Hugging Face's logo

maximellerbach
/
folding_lingbot

Robotics
LeRobot
Safetensors
lingbot_va
Model card Files
xet
 Community
maximellerbach commited on
Commit
0dd11e7
·
verified ·
1 Parent(s): 74d6990

lingbotva 10k steps

Browse files
Files changed (7) hide show
  1. config.json +144 -0
  2. model.safetensors +3 -0
  3. policy_postprocessor.json +38 -0
  4. policy_postprocessor_step_0_unnormalizer_processor.safetensors +3 -0
  5. policy_preprocessor.json +97 -0
  6. policy_preprocessor_step_3_normalizer_processor.safetensors +3 -0
  7. train_config.json +308 -0
config.json ADDED
@@ -0,0 +1,144 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "lingbot_va",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.images.cam_high": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 256,
10
+ 256
11
+ ]
12
+ },
13
+ "observation.images.cam_left_wrist": {
14
+ "type": "VISUAL",
15
+ "shape": [
16
+ 3,
17
+ 256,
18
+ 256
19
+ ]
20
+ },
21
+ "observation.images.cam_right_wrist": {
22
+ "type": "VISUAL",
23
+ "shape": [
24
+ 3,
25
+ 256,
26
+ 256
27
+ ]
28
+ }
29
+ },
30
+ "output_features": {
31
+ "action": {
32
+ "type": "ACTION",
33
+ "shape": [
34
+ 16
35
+ ]
36
+ }
37
+ },
38
+ "device": "cuda",
39
+ "use_amp": false,
40
+ "use_peft": false,
41
+ "push_to_hub": false,
42
+ "repo_id": "maximellerbach/folding_lingbot",
43
+ "private": null,
44
+ "tags": null,
45
+ "license": null,
46
+ "pretrained_path": "lerobot/lingbot_va_base",
47
+ "pretrained_revision": null,
48
+ "patch_size": [
49
+ 1,
50
+ 2,
51
+ 2
52
+ ],
53
+ "num_attention_heads": 24,
54
+ "attention_head_dim": 128,
55
+ "in_channels": 48,
56
+ "out_channels": 48,
57
+ "action_dim": 30,
58
+ "text_dim": 4096,
59
+ "freq_dim": 256,
60
+ "ffn_dim": 14336,
61
+ "num_layers": 30,
62
+ "cross_attn_norm": true,
63
+ "eps": 1e-06,
64
+ "rope_max_seq_len": 1024,
65
+ "attn_mode": "flex",
66
+ "wan_pretrained_path": "robbyant/lingbot-va-base",
67
+ "dtype": "bfloat16",
68
+ "text_encoder_device": "cuda",
69
+ "obs_cam_keys": [
70
+ "observation.images.cam_high",
71
+ "observation.images.cam_left_wrist",
72
+ "observation.images.cam_right_wrist"
73
+ ],
74
+ "image_hflip": false,
75
+ "camera_layout": "width_concat",
76
+ "height": 256,
77
+ "width": 320,
78
+ "action_per_frame": 16,
79
+ "frame_chunk_size": 2,
80
+ "attn_window": 72,
81
+ "num_inference_steps": 25,
82
+ "video_exec_step": -1,
83
+ "action_num_inference_steps": 50,
84
+ "guidance_scale": 5.0,
85
+ "action_guidance_scale": 1.0,
86
+ "snr_shift": 5.0,
87
+ "action_snr_shift": 1.0,
88
+ "max_sequence_length": 512,
89
+ "used_action_channel_ids": [
90
+ 0,
91
+ 1,
92
+ 2,
93
+ 3,
94
+ 4,
95
+ 5,
96
+ 6,
97
+ 7,
98
+ 8,
99
+ 9,
100
+ 10,
101
+ 11,
102
+ 12,
103
+ 13,
104
+ 14,
105
+ 15
106
+ ],
107
+ "use_relative_actions": true,
108
+ "relative_exclude_joints": [
109
+ "gripper"
110
+ ],
111
+ "action_feature_names": [
112
+ "right_joint_1.pos",
113
+ "right_joint_2.pos",
114
+ "right_joint_3.pos",
115
+ "right_joint_4.pos",
116
+ "right_joint_5.pos",
117
+ "right_joint_6.pos",
118
