Hugging Face's logo

kd-forge
/
diffusion-demo-all-sorting-2

Robotics
LeRobot
Safetensors
diffusion
Model card Files
xet
Community
tinkhireeva commited on
Commit
ac8e981
·
verified ·
1 Parent(s): 1827a85

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +95 -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 +72 -0
  6. policy_preprocessor_step_3_normalizer_processor.safetensors +3 -0
  7. train_config.json +227 -0
config.json ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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.up": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 480,
16
+ 640
17
+ ]
18
+ },
19
+ "observation.images.right": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 480,
24
+ 640
25
+ ]
26
+ },
27
+ "observation.images.left": {
28
+ "type": "VISUAL",
29
+ "shape": [
30
+ 3,
31
+ 480,
32
+ 640
33
+ ]
34
+ }
35
+ },
36
+ "output_features": {
37
+ "action": {
38
+ "type": "ACTION",
39
+ "shape": [
40
+ 12
41
+ ]
42
+ }
43
+ },
44
+ "device": "cuda",
45
+ "use_amp": false,
46
+ "push_to_hub": true,
47
+ "repo_id": "kd-forge/diffusion-demo-all-sorting-2",
48
+ "private": null,
49
+ "tags": null,
50
+ "license": null,
51
+ "pretrained_path": null,
52
+ "horizon": 16,
53
+ "n_action_steps": 8,
54
+ "normalization_mapping": {
55
+ "VISUAL": "MEAN_STD",
56
+ "STATE": "MIN_MAX",
57
+ "ACTION": "MIN_MAX"
58
+ },
59
+ "drop_n_last_frames": 2,
60
+ "vision_backbone": "resnet34",
61
+ "crop_shape": null,
62
+ "crop_is_random": true,
63
+ "pretrained_backbone_weights": null,
64
+ "use_group_norm": true,
65
+ "spatial_softmax_num_keypoints": 32,
66
+ "use_separate_rgb_encoder_per_camera": true,
67
+ "down_dims": [
68
+ 512,
69
+ 1024,
70
+ 2048
71
+ ],
72
+ "kernel_size": 5,
73
+ "n_groups": 8,
74
+ "diffusion_step_embed_dim": 128,
75
+ "use_film_scale_modulation": true,
76
+ "noise_scheduler_type": "DDIM",
77
+ "num_train_timesteps": 100,
78
+ "beta_schedule": "squaredcos_cap_v2",
79
+ "beta_start": 0.0001,
80
+ "beta_end": 0.02,
81
+ "prediction_type": "epsilon",
82
+ "clip_sample": true,
83
+ "clip_sample_range": 1.0,
84
+ "num_inference_steps": 10,
85
+ "do_mask_loss_for_padding": false,
86
+ "optimizer_lr": 0.0001,
87
+ "optimizer_betas": [
88
+ 0.95,
89
+ 0.999
90
+ ],
91
+ "optimizer_eps": 1e-08,
92
+ "optimizer_weight_decay": 1e-06,
93
+ "scheduler_name": "cosine",
94
+ "scheduler_warmup_steps": 500
95
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:541b6c1bde2cb809df974d65ef062255d81eb75e27c40b448c4d34d3ca62e4c9
3
+ size 1293573184
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 sha256:7bdf21d2d6bd6c7c1bfe6e88e8b8ba9ef9a89e2ab7578fded03310ea33956d24
3
+ size 9008
policy_preprocessor.json ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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.up": {
33
+ "type": "VISUAL",
34
+ "shape": [
35
+ 3,
36
+ 480,
37
+ 640
38
+ ]
39
+ },
40
+ "observation.images.right": {
41
+ "type": "VISUAL",
42
+ "shape": [
43
+ 3,
44
+ 480,
45
+ 640
46
+ ]
47
+ },
48
+ "observation.images.left": {
49
+ "type": "VISUAL",
50
+ "shape": [
51
+ 3,
52
+ 480,
53
+ 640
54
+ ]
55
+ },
56
+ "action": {
57
+ "type": "ACTION",
58
+ "shape": [
59
+ 12
60
+ ]
61
+ }
62
+ },
63
+ "norm_map": {
64
+ "VISUAL": "MEAN_STD",
65
+ "STATE": "MIN_MAX",
66
+ "ACTION": "MIN_MAX"
67
+ }
68
+ },
69
+ "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors"
70
+ }
71
+ ]
72
+ }
