Hugging Face's logo

luojianaiw
/
multiframe

Safetensors
Model card Files
xet
 Community
luojianaiw commited on
Commit
21b113c
·
verified ·
1 Parent(s): e61d338

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +119 -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 +95 -0
  6. policy_preprocessor_step_3_normalizer_processor.safetensors +3 -0
  7. train_config.json +262 -0
config.json ADDED
@@ -0,0 +1,119 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "pi05",
3
+ "n_obs_steps": 5,
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
+ 32
41
+ ]
42
+ }
43
+ },
44
+ "device": "cuda",
45
+ "use_amp": false,
46
+ "push_to_hub": false,
47
+ "repo_id": null,
48
+ "private": null,
49
+ "tags": null,
50
+ "license": null,
51
+ "pretrained_path": "/data/output-train/output-lrbtwin_pi05_multiframe_delta/pi05_multiframe5_delta_mixed12345_8gpu80g/checkpoints/003000/pretrained_model",
52
+ "paligemma_variant": "gemma_2b",
53
+ "action_expert_variant": "gemma_300m",
54
+ "dtype": "bfloat16",
55
+ "use_peft": false,
56
+ "rtc_config": null,
57
+ "freeze_vision_encoder": false,
58
+ "train_expert_only": false,
59
+ "chunk_size": 50,
60
+ "n_action_steps": 50,
61
+ "max_state_dim": 16,
62
+ "max_action_dim": 32,
63
+ "num_inference_steps": 10,
64
+ "time_sampling_beta_alpha": 1.5,
65
+ "time_sampling_beta_beta": 1.0,
66
+ "time_sampling_scale": 0.999,
67
+ "time_sampling_offset": 0.001,
68
+ "min_period": 0.004,
69
+ "max_period": 4.0,
70
+ "image_resolution": [
71
+ 224,
72
+ 224
73
+ ],
74
+ "empty_cameras": 0,
75
+ "tokenizer_max_length": 200,
76
+ "normalization_mapping": {
77
+ "VISUAL": "IDENTITY",
78
+ "STATE": "QUANTILES",
79
+ "ACTION": "QUANTILES"
80
+ },
81
+ "gradient_checkpointing": true,
82
+ "compile_model": false,
83
+ "compile_mode": "max-autotune",
84
+ "keep_vision_float32": false,
85
+ "history_max_image_frames": 0,
86
+ "history_max_action_steps": 0,
87
+ "history_action_memory_time": 0.0,
88
+ "train_with_history_kv": false,
89
+ "history_training_rollout_steps": 0,
90
+ "history_reference_window_training": false,
91
+ "history_anchor_stride": 0,
92
+ "history_supervise_last_block_only": false,
93
+ "history_training_chunk_supervision_steps": 0,
94
+ "history_training_chunk_supervision_decay": 1.0,
95
+ "gripper_loss_weight": 1.0,
96
+ "gripper_action_dim": -1,
97
+ "use_delta_joint_actions": true,
98
+ "delta_action_mask_dims": [
99
+ 7,
100
+ -1,
101
+ 7,
102
+ -1
103
+ ],
104
+ "conditioning_mode": "global_task_plus_state_block",
105
+ "include_state_in_task_prompt": false,
106
+ "append_source_newline_token": true,
107
+ "use_state_input": true,
108
+ "optimizer_lr": 1e-05,
109
+ "optimizer_betas": [
110
+ 0.9,
111
+ 0.95
112
+ ],
113
+ "optimizer_eps": 1e-08,
114
+ "optimizer_weight_decay": 0.01,
115
+ "optimizer_grad_clip_norm": 1.0,
116
+ "scheduler_warmup_steps": 1000,
117
+ "scheduler_decay_steps": 100000,
118
+ "scheduler_decay_lr": 1e-06
119
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5ee153e400adddab4942b801d667abde222210357b8755adcc7ee9ae693e12b
3
+ size 7237986856
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 sha256:3f0d00c9164e3deadb3c8025dcece9fd7d48ddf56720de534f55b0e7cb8e77b5
3
+ size 9576