+ "right_joint_7.pos",
119
+ "right_gripper.pos",
120
+ "left_joint_1.pos",
121
+ "left_joint_2.pos",
122
+ "left_joint_3.pos",
123
+ "left_joint_4.pos",
124
+ "left_joint_5.pos",
125
+ "left_joint_6.pos",
126
+ "left_joint_7.pos",
127
+ "left_gripper.pos"
128
+ ],
129
+ "save_predicted_video": false,
130
+ "normalization_mapping": {
131
+ "VISUAL": "IDENTITY",
132
+ "STATE": "IDENTITY",
133
+ "ACTION": "QUANTILES"
134
+ },
135
+ "optimizer_lr": 1e-05,
136
+ "optimizer_betas": [
137
+ 0.9,
138
+ 0.95
139
+ ],
140
+ "optimizer_eps": 1e-08,
141
+ "optimizer_weight_decay": 0.0001,
142
+ "optimizer_grad_clip_norm": 1.0,
143
+ "scheduler_warmup_steps": 1000
144
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f42ede5871ae989c32bbf756d351d65f7eed0fee32f9581bf034611daca26c9d
3
+ size 20355587160
policy_postprocessor.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "IDENTITY",
19
+ "ACTION": "QUANTILES"
20
+ }
21
+ },
22
+ "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors"
23
+ },
24
+ {
25
+ "registry_name": "absolute_actions_processor",
26
+ "config": {
27
+ "enabled": true
28
+ }
29
+ },
30
+ {
31
+ "registry_name": "device_processor",
32
+ "config": {
33
+ "device": "cpu",
34
+ "float_dtype": null
35
+ }
36
+ }
37
+ ]
38
+ }
policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f79a51c2af467059f625113f1cb4b88c94a5c8a389e77143be181cf45363a34c
3
+ size 9456
policy_preprocessor.json ADDED
@@ -0,0 +1,97 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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.cam_high",
9
+ "observation.images.left_wrist": "observation.images.cam_left_wrist",
10
+ "observation.images.right_wrist": "observation.images.cam_right_wrist"
11
+ }
12
+ }
13
+ },
14
+ {
15
+ "registry_name": "to_batch_processor",
16
+ "config": {}
17
+ },
18
+ {
19
+ "registry_name": "relative_actions_processor",
20
+ "config": {
21
+ "enabled": true,
22
+ "exclude_joints": [
23
+ "gripper"
24
+ ],
25
+ "action_names": [
26
+ "right_joint_1.pos",
27
+ "right_joint_2.pos",
28
+ "right_joint_3.pos",
29
+ "right_joint_4.pos",
30
+ "right_joint_5.pos",
31
+ "right_joint_6.pos",
32
+ "right_joint_7.pos",
33
+ "right_gripper.pos",
34
+ "left_joint_1.pos",
35
+ "left_joint_2.pos",
36
+ "left_joint_3.pos",
37
+ "left_joint_4.pos",
38
+ "left_joint_5.pos",
39
+ "left_joint_6.pos",
40
+ "left_joint_7.pos",
41
+ "left_gripper.pos"
42
+ ]
43
+ }
44
+ },
45
+ {
46
+ "registry_name": "normalizer_processor",
47
+ "config": {
48
+ "eps": 1e-08,
49
+ "features": {
50
+ "observation.images.cam_high": {
51
+ "type": "VISUAL",
52
+ "shape": [
53
+ 3,
54
+ 256,
55
+ 256
56
+ ]
57
+ },
58
+ "observation.images.cam_left_wrist": {
59
+ "type": "VISUAL",
60
+ "shape": [
61
+ 3,
62
+ 256,
63
+ 256
64
+ ]
65
+ },
66
+ "observation.images.cam_right_wrist": {
67
+ "type": "VISUAL",
68
+ "shape": [
69
+ 3,
70
+ 256,
71
+ 256
72
+ ]
73
+ },
74
+ "action": {
75
+ "type": "ACTION",
76
+ "shape": [
77
+ 16
78
+ ]
79
+ }
80
+ },
81
+ "norm_map": {
82
+ "VISUAL": "IDENTITY",
83
+ "STATE": "IDENTITY",
84
