rename_map": {
8
+ "observation.images.base": "observation.images.base_0_rgb",
9
+ "observation.images.left_wrist": "observation.images.left_wrist_0_rgb",
10
+ "observation.images.right_wrist": "observation.images.right_wrist_0_rgb"
11
+ }
12
+ }
13
+ },
14
+ {
15
+ "registry_name": "to_batch_processor",
16
+ "config": {}
17
+ },
18
+ {
19
+ "registry_name": "normalizer_processor",
20
+ "config": {
21
+ "eps": 1e-08,
22
+ "features": {
23
+ "observation.images.base_0_rgb": {
24
+ "type": "VISUAL",
25
+ "shape": [
26
+ 3,
27
+ 224,
28
+ 224
29
+ ]
30
+ },
31
+ "observation.images.left_wrist_0_rgb": {
32
+ "type": "VISUAL",
33
+ "shape": [
34
+ 3,
35
+ 224,
36
+ 224
37
+ ]
38
+ },
39
+ "observation.images.right_wrist_0_rgb": {
40
+ "type": "VISUAL",
41
+ "shape": [
42
+ 3,
43
+ 224,
44
+ 224
45
+ ]
46
+ },
47
+ "observation.state": {
48
+ "type": "STATE",
49
+ "shape": [
50
+ 32
51
+ ]
52
+ },
53
+ "action": {
54
+ "type": "ACTION",
55
+ "shape": [
56
+ 16
57
+ ]
58
+ }
59
+ },
60
+ "norm_map": {
61
+ "VISUAL": "IDENTITY",
62
+ "STATE": "MEAN_STD",
63
+ "ACTION": "MEAN_STD"
64
+ }
65
+ },
66
+ "state_file": "policy_preprocessor_step_2_normalizer_processor.safetensors"
67
+ },
68
+ {
69
+ "registry_name": "pi05_prepare_state_tokenizer_processor_step",
70
+ "config": {}
71
+ },
72
+ {
73
+ "registry_name": "tokenizer_processor",
74
+ "config": {
75
+ "max_length": 200,
76
+ "task_key": "task",
77
+ "padding_side": "right",
78
+ "padding": "max_length",
79
+ "truncation": true,
80
+ "tokenizer_name": "google/paligemma-3b-pt-224"
81
+ }
82
+ },
83
+ {
84
+ "registry_name": "action_tokenizer_processor",
85
+ "config": {
86
+ "trust_remote_code": true,
87
+ "max_action_tokens": 256,
88
+ "action_tokenizer_name": "jadechoghari/fast-folding-tokenizer"
89
+ }
90
+ },
91
+ {
92
+ "registry_name": "device_processor",
93
+ "config": {
94
+ "device": "cuda",
95
+ "float_dtype": null
96
+ }
97
+ }
98
+ ]
99
+ }
policy_preprocessor_step_2_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:000d96b15bc3d22c1cfe7ca25419b2981cf281147cbc00bcedfc64e292253700
3
+ size 9456
train_config.json ADDED
@@ -0,0 +1,232 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "local",
4
+ "root": "/fsx/jade_choghari/data/folding-data",
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": "pi0_fast",
85
+ "n_obs_steps": 1,
86
+ "input_features": {
87
+ "observation.images.base_0_rgb": {
88
+ "type": "VISUAL",
89
+ "shape": [
90
+ 3,
91
+ 224,
92
+ 224
93
+ ]
94
+ },
95
+ "observation.images.left_wrist_0_rgb": {
96
+ "type": "VISUAL",
97
+ "shape": [
98
+ 3,
99
+ 224,
100
+ 224
101
+ ]
102
+ },
103
+ "observation.images.right_wrist_0_rgb": {
104
+ "type": "VISUAL",
105
+ "shape": [
106
+ 3,
107
+ 224,
108
+ 224
109
+ ]
110
+ },
111
+ "observation.state": {
112
+ "type": "STATE",
113
+ "shape": [
114
+ 32
115
+ ]
116
+ }
117
+ },
118
+ "output_features": {
119
+ "action": {
120
+ "type": "ACTION",
121
+ "shape": [
122
+ 16
123
+ ]
124
+ }
125
+ },
126
+ "device": "cuda",
127