policy_preprocessor.json ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "policy_preprocessor",
3
+ "steps": [
4
+ {
5
+ "registry_name": "rename_observations_processor",
6
+ "config": {
7
+ "rename_map": {
8
+ "observation.images.cam_high": "observation.images.base_0_rgb",
9
+ "observation.images.cam_left_wrist": "observation.images.left_wrist_0_rgb",
10
+ "observation.images.cam_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": "pi05_delta_joint_actions_processor_step",
20
+ "config": {}
21
+ },
22
+ {
23
+ "registry_name": "normalizer_processor",
24
+ "config": {
25
+ "eps": 1e-08,
26
+ "features": {
27
+ "observation.images.base_0_rgb": {
28
+ "type": "VISUAL",
29
+ "shape": [
30
+ 3,
31
+ 224,
32
+ 224
33
+ ]
34
+ },
35
+ "observation.images.left_wrist_0_rgb": {
36
+ "type": "VISUAL",
37
+ "shape": [
38
+ 3,
39
+ 224,
40
+ 224
41
+ ]
42
+ },
43
+ "observation.images.right_wrist_0_rgb": {
44
+ "type": "VISUAL",
45
+ "shape": [
46
+ 3,
47
+ 224,
48
+ 224
49
+ ]
50
+ },
51
+ "observation.state": {
52
+ "type": "STATE",
53
+ "shape": [
54
+ 32
55
+ ]
56
+ },
57
+ "action": {
58
+ "type": "ACTION",
59
+ "shape": [
60
+ 32
61
+ ]
62
+ }
63
+ },
64
+ "norm_map": {
65
+ "VISUAL": "IDENTITY",
66
+ "STATE": "QUANTILES",
67
+ "ACTION": "QUANTILES"
68
+ }
69
+ },
70
+ "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors"
71
+ },
72
+ {
73
+ "registry_name": "pi05_prepare_state_tokenizer_processor_step",
74
+ "config": {}
75
+ },
76
+ {
77
+ "registry_name": "tokenizer_processor",
78
+ "config": {
79
+ "max_length": 200,
80
+ "task_key": "task",
81
+ "padding_side": "right",
82
+ "padding": "max_length",
83
+ "truncation": true,
84
+ "tokenizer_name": "google/paligemma-3b-pt-224"
85
+ }
86
+ },
87
+ {
88
+ "registry_name": "device_processor",
89
+ "config": {
90
+ "device": "cuda",
91
+ "float_dtype": null
92
+ }
93
+ }
94
+ ]
95
+ }
policy_preprocessor_step_3_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f0d00c9164e3deadb3c8025dcece9fd7d48ddf56720de534f55b0e7cb8e77b5
3
+ size 9576
train_config.json ADDED
@@ -0,0 +1,262 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "robotwin_franka_clean_50_parquet",
4
+ "root": "/data/datasets/robotwin_franka_clean_50_parquet",
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": 5,
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
+ 32
123
+ ]
124
+ }
125
+ },
126
+ "device": "cuda",
127
+ "use_amp": false,
128
+ "push_to_hub": false,
129
+ "repo_id": null,
130
+ "private": null,
131
+ "tags": null,
132
+ "license": null,
133
+ "pretrained_path": "/data/output-train/output-lrbtwin_pi05_multiframe_delta/pi05_multiframe5_delta_mixed12345_8gpu80g/checkpoints/003000/pretrained_model",
134
+ "paligemma_variant": "gemma_2b",
135
+ "action_expert_variant": "gemma_300m",
136
+ "dtype": "bfloat16",
137
+ "use_peft": false,
138
+ "rtc_config": null,
139
+ "freeze_vision_encoder": false,
140
+ "train_expert_only": false,
141
+ "chunk_size": 50,
142
+ "n_action_steps": 50,
143
+ "max_state_dim": 16,
144
+ "max_action_dim": 32,
145
+ "num_inference_steps": 10,
146
+ "time_sampling_beta_alpha": 1.5,
147
+ "time_sampling_beta_beta": 1.0,
148
+ "time_sampling_scale": 0.999,
149
+ "time_sampling_offset": 0.001,
150
+ "min_period": 0.004,
151
+ "max_period": 4.0,
152