+ "ACTION": "QUANTILES"
85
+ }
86
+ },
87
+ "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors"
88
+ },
89
+ {
90
+ "registry_name": "device_processor",
91
+ "config": {
92
+ "device": "cuda",
93
+ "float_dtype": null
94
+ }
95
+ }
96
+ ]
97
+ }
policy_preprocessor_step_3_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f79a51c2af467059f625113f1cb4b88c94a5c8a389e77143be181cf45363a34c
3
+ size 9456
train_config.json ADDED
@@ -0,0 +1,308 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "lerobot/high_quality_folding",
4
+ "root": "/fsx/maxime_ellerbach/datasets/high_quality_folding_rel",
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": false,
79
+ "video_backend": "pyav",
80
+ "return_uint8": false,
81
+ "depth_output_unit": "mm",
82
+ "streaming": false,
83
+ "eval_split": 0.0
84
+ },
85
+ "env": null,
86
+ "policy": {
87
+ "type": "lingbot_va",
88
+ "n_obs_steps": 1,
89
+ "input_features": {
90
+ "observation.images.cam_high": {
91
+ "type": "VISUAL",
92
+ "shape": [
93
+ 3,
94
+ 256,
95
+ 256
96
+ ]
97
+ },
98
+ "observation.images.cam_left_wrist": {
99
+ "type": "VISUAL",
100
+ "shape": [
101
+ 3,
102
+ 256,
103
+ 256
104
+ ]
105
+ },
106
+ "observation.images.cam_right_wrist": {
107
+ "type": "VISUAL",
108
+ "shape": [
109
+ 3,
110
+ 256,
111
+ 256
112
+ ]
113
+ }
114
+ },
115
+ "output_features": {
116
+ "action": {
117
+ "type": "ACTION",
118
+ "shape": [
119
+ 16
120
+ ]
121
+ }
122
+ },
123
+ "device": "cuda",
124
+ "use_amp": false,
125
+ "use_peft": false,
126
+ "push_to_hub": false,
127
+ "repo_id": "maximellerbach/folding_lingbot",
128
+ "private": null,
129
+ "tags": null,
130
+ "license": null,
131
+ "pretrained_path": "lerobot/lingbot_va_base",
132
+ "pretrained_revision": null,
133
+ "patch_size": [
134
+ 1,
135
+ 2,
136
+ 2
137
+ ],
138
+ "num_attention_heads": 24,
139
+ "attention_head_dim": 128,
140
+ "in_channels": 48,
141
+ "out_channels": 48,
142
+ "action_dim": 30,
143
+ "text_dim": 4096,
144
+ "freq_dim": 256,
145
+ "ffn_dim": 14336,
146
+ "num_layers": 30,
147
+ "cross_attn_norm": true,
148
+ "eps": 1e-06,
149
+ "rope_max_seq_len": 1024,
150
+ "attn_mode": "flex",
151
+ "wan_pretrained_path": "robbyant/lingbot-va-base",
152
+ "dtype": "bfloat16",
153
+ "text_encoder_device": "cuda",
154
+ "obs_cam_keys": [
155
+ "observation.images.cam_high",
156
+ "observation.images.cam_left_wrist",
157
+ "observation.images.cam_right_wrist"
158
+ ],
159
+ "image_hflip": false,
160
+ "camera_layout": "width_concat",
161
+ "height": 256,
162
+ "width": 320,
163
+ "action_per_frame": 16,
164
+ "frame_chunk_size": 2,
165
+ "attn_window": 72,
166
+ "num_inference_steps": 25,
167
+ "video_exec_step": -1,
168
+ "action_num_inference_steps": 50,
169
+ "guidance_scale": 5.0,
170
+ "action_guidance_scale": 1.0,
171
+ "snr_shift": 5.0,
172
+ "action_snr_shift": 1.0,
173
+ "max_sequence_length": 512,
174
+ "used_action_channel_ids": [
175
+ 0,
176
+ 1,
177
+ 2,
178
+ 3,
179
+ 4,
180
+ 5,
181
+ 6,
182
+ 7,
183
+ 8,
184
+ 9,
185
+ 10,
186
+ 11,
187
+ 12,
188
+ 13,
189
+ 14,
190
+ 15
191
+ ],
