Hugging Face's logo

kd-forge
/
diffusion-wb2-sort_sharp_merged_27ep

Safetensors
Model card Files
xet
 Community
msmandelbrot commited on
Commit
cfeef07
·
verified ·
1 Parent(s): c73b488

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +104 -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 +80 -0
  6. policy_preprocessor_step_3_normalizer_processor.safetensors +3 -0
  7. train_config.json +243 -0
config.json ADDED
@@ -0,0 +1,104 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "diffusion",
3
+ "n_obs_steps": 2,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 12
9
+ ]
10
+ },
11
+ "observation.images.left_up": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 480,
16
+ 640
17
+ ]
18
+ },
19
+ "observation.images.left_right": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 480,
24
+ 640
25
+ ]
26
+ },
27
+ "observation.images.left_left": {
28
+ "type": "VISUAL",
29
+ "shape": [
30
+ 3,
31
+ 480,
32
+ 640
33
+ ]
34
+ },
35
+ "observation.images.left_front": {
36
+ "type": "VISUAL",
37
+ "shape": [
38
+ 3,
39
+ 480,
40
+ 640
41
+ ]
42
+ }
43
+ },
44
+ "output_features": {
45
+ "action": {
46
+ "type": "ACTION",
47
+ "shape": [
48
+ 12
49
+ ]
50
+ }
51
+ },
52
+ "device": "cuda",
53
+ "use_amp": false,
54
+ "use_peft": false,
55
+ "push_to_hub": false,
56
+ "repo_id": "/diffusion_sort_sharp_merged",
57
+ "private": null,
58
+ "tags": null,
59
+ "license": null,
60
+ "pretrained_path": null,
61
+ "horizon": 16,
62
+ "n_action_steps": 8,
63
+ "normalization_mapping": {
64
+ "VISUAL": "MEAN_STD",
65
+ "STATE": "MIN_MAX",
66
+ "ACTION": "MIN_MAX"
67
+ },
68
+ "drop_n_last_frames": 7,
69
+ "vision_backbone": "resnet18",
70
+ "crop_shape": null,
71
+ "crop_is_random": true,
72
+ "pretrained_backbone_weights": null,
73
+ "use_group_norm": true,
74
+ "spatial_softmax_num_keypoints": 32,
75
+ "use_separate_rgb_encoder_per_camera": false,
76
+ "down_dims": [
77
+ 512,
78
+ 1024,
79
+ 2048
80
+ ],
81
+ "kernel_size": 5,
82
+ "n_groups": 8,
83
+ "diffusion_step_embed_dim": 128,
84
+ "use_film_scale_modulation": true,
85
+ "noise_scheduler_type": "DDPM",
86
+ "num_train_timesteps": 100,
87
+ "beta_schedule": "squaredcos_cap_v2",
88
+ "beta_start": 0.0001,
89
+ "beta_end": 0.02,
90
+ "prediction_type": "epsilon",
91
+ "clip_sample": true,
92
+ "clip_sample_range": 1.0,
93
+ "num_inference_steps": 10,
94
+ "do_mask_loss_for_padding": false,
95
+ "optimizer_lr": 0.0001,
96
+ "optimizer_betas": [
97
+ 0.95,
98
+ 0.999
99
+ ],
100
+ "optimizer_eps": 1e-08,
101
+ "optimizer_weight_decay": 1e-06,
102
+ "scheduler_name": "cosine",
103
+ "scheduler_warmup_steps": 500
104
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e512b21f15055a01d155967d43ab7fbe78acd35c018f31fc8efcb8fb929f1b5
3
+ size 1097341152
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
+ 12
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 sha256:46525c6698fc365a697c1324cb8128f18bbe43e2663595a18a6cac16436b0348
3
+ size 10208
policy_preprocessor.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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.state": {
27
+ "type": "STATE",
28
+ "shape": [
29
+ 12
30
+ ]
31
+ },
32
+ "observation.images.left_up": {
33
+ "type": "VISUAL",
34
+ "shape": [
35
+ 3,
36
+ 480,
37
+ 640
38
+ ]
39
+ },
40
+ "observation.images.left_right": {
41
+ "type": "VISUAL",
42
+ "shape": [
43
+ 3,
44
+ 480,
45
+ 640
46
+ ]
47
+ },
48
+ "observation.images.left_left": {
49
+ "type": "VISUAL",
50
+ "shape": [
51
+ 3,
52
+ 480,
53
+ 640
54
+ ]
55
+ },
56
+ "observation.images.left_front": {
57
+ "type": "VISUAL",
58
+ "shape": [
59
+ 3,
60
+ 480,
61
+ 640
62
+ ]
63
+ },
64
+ "action": {
65
+ "type": "ACTION",
66
+ "shape": [
67
+ 12
68
+ ]
69
+ }
70
+ },
71
+ "norm_map": {
72
+ "VISUAL": "MEAN_STD",
73
+ "STATE": "MIN_MAX",
74
+ "ACTION": "MIN_MAX"
75
+ }
76
+ },
77
+ "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors"
78
+ }
79
+ ]
80
+ }