+ "use_amp": false,
128
+ "push_to_hub": true,
129
+ "repo_id": "jadechoghari/pi0fast-folding",
130
+ "private": null,
131
+ "tags": null,
132
+ "license": null,
133
+ "pretrained_path": "/fsx/jade_choghari/models/pi0fast-base",
134
+ "paligemma_variant": "gemma_2b",
135
+ "action_expert_variant": "gemma_300m",
136
+ "dtype": "bfloat16",
137
+ "chunk_size": 10,
138
+ "n_action_steps": 10,
139
+ "max_state_dim": 32,
140
+ "max_action_dim": 32,
141
+ "max_action_tokens": 256,
142
+ "rtc_config": null,
143
+ "image_resolution": [
144
+ 224,
145
+ 224
146
+ ],
147
+ "empty_cameras": 0,
148
+ "tokenizer_max_length": 200,
149
+ "text_tokenizer_name": "google/paligemma-3b-pt-224",
150
+ "action_tokenizer_name": "jadechoghari/fast-folding-tokenizer",
151
+ "temperature": 0.0,
152
+ "max_decoding_steps": 256,
153
+ "fast_skip_tokens": 128,
154
+ "validate_action_token_prefix": true,
155
+ "use_kv_cache": true,
156
+ "normalization_mapping": {
157
+ "VISUAL": "IDENTITY",
158
+ "STATE": "MEAN_STD",
159
+ "ACTION": "MEAN_STD"
160
+ },
161
+ "gradient_checkpointing": true,
162
+ "compile_model": false,
163
+ "compile_mode": "max-autotune",
164
+ "optimizer_lr": 2.5e-05,
165
+ "optimizer_betas": [
166
+ 0.9,
167
+ 0.95
168
+ ],
169
+ "optimizer_eps": 1e-08,
170
+ "optimizer_weight_decay": 0.01,
171
+ "optimizer_grad_clip_norm": 1.0,
172
+ "scheduler_warmup_steps": 4000,
173
+ "scheduler_decay_steps": 100000,
174
+ "scheduler_decay_lr": 1e-05
175
+ },
176
+ "output_dir": "/fsx/jade_choghari/outputs/folding_training_fast_mean",
177
+ "job_name": "folding_training_fast",
178
+ "resume": false,
179
+ "seed": 1000,
180
+ "num_workers": 4,
181
+ "batch_size": 4,
182
+ "steps": 100000,
183
+ "eval_freq": 20000,
184
+ "log_freq": 200,
185
+ "tolerance_s": 0.0001,
186
+ "save_checkpoint": true,
187
+ "save_freq": 20000,
188
+ "use_policy_training_preset": true,
189
+ "optimizer": {
190
+ "type": "adamw",
191
+ "lr": 2.5e-05,
192
+ "weight_decay": 0.01,
193
+ "grad_clip_norm": 1.0,
194
+ "betas": [
195
+ 0.9,
196
+ 0.95
197
+ ],
198
+ "eps": 1e-08
199
+ },
200
+ "scheduler": {
201
+ "type": "cosine_decay_with_warmup",
202
+ "num_warmup_steps": 4000,
203
+ "num_decay_steps": 100000,
204
+ "peak_lr": 2.5e-05,
205
+ "decay_lr": 1e-05
206
+ },
207
+ "eval": {
208
+ "n_episodes": 50,
209
+ "batch_size": 50,
210
+ "use_async_envs": false
211
+ },
212
+ "wandb": {
213
+ "enable": true,
214
+ "disable_artifact": true,
215
+ "project": "pi0fast-folding-training",
216
+ "entity": null,
217
+ "notes": null,
218
+ "run_id": "b89fyg55",
219
+ "mode": null
220
+ },
221
+ "use_rabc": false,
222
+ "rabc_progress_path": null,
223
+ "rabc_kappa": 0.01,
224
+ "rabc_epsilon": 1e-06,
225
+ "rabc_head_mode": "sparse",
226
+ "rename_map": {
227
+ "observation.images.base": "observation.images.base_0_rgb",
228
+ "observation.images.left_wrist": "observation.images.left_wrist_0_rgb",
229
+ "observation.images.right_wrist": "observation.images.right_wrist_0_rgb"
230
+ },
231
+ "checkpoint_path": null
232
+ }