policy_preprocessor_step_3_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7bdf21d2d6bd6c7c1bfe6e88e8b8ba9ef9a89e2ab7578fded03310ea33956d24
3
+ size 9008
train_config.json ADDED
@@ -0,0 +1,227 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "kd-forge/demo-all-sorting",
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.up": {
94
+ "type": "VISUAL",
95
+ "shape": [
96
+ 3,
97
+ 480,
98
+ 640
99
+ ]
100
+ },
101
+ "observation.images.right": {
102
+ "type": "VISUAL",
103
+ "shape": [
104
+ 3,
105
+ 480,
106
+ 640
107
+ ]
108
+ },
109
+ "observation.images.left": {
110
+ "type": "VISUAL",
111
+ "shape": [
112
+ 3,
113
+ 480,
114
+ 640
115
+ ]
116
+ }
117
+ },
118
+ "output_features": {
119
+ "action": {
120
+ "type": "ACTION",
121
+ "shape": [
122
+ 12
123
+ ]
124
+ }
125
+ },
126
+ "device": "cuda",
127
+ "use_amp": false,
128
+ "push_to_hub": true,
129
+ "repo_id": "kd-forge/diffusion-demo-all-sorting-2",
130
+ "private": null,
131
+ "tags": null,
132
+ "license": null,
133
+ "pretrained_path": null,
134
+ "horizon": 16,
135
+ "n_action_steps": 8,
136
+ "normalization_mapping": {
137
+ "VISUAL": "MEAN_STD",
138
+ "STATE": "MIN_MAX",
139
+ "ACTION": "MIN_MAX"
140
+ },
141
+ "drop_n_last_frames": 2,
142
+ "vision_backbone": "resnet34",
143
+ "crop_shape": null,
144
+ "crop_is_random": true,
145
+ "pretrained_backbone_weights": null,
146
+ "use_group_norm": true,
147
+ "spatial_softmax_num_keypoints": 32,
148
+ "use_separate_rgb_encoder_per_camera": true,
149
+ "down_dims": [
150
+ 512,
151
+ 1024,
152
+ 2048
153
+ ],
154
+ "kernel_size": 5,
155
+ "n_groups": 8,
156
+ "diffusion_step_embed_dim": 128,
157
+ "use_film_scale_modulation": true,
158
+ "noise_scheduler_type": "DDIM",
159
+ "num_train_timesteps": 100,
160
+ "beta_schedule": "squaredcos_cap_v2",
161
+ "beta_start": 0.0001,
162
+ "beta_end": 0.02,
163
+ "prediction_type": "epsilon",
164
+ "clip_sample": true,
165
+ "clip_sample_range": 1.0,
166
+ "num_inference_steps": 10,
167
+ "do_mask_loss_for_padding": false,
168
+ "optimizer_lr": 0.0001,
169
+ "optimizer_betas": [
170
+ 0.95,
171
+ 0.999
172
+ ],
173
+ "optimizer_eps": 1e-08,
174
+ "optimizer_weight_decay": 1e-06,
175
+ "scheduler_name": "cosine",
176
+ "scheduler_warmup_steps": 500
177
+ },
178
+ "output_dir": "outputs/kd-forge/diffusion-demo-all-sorting-2",
179
+ "job_name": "diffusion",
180
+ "resume": false,
181
+ "seed": 1000,
182
+ "num_workers": 4,
183
+ "batch_size": 8,
184
+ "steps": 300000,
185
+ "eval_freq": 20000,
186
+ "log_freq": 200,
187
+ "save_checkpoint": true,
188
+ "save_freq": 20000,
189
+ "use_policy_training_preset": true,
190
+ "optimizer": {
191
+ "type": "adam",
192
+ "lr": 0.0001,
193
+ "weight_decay": 1e-06,
194
+ "grad_clip_norm": 10.0,
195
+ "betas": [
196
+ 0.95,
197
+ 0.999
198
+ ],
199
+ "eps": 1e-08
200
+ },
201
+ "scheduler": {
202
+ "type": "diffuser",
203
+ "num_warmup_steps": 500,
204
+ "name": "cosine"
205
+ },
206
+ "eval": {
207
+ "n_episodes": 50,
208
+ "batch_size": 50,
209
+ "use_async_envs": false
210
+ },
211
+ "wandb": {
212
+ "enable": false,
213
+ "disable_artifact": false,
214
+ "project": "lerobot",
215
+ "entity": null,
216
+ "notes": null,
217
+ "run_id": null,
218
+ "mode": null
219
+ },
220
+ "clearml": {
221
+ "enable": true,
222
+ "project": "lerobot/diffusion",
223
+ "output_uri": null
224
+ },
225
+ "checkpoint_path": null,
226
+ "rename_map": {}
227
+ }