Hugging Face's logo

kimho
/
diffusion_aloha_transfer_cube_chunk64_bs64

Safetensors
Model card Files
xet
Community
kimho commited on
Commit
3a8720e
·
verified ·
1 Parent(s): 756b85c

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +82 -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 +56 -0
  6. policy_preprocessor_step_3_normalizer_processor.safetensors +3 -0
  7. train_config.json +248 -0
config.json ADDED
@@ -0,0 +1,82 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "diffusion",
3
+ "n_obs_steps": 2,
4
+ "input_features": {
5
+ "observation.images.top": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 480,
10
+ 640
11
+ ]
12
+ },
13
+ "observation.state": {
14
+ "type": "STATE",
15
+ "shape": [
16
+ 14
17
+ ]
18
+ }
19
+ },
20
+ "output_features": {
21
+ "action": {
22
+ "type": "ACTION",
23
+ "shape": [
24
+ 14
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": null,
36
+ "horizon": 64,
37
+ "n_action_steps": 64,
38
+ "normalization_mapping": {
39
+ "VISUAL": "MEAN_STD",
40
+ "STATE": "MIN_MAX",
41
+ "ACTION": "MIN_MAX"
42
+ },
43
+ "drop_n_last_frames": 7,
44
+ "vision_backbone": "resnet18",
45
+ "crop_shape": [
46
+ 480,
47
+ 640
48
+ ],
49
+ "crop_is_random": false,
50
+ "pretrained_backbone_weights": null,
51
+ "use_group_norm": true,
52
+ "spatial_softmax_num_keypoints": 32,
53
+ "use_separate_rgb_encoder_per_camera": false,
54
+ "down_dims": [
55
+ 512,
56
+ 1024,
57
+ 2048
58
+ ],
59
+ "kernel_size": 5,
60
+ "n_groups": 8,
61
+ "diffusion_step_embed_dim": 128,
62
+ "use_film_scale_modulation": true,
63
+ "noise_scheduler_type": "DDPM",
64
+ "num_train_timesteps": 100,
65
+ "beta_schedule": "squaredcos_cap_v2",
66
+ "beta_start": 0.0001,
67
+ "beta_end": 0.02,
68
+ "prediction_type": "epsilon",
69
+ "clip_sample": true,
70
+ "clip_sample_range": 1.0,
71
+ "num_inference_steps": null,
72
+ "do_mask_loss_for_padding": false,
73
+ "optimizer_lr": 0.0001,
74
+ "optimizer_betas": [
75
+ 0.95,
76
+ 0.999
77
+ ],
78
+ "optimizer_eps": 1e-08,
79
+ "optimizer_weight_decay": 1e-06,
80
+ "scheduler_name": "cosine",
81
+ "scheduler_warmup_steps": 500
82
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3d11b94bbceeb7d74fdf06a295370354a98404216245d338d84871412b82508
3
+ size 1053788376
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
+ 14
13
+ ]
14
+ }
15
+ },
16
+ "norm_map": {
17
+ "VISUAL": "MEAN_STD",
18
+ "STATE": "MIN_MAX",
19
+ "ACTION": "MIN_MAX"
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 sha256:797777e6e870b9c70a950b6860aa23d246ee8292a6f10f4bca727ab564e39348
3
+ size 3892
policy_preprocessor.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "device_processor",
16
+ "config": {
17
+ "device": "cuda",
18
+ "float_dtype": null
19
+ }
20
+ },
21
+ {
22
+ "registry_name": "normalizer_processor",
23
+ "config": {
24
+ "eps": 1e-08,
25
+ "features": {
26
+ "observation.images.top": {
27
+ "type": "VISUAL",
28
+ "shape": [
29
+ 3,
30
+ 480,
31
+ 640
32
+ ]
33
+ },
34
+ "observation.state": {
35
+ "type": "STATE",
36
+ "shape": [
37
+ 14
38
+ ]
39
+ },
40
+ "action": {
41
+ "type": "ACTION",
42
+ "shape": [
43
+ 14
44
+ ]
45
+ }
46
+ },
47
+ "norm_map": {
48
+ "VISUAL": "MEAN_STD",
49
+ "STATE": "MIN_MAX",
50
+ "ACTION": "MIN_MAX"
51
+ }
52
+ },
53
+ "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors"
54
+ }
55
+ ]
56
+ }
policy_preprocessor_step_3_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:797777e6e870b9c70a950b6860aa23d246ee8292a6f10f4bca727ab564e39348
3
+ size 3892
train_config.json ADDED
@@ -0,0 +1,248 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "lerobot/aloha_sim_transfer_cube_human",
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": {
83
+ "type": "aloha",
84
+ "task": "AlohaTransferCube-v0",