+ "image_resolution": [
153
+ 224,
154
+ 224
155
+ ],
156
+ "empty_cameras": 0,
157
+ "tokenizer_max_length": 200,
158
+ "normalization_mapping": {
159
+ "VISUAL": "IDENTITY",
160
+ "STATE": "QUANTILES",
161
+ "ACTION": "QUANTILES"
162
+ },
163
+ "gradient_checkpointing": true,
164
+ "compile_model": false,
165
+ "compile_mode": "max-autotune",
166
+ "keep_vision_float32": false,
167
+ "history_max_image_frames": 0,
168
+ "history_max_action_steps": 0,
169
+ "history_action_memory_time": 0.0,
170
+ "train_with_history_kv": false,
171
+ "history_training_rollout_steps": 0,
172
+ "history_reference_window_training": false,
173
+ "history_anchor_stride": 0,
174
+ "history_supervise_last_block_only": false,
175
+ "history_training_chunk_supervision_steps": 0,
176
+ "history_training_chunk_supervision_decay": 1.0,
177
+ "gripper_loss_weight": 1.0,
178
+ "gripper_action_dim": -1,
179
+ "use_delta_joint_actions": true,
180
+ "delta_action_mask_dims": [
181
+ 7,
182
+ -1,
183
+ 7,
184
+ -1
185
+ ],
186
+ "conditioning_mode": "global_task_plus_state_block",
187
+ "include_state_in_task_prompt": false,
188
+ "append_source_newline_token": true,
189
+ "use_state_input": true,
190
+ "optimizer_lr": 1e-05,
191
+ "optimizer_betas": [
192
+ 0.9,
193
+ 0.95
194
+ ],
195
+ "optimizer_eps": 1e-08,
196
+ "optimizer_weight_decay": 0.01,
197
+ "optimizer_grad_clip_norm": 1.0,
198
+ "scheduler_warmup_steps": 1000,
199
+ "scheduler_decay_steps": 100000,
200
+ "scheduler_decay_lr": 1e-06
201
+ },
202
+ "output_dir": "/data/output-train/output-lrbtwin_pi05_multiframe_delta/pi05_multiframe5_delta_mixed12345_8gpu80g",
203
+ "job_name": "pi05_multiframe5_delta_mixed12345_8gpu80g",
204
+ "resume": true,
205
+ "seed": 1000,
206
+ "num_workers": 4,
207
+ "batch_size": 8,
208
+ "gradient_accumulation_steps": 4,
209
+ "mixed_frame_counts": [
210
+ 1,
211
+ 2,
212
+ 3,
213
+ 4,
214
+ 5
215
+ ],
216
+ "steps": 100000,
217
+ "eval_freq": 0,
218
+ "log_freq": 10,
219
+ "save_checkpoint": true,
220
+ "save_freq": 200,
221
+ "save_best_only": false,
222
+ "latest_save_freq": null,
223
+ "use_policy_training_preset": true,
224
+ "optimizer": {
225
+ "type": "adamw",
226
+ "lr": 1e-05,
227
+ "weight_decay": 0.01,
228
+ "grad_clip_norm": 1.0,
229
+ "betas": [
230
+ 0.9,
231
+ 0.95
232
+ ],
233
+ "eps": 1e-08
234
+ },
235
+ "scheduler": {
236
+ "type": "cosine_decay_with_warmup",
237
+ "num_warmup_steps": 1000,
238
+ "num_decay_steps": 100000,
239
+ "peak_lr": 1e-05,
240
+ "decay_lr": 1e-06
241
+ },
242
+ "eval": {
243
+ "n_episodes": 50,
244
+ "batch_size": 50,
245
+ "use_async_envs": false
246
+ },
247
+ "wandb": {
248
+ "enable": true,
249
+ "disable_artifact": true,
250
+ "project": "robotwin-pi05-multiframe",
251
+ "entity": null,
252
+ "notes": null,
253
+ "run_id": "268orato",
254
+ "mode": "offline"
255
+ },
256
+ "checkpoint_path": "/data/output-train/output-lrbtwin_pi05_multiframe_delta/pi05_multiframe5_delta_mixed12345_8gpu80g/checkpoints/003000",
257
+ "rename_map": {
258
+ "observation.images.cam_high": "observation.images.base_0_rgb",
259
+ "observation.images.cam_left_wrist": "observation.images.left_wrist_0_rgb",
260
+ "observation.images.cam_right_wrist": "observation.images.right_wrist_0_rgb"
261
+ }
262
+ }