policy_preprocessor_step_3_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:46525c6698fc365a697c1324cb8128f18bbe43e2663595a18a6cac16436b0348
3
+ size 10208
train_config.json ADDED
@@ -0,0 +1,243 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "kd-forge/sort_sharp_merged",
4
+ "root": null,
5
+ "episodes": null,
6
+ "image_transforms": {
7
+ "enable": true,
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": "diffusion",
85
+ "n_obs_steps": 2,
86
+ "input_features": {
87
+ "observation.state": {
88
+ "type": "STATE",
89
+ "shape": [
90
+ 12
91
+ ]
92
+ },
93
+ "observation.images.left_up": {
94
+ "type": "VISUAL",
95
+ "shape": [
96
+ 3,
97
+ 480,
98
+ 640
99
+ ]
100
+ },
101
+ "observation.images.left_right": {
102
+ "type": "VISUAL",
103
+ "shape": [
104
+ 3,
105
+ 480,
106
+ 640
107
+ ]
108
+ },
109
+ "observation.images.left_left": {
110
+ "type": "VISUAL",
111
+ "shape": [
112
+ 3,
113
+ 480,
114
+ 640
115
+ ]
116
+ },
117
+ "observation.images.left_front": {
118
+ "type": "VISUAL",
119
+ "shape": [
120
+ 3,
121
+ 480,
122
+ 640
123
+ ]
124
+ }
125
+ },
126
+ "output_features": {
127
+ "action": {
128
+ "type": "ACTION",
129
+ "shape": [
130
+ 12
131
+ ]
132
+ }
133
+ },
134
+ "device": "cuda",
135
+ "use_amp": false,
136
+ "use_peft": false,
137
+ "push_to_hub": false,
138
+ "repo_id": "/diffusion_sort_sharp_merged",
139
+ "private": null,
140
+ "tags": null,
141
+ "license": null,
142
+ "pretrained_path": null,
143
+ "horizon": 16,
144
+ "n_action_steps": 8,
145
+ "normalization_mapping": {
146
+ "VISUAL": "MEAN_STD",
147
+ "STATE": "MIN_MAX",
148
+ "ACTION": "MIN_MAX"
149
+ },
150
+ "drop_n_last_frames": 7,
151
+ "vision_backbone": "resnet18",
152
+ "crop_shape": null,
153
+ "crop_is_random": true,
154
+ "pretrained_backbone_weights": null,
155
+ "use_group_norm": true,
156
+ "spatial_softmax_num_keypoints": 32,
157
+ "use_separate_rgb_encoder_per_camera": false,
158
+ "down_dims": [
159
+ 512,
160
+ 1024,
161
+ 2048
162
+ ],
163
+ "kernel_size": 5,
164
+ "n_groups": 8,
165
+ "diffusion_step_embed_dim": 128,
166
+ "use_film_scale_modulation": true,
167
+ "noise_scheduler_type": "DDPM",
168
+ "num_train_timesteps": 100,
169
+ "beta_schedule": "squaredcos_cap_v2",
170
+ "beta_start": 0.0001,
171
+ "beta_end": 0.02,
172
+ "prediction_type": "epsilon",
173
+ "clip_sample": true,
174
+ "clip_sample_range": 1.0,
175
+ "num_inference_steps": 10,
176
+ "do_mask_loss_for_padding": false,
177
+ "optimizer_lr": 0.0001,
178
+ "optimizer_betas": [
179
+ 0.95,
180
+ 0.999
181
+ ],
182
+ "optimizer_eps": 1e-08,
183
+ "optimizer_weight_decay": 1e-06,
184
+ "scheduler_name": "cosine",
185
+ "scheduler_warmup_steps": 500
186
+ },
187
+ "output_dir": "outputs/train/diffusion_sort_sharp_merged",
188
+ "job_name": "diffusion_sort_sharp_merged",
189
+ "resume": false,
190
+ "seed": 1000,
191
+ "num_workers": 4,
192
+ "batch_size": 10,
193
+ "steps": 1000000,
194
+ "eval_freq": 20000,
195
+ "log_freq": 100,
196
+ "tolerance_s": 0.0001,
197
+ "save_checkpoint": true,
198
+ "save_freq": 50000,
199
+ "use_policy_training_preset": true,
200
+ "optimizer": {
201
+ "type": "adam",
202
+ "lr": 0.0001,
203
+ "weight_decay": 1e-06,
204
+ "grad_clip_norm": 10.0,
205
+ "betas": [
206
+ 0.95,
207
+ 0.999
208
+ ],
209
+ "eps": 1e-08
210
+ },
211
+ "scheduler": {
212
+ "type": "diffuser",
213
+ "num_warmup_steps": 500,
214
+ "name": "cosine"
215
+ },
216
+ "eval": {
217
+ "n_episodes": 50,
218
+ "batch_size": 50,
219
+ "use_async_envs": false
220
+ },
221
+ "wandb": {
222
+ "enable": false,
223
+ "disable_artifact": false,
224
+ "project": "lerobot",
225
+ "entity": null,
226
+ "notes": null,
227
+ "run_id": null,
228
+ "mode": null
229
+ },
230
+ "clearml": {
231
+ "enable": true,
232
+ "project": "lerobot/diffusion",
233
+ "output_uri": null
234
+ },
235
+ "peft": null,
236
+ "checkpoint_path": null,
237
+ "rename_map": {},
238
+ "use_rabc": false,
239
+ "rabc_progress_path": null,
240
+ "rabc_head_mode": "sparse",
241
+ "rabc_kappa": 0.01,
242
+ "rabc_epsilon": 1e-06
243
+ }