Hugging Face's logo

Sakits
/
pi05_async8_sorting_20260218

Safetensors
Model card Files
xet
 Community
Sakits commited on
Commit
8e5efcf
·
verified ·
1 Parent(s): b81428d

Upload folder using huggingface_hub

Browse files
Files changed (3) hide show
  1. config.json +81 -0
  2. model.safetensors +3 -0
  3. train_config.json +233 -0
config.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ },
20
+ "output_features": {
21
+ "action": {
22
+ "type": "ACTION",
23
+ "shape": [
24
+ 6
25
+ ]
26
+ }
27
+ },
28
+ "device": "cuda",
29
+ "use_amp": false,
30
+ "push_to_hub": false,
31
+ "repo_id": null,
32
+ "private": null,
33
+ "tags": null,
34
+ "license": null,
35
+ "pretrained_path": "lerobot/pi05_base",
36
+ "paligemma_variant": "gemma_2b",
37
+ "action_expert_variant": "gemma_300m",
38
+ "dtype": "bfloat16",
39
+ "chunk_size": 50,
40
+ "n_action_steps": 50,
41
+ "max_state_dim": 32,
42
+ "max_action_dim": 32,
43
+ "state_cond": true,
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
+ "image_resolution": [
52
+ 224,
53
+ 224
54
+ ],
55
+ "empty_cameras": 0,
56
+ "tokenizer_max_length": 200,
57
+ "normalization_mapping": {
58
+ "VISUAL": "IDENTITY",
59
+ "STATE": "MEAN_STD",
60
+ "ACTION": "MEAN_STD"
61
+ },
62
+ "gradient_checkpointing": false,
63
+ "compile_model": false,
64
+ "compile_mode": "max-autotune",
65
+ "fuse_qkv": false,
66
+ "fuse_gate_up": false,
67
+ "fuse_adarms": 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
+ "vlm_config": {},
80
+ "action_expert_config": {}
81
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9775b468cceb1808b932a255efcc893b28f20ef9ead16a43272f18a6f0d64710
3
+ size 7481485688
train_config.json ADDED
@@ -0,0 +1,233 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "Sakits/so101_sorting_20260218",
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.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
+ },
102
+ "output_features": {
103
+ "action": {
104
+ "type": "ACTION",
105
+ "shape": [
106
+ 6
107
+ ]
108
+ }
109
+ },
110
+ "device": "cuda",
111
+ "use_amp": false,
112
+ "push_to_hub": false,
113
+ "repo_id": null,
114
+ "private": null,
115
+ "tags": null,
116
+ "license": null,
117
+ "pretrained_path": "lerobot/pi05_base",
118
+ "paligemma_variant": "gemma_2b",
119
+ "action_expert_variant": "gemma_300m",
120
+ "dtype": "bfloat16",
121
+ "chunk_size": 50,
122
+ "n_action_steps": 50,
123
+ "max_state_dim": 32,
124
+ "max_action_dim": 32,
125
+ "state_cond": true,
126
+ "num_inference_steps": 10,
127
+ "time_sampling_beta_alpha": 1.5,
128
+ "time_sampling_beta_beta": 1.0,
129
+ "time_sampling_scale": 0.999,
130
+ "time_sampling_offset": 0.001,
131
+ "min_period": 0.004,
132
+ "max_period": 4.0,
133
+ "image_resolution": [
134
+ 224,
135
+ 224
136
+ ],
137
+ "empty_cameras": 0,
138
+ "tokenizer_max_length": 200,
139
+ "normalization_mapping": {
140
+ "VISUAL": "IDENTITY",
141
+ "STATE": "MEAN_STD",
142
+ "ACTION": "MEAN_STD"
143
+ },
144
+ "gradient_checkpointing": false,
145
+ "compile_model": false,
146
+ "compile_mode": "max-autotune",
147
+ "fuse_qkv": false,
148
+ "fuse_gate_up": false,
149
+ "fuse_adarms": false,
150
+ "optimizer_lr": 2.5e-05,
151
+ "optimizer_betas": [
152
+ 0.9,
153
+ 0.95
154
+ ],
155
+ "optimizer_eps": 1e-08,
156
+ "optimizer_weight_decay": 0.01,
157
+ "optimizer_grad_clip_norm": 1.0,
158
+ "scheduler_warmup_steps": 1000,
159
+ "scheduler_decay_steps": 30000,
160
+ "scheduler_decay_lr": 2.5e-06,
161
+ "vlm_config": {},
162
+ "action_expert_config": {}
163
+ },
164
+ "output_dir": "outputs/train/pi05_async8_stacking_20260218",
165
+ "job_name": "pi05_async8_stacking_20260218",
166
+ "resume": false,
167
+ "seed": 1000,
168
+ "num_workers": 4,
169
+ "batch_size": 8,
170
+ "steps": 50000,
171
+ "eval_freq": 20000,
172
+ "log_freq": 200,
173
+ "save_checkpoint": true,
174
+ "save_freq": 10000,
175
+ "use_policy_training_preset": false,
176
+ "optimizer": {
177
+ "type": "adamw",
178
+ "lr": 5e-05,
179
+ "weight_decay": 1e-10,
180
+ "grad_clip_norm": 10.0,
181
+ "betas": [
182
+ 0.9,
183
+ 0.95
184
+ ],
185
+ "eps": 1e-08
186
+ },
187
+ "scheduler": {
188
+ "type": "cosine_decay_with_warmup",
189
+ "num_warmup_steps": 1000,
190
+ "num_decay_steps": 50000,
191
+ "peak_lr": 5e-05,
192
+ "decay_lr": 2.5e-06
193
+ },
194
+ "eval": {
195
+ "n_episodes": 50,
196
+ "batch_size": 50,
197
+ "use_async_envs": false
198
+ },
199
+ "wandb": {
200
+ "enable": true,
201
+ "disable_artifact": true,
202
+ "project": "vlash",
203
+ "entity": null,
204
+ "notes": null,
205
+ "run_id": "htrb1cka",
206
+ "mode": null
207
+ },
208
+ "checkpoint_path": null,
209
+ "rename_map": {},
210
+ "max_delay_steps": 8,
211
+ "grad_accum_steps": 1,
212
+ "shared_observation": true,
213
+ "lora": {
214
+ "enable": false,
215
+ "backend": "peft",
216
+ "r": 16,
217
+ "alpha": 16,
218
+ "dropout": 0.0,
219
+ "extra_trainable_modules": [],
220
+ "target_modules": [
221
+ "q_proj",
222
+ "k_proj",
223
+ "v_proj",
224
+ "o_proj",
225
+ "up_proj",
226
+ "down_proj",
227
+ "gate_proj"
228
+ ],
229
+ "use_qlora": false,
230
+ "qlora_quant_type": "nf4",
231
+ "qlora_compute_dtype": "bfloat16"
232
+ }
233
+ }