192
+ "use_relative_actions": true,
193
+ "relative_exclude_joints": [
194
+ "gripper"
195
+ ],
196
+ "action_feature_names": [
197
+ "right_joint_1.pos",
198
+ "right_joint_2.pos",
199
+ "right_joint_3.pos",
200
+ "right_joint_4.pos",
201
+ "right_joint_5.pos",
202
+ "right_joint_6.pos",
203
+ "right_joint_7.pos",
204
+ "right_gripper.pos",
205
+ "left_joint_1.pos",
206
+ "left_joint_2.pos",
207
+ "left_joint_3.pos",
208
+ "left_joint_4.pos",
209
+ "left_joint_5.pos",
210
+ "left_joint_6.pos",
211
+ "left_joint_7.pos",
212
+ "left_gripper.pos"
213
+ ],
214
+ "save_predicted_video": false,
215
+ "normalization_mapping": {
216
+ "VISUAL": "IDENTITY",
217
+ "STATE": "IDENTITY",
218
+ "ACTION": "QUANTILES"
219
+ },
220
+ "optimizer_lr": 1e-05,
221
+ "optimizer_betas": [
222
+ 0.9,
223
+ 0.95
224
+ ],
225
+ "optimizer_eps": 1e-08,
226
+ "optimizer_weight_decay": 0.0001,
227
+ "optimizer_grad_clip_norm": 1.0,
228
+ "scheduler_warmup_steps": 1000
229
+ },
230
+ "reward_model": null,
231
+ "output_dir": "outputs/train/folding_lingbot_full",
232
+ "job_name": "folding_lingbot",
233
+ "resume": false,
234
+ "seed": 1000,
235
+ "cudnn_deterministic": false,
236
+ "num_workers": 8,
237
+ "batch_size": 32,
238
+ "prefetch_factor": 4,
239
+ "persistent_workers": true,
240
+ "steps": 12711,
241
+ "env_eval_freq": 20000,
242
+ "log_freq": 50,
243
+ "eval_steps": 0,
244
+ "max_eval_samples": 0,
245
+ "tolerance_s": 0.0001,
246
+ "save_checkpoint": true,
247
+ "save_freq": 5000,
248
+ "use_policy_training_preset": true,
249
+ "optimizer": {
250
+ "type": "adamw",
251
+ "lr": 1e-05,
252
+ "weight_decay": 0.0001,
253
+ "grad_clip_norm": 1.0,
254
+ "betas": [
255
+ 0.9,
256
+ 0.95
257
+ ],
258
+ "eps": 1e-08
259
+ },
260
+ "scheduler": {
261
+ "type": "constant_with_warmup",
262
+ "num_warmup_steps": 1000
263
+ },
264
+ "eval": {
265
+ "n_episodes": 50,
266
+ "batch_size": 50,
267
+ "use_async_envs": true,
268
+ "recording": false,
269
+ "recording_repo_id": null,
270
+ "recording_private": false
271
+ },
272
+ "wandb": {
273
+ "enable": true,
274
+ "disable_artifact": true,
275
+ "project": "folding-tshirt",
276
+ "entity": null,
277
+ "notes": null,
278
+ "run_id": "3nnse2xc",
279
+ "mode": null,
280
+ "add_tags": true
281
+ },
282
+ "peft": null,
283
+ "job": {
284
+ "target": null,
285
+ "image": "huggingface/lerobot-gpu:latest",
286
+ "timeout": "2d",
287
+ "detach": false,
288
+ "tags": []
289
+ },
290
+ "save_checkpoint_to_hub": false,
291
+ "sample_weighting": {
292
+ "type": "rabc",
293
+ "progress_path": "hf://datasets/lerobot/high_quality_folding/sarm_progress.parquet",
294
+ "head_mode": "sparse",
295
+ "kappa": 0.05,
296
+ "epsilon": 1e-06,
297
+ "weighted_losses": [
298
+ "action"
299
+ ],
300
+ "extra_params": {}
301
+ },
302
+ "rename_map": {
303
+ "observation.images.base": "observation.images.cam_high",
304
+ "observation.images.left_wrist": "observation.images.cam_left_wrist",
305
+ "observation.images.right_wrist": "observation.images.cam_right_wrist"
306
+ },
307
+ "checkpoint_path": null
308
+ }