85
+ "fps": 50,
86
+ "features": {
87
+ "action": {
88
+ "type": "ACTION",
89
+ "shape": [
90
+ 14
91
+ ]
92
+ },
93
+ "agent_pos": {
94
+ "type": "STATE",
95
+ "shape": [
96
+ 14
97
+ ]
98
+ },
99
+ "pixels/top": {
100
+ "type": "VISUAL",
101
+ "shape": [
102
+ 480,
103
+ 640,
104
+ 3
105
+ ]
106
+ }
107
+ },
108
+ "features_map": {
109
+ "action": "action",
110
+ "agent_pos": "observation.state",
111
+ "top": "observation.image.top",
112
+ "pixels/top": "observation.images.top"
113
+ },
114
+ "max_parallel_tasks": 1,
115
+ "disable_env_checker": true,
116
+ "episode_length": 400,
117
+ "obs_type": "pixels_agent_pos",
118
+ "observation_height": 480,
119
+ "observation_width": 640,
120
+ "render_mode": "rgb_array"
121
+ },
122
+ "policy": {
123
+ "type": "diffusion",
124
+ "n_obs_steps": 2,
125
+ "input_features": {
126
+ "observation.images.top": {
127
+ "type": "VISUAL",
128
+ "shape": [
129
+ 3,
130
+ 480,
131
+ 640
132
+ ]
133
+ },
134
+ "observation.state": {
135
+ "type": "STATE",
136
+ "shape": [
137
+ 14
138
+ ]
139
+ }
140
+ },
141
+ "output_features": {
142
+ "action": {
143
+ "type": "ACTION",
144
+ "shape": [
145
+ 14
146
+ ]
147
+ }
148
+ },
149
+ "device": "cuda",
150
+ "use_amp": false,
151
+ "push_to_hub": false,
152
+ "repo_id": null,
153
+ "private": null,
154
+ "tags": null,
155
+ "license": null,
156
+ "pretrained_path": null,
157
+ "horizon": 64,
158
+ "n_action_steps": 64,
159
+ "normalization_mapping": {
160
+ "VISUAL": "MEAN_STD",
161
+ "STATE": "MIN_MAX",
162
+ "ACTION": "MIN_MAX"
163
+ },
164
+ "drop_n_last_frames": 7,
165
+ "vision_backbone": "resnet18",
166
+ "crop_shape": [
167
+ 480,
168
+ 640
169
+ ],
170
+ "crop_is_random": false,
171
+ "pretrained_backbone_weights": null,
172
+ "use_group_norm": true,
173
+ "spatial_softmax_num_keypoints": 32,
174
+ "use_separate_rgb_encoder_per_camera": false,
175
+ "down_dims": [
176
+ 512,
177
+ 1024,
178
+ 2048
179
+ ],
180
+ "kernel_size": 5,
181
+ "n_groups": 8,
182
+ "diffusion_step_embed_dim": 128,
183
+ "use_film_scale_modulation": true,
184
+ "noise_scheduler_type": "DDPM",
185
+ "num_train_timesteps": 100,
186
+ "beta_schedule": "squaredcos_cap_v2",
187
+ "beta_start": 0.0001,
188
+ "beta_end": 0.02,
189
+ "prediction_type": "epsilon",
190
+ "clip_sample": true,
191
+ "clip_sample_range": 1.0,
192
+ "num_inference_steps": null,
193
+ "do_mask_loss_for_padding": false,
194
+ "optimizer_lr": 0.0001,
195
+ "optimizer_betas": [
196
+ 0.95,
197
+ 0.999
198
+ ],
199
+ "optimizer_eps": 1e-08,
200
+ "optimizer_weight_decay": 1e-06,
201
+ "scheduler_name": "cosine",
202
+ "scheduler_warmup_steps": 500
203
+ },
204
+ "output_dir": "outputs/train/2025_11_29/diffusion_aloha_transfer_cube_chunk64_bs64",
205
+ "job_name": "diffusion_transfer_cube_chunk64_bs64",
206
+ "resume": false,
207
+ "seed": 1000,
208
+ "num_workers": 4,
209
+ "batch_size": 64,
210
+ "steps": 200000,
211
+ "eval_freq": 50000,
212
+ "log_freq": 200,
213
+ "save_checkpoint": true,
214
+ "save_freq": 50000,
215
+ "use_policy_training_preset": true,
216
+ "optimizer": {
217
+ "type": "adam",
218
+ "lr": 0.0001,
219
+ "weight_decay": 1e-06,
220
+ "grad_clip_norm": 10.0,
221
+ "betas": [
222
+ 0.95,
223
+ 0.999
224
+ ],
225
+ "eps": 1e-08
226
+ },
227
+ "scheduler": {
228
+ "type": "diffuser",
229
+ "num_warmup_steps": 500,
230
+ "name": "cosine"
231
+ },
232
+ "eval": {
233
+ "n_episodes": 10,
234
+ "batch_size": 10,
235
+ "use_async_envs": false
236
+ },
237
+ "wandb": {
238
+ "enable": true,
239
+ "disable_artifact": false,
240
+ "project": "lerobot",
241
+ "entity": null,
242
+ "notes": null,
243
+ "run_id": "6juq1zrc",
244
+ "mode": null
245
+ },
246
+ "checkpoint_path": null,
247
+ "rename_map": {}
248
+ }