Hugging Face's logo

AleksanderWojsz
/
diffusion_training

Safetensors
Model card Files
xet
 Community
AleksanderWojsz commited on
Commit
b26ad4a
·
verified ·
1 Parent(s): fc24ef4

Upload model from diffusion_training

Browse files
Files changed (7) hide show
  1. config.json +80 -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 +214 -0
config.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "diffusion",
3
+ "n_obs_steps": 2,
4
+ "input_features": {
5
+ "observation.images.top_camera": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 224,
10
+ 224
11
+ ]
12
+ },
13
+ "observation.state": {
14
+ "type": "STATE",
15
+ "shape": [
16
+ 2
17
+ ]
18
+ }
19
+ },
20
+ "output_features": {
21
+ "action": {
22
+ "type": "ACTION",
23
+ "shape": [
24
+ 2
25
+ ]
26
+ }
27
+ },
28
+ "device": "cuda",
29
+ "use_amp": false,
30
+ "use_peft": false,
31
+ "push_to_hub": false,
32
+ "repo_id": null,
33
+ "private": null,
34
+ "tags": null,
35
+ "license": null,
36
+ "pretrained_path": null,
37
+ "horizon": 32,
38
+ "n_action_steps": 16,
39
+ "normalization_mapping": {
40
+ "VISUAL": "MEAN_STD",
41
+ "STATE": "MIN_MAX",
42
+ "ACTION": "MIN_MAX"
43
+ },
44
+ "drop_n_last_frames": 7,
45
+ "vision_backbone": "resnet18",
46
+ "crop_shape": null,
47
+ "crop_is_random": true,
48
+ "pretrained_backbone_weights": null,
49
+ "use_group_norm": true,
50
+ "spatial_softmax_num_keypoints": 32,
51
+ "use_separate_rgb_encoder_per_camera": false,
52
+ "down_dims": [
53
+ 512,
54
+ 1024,
55
+ 2048
56
+ ],
57
+ "kernel_size": 5,
58
+ "n_groups": 8,
59
+ "diffusion_step_embed_dim": 128,
60
+ "use_film_scale_modulation": true,
61
+ "noise_scheduler_type": "DDPM",
62
+ "num_train_timesteps": 100,
63
+ "beta_schedule": "squaredcos_cap_v2",
64
+ "beta_start": 0.0001,
65
+ "beta_end": 0.02,
66
+ "prediction_type": "epsilon",
67
+ "clip_sample": true,
68
+ "clip_sample_range": 1.0,
69
+ "num_inference_steps": null,
70
+ "do_mask_loss_for_padding": false,
71
+ "optimizer_lr": 0.0001,
72
+ "optimizer_betas": [
73
+ 0.95,
74
+ 0.999
75
+ ],
76
+ "optimizer_eps": 1e-08,
77
+ "optimizer_weight_decay": 1e-06,
78
+ "scheduler_name": "cosine",
79
+ "scheduler_warmup_steps": 500
80
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:82f2cb8bb21b7a305d2ebe5af5fb2bd3b8fad35bc775d72629b978479e08a8ca
3
+ size 1050861768
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
+ 2
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 sha256:0c857034c6535d6f765755e721fb48fd855d019774f517e13ec105987bf45a60
3
+ size 6304
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_camera": {
27
+ "type": "VISUAL",
28
+ "shape": [
29
+ 3,
30
+ 224,
31
+ 224
32
+ ]
33
+ },
34
+ "observation.state": {
35
+ "type": "STATE",
36
+ "shape": [
37
+ 2
38
+ ]
39
+ },
40
+ "action": {
41
+ "type": "ACTION",
42
+ "shape": [
43
+ 2
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 sha256:0c857034c6535d6f765755e721fb48fd855d019774f517e13ec105987bf45a60
3
+ size 6304
train_config.json ADDED
@@ -0,0 +1,214 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "pusht",
4
+ "root": "data",
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": "diffusion",
85
+ "n_obs_steps": 2,
86
+ "input_features": {
87
+ "observation.images.top_camera": {
88
+ "type": "VISUAL",
89
+ "shape": [
90
+ 3,
91
+ 224,
92
+ 224
93
+ ]
94
+ },
95
+ "observation.state": {
96
+ "type": "STATE",
97
+ "shape": [
98
+ 2
99
+ ]
100
+ }
101
+ },
102
+ "output_features": {
103
+ "action": {
104
+ "type": "ACTION",
105
+ "shape": [
106
+ 2
107
+ ]
108
+ }
109
+ },
110
+ "device": "cuda",
111
+ "use_amp": false,
112
+ "use_peft": false,
113
+ "push_to_hub": false,
114
+ "repo_id": null,
115
+ "private": null,
116
+ "tags": null,
117
+ "license": null,
118
+ "pretrained_path": null,
119
+ "horizon": 32,
120
+ "n_action_steps": 16,
121
+ "normalization_mapping": {
122
+ "VISUAL": "MEAN_STD",
123
+ "STATE": "MIN_MAX",
124
+ "ACTION": "MIN_MAX"
125
+ },
126
+ "drop_n_last_frames": 7,
127
+ "vision_backbone": "resnet18",
128
+ "crop_shape": null,
129
+ "crop_is_random": true,
130
+ "pretrained_backbone_weights": null,
131
+ "use_group_norm": true,
132
+ "spatial_softmax_num_keypoints": 32,
133
+ "use_separate_rgb_encoder_per_camera": false,
134
+ "down_dims": [
135
+ 512,
136
+ 1024,
137
+ 2048
138
+ ],
139
+ "kernel_size": 5,
140
+ "n_groups": 8,
141
+ "diffusion_step_embed_dim": 128,
142
+ "use_film_scale_modulation": true,
143
+ "noise_scheduler_type": "DDPM",
144
+ "num_train_timesteps": 100,
145
+ "beta_schedule": "squaredcos_cap_v2",
146
+ "beta_start": 0.0001,
147
+ "beta_end": 0.02,
148
+ "prediction_type": "epsilon",
149
+ "clip_sample": true,
150
+ "clip_sample_range": 1.0,
151
+ "num_inference_steps": null,
152
+ "do_mask_loss_for_padding": false,
153
+ "optimizer_lr": 0.0001,
154
+ "optimizer_betas": [
155
+ 0.95,
156
+ 0.999
157
+ ],
158
+ "optimizer_eps": 1e-08,
159
+ "optimizer_weight_decay": 1e-06,
160
+ "scheduler_name": "cosine",
161
+ "scheduler_warmup_steps": 500
162
+ },
163
+ "output_dir": "outputs/diffusion_training",
164
+ "job_name": "diffusion_training_local",
165
+ "resume": false,
166
+ "seed": 1000,
167
+ "num_workers": 4,
168
+ "batch_size": 32,
169
+ "steps": 100000,
170
+ "eval_freq": 20000,
171
+ "log_freq": 200,
172
+ "tolerance_s": 0.0001,
173
+ "save_checkpoint": true,
174
+ "save_freq": 10000,
175
+ "use_policy_training_preset": true,
176
+ "optimizer": {
177
+ "type": "adam",
178
+ "lr": 0.0001,
179
+ "weight_decay": 1e-06,
180
+ "grad_clip_norm": 10.0,
181
+ "betas": [
182
+ 0.95,
183
+ 0.999
184
+ ],
185
+ "eps": 1e-08
186
+ },
187
+ "scheduler": {
188
+ "type": "diffuser",
189
+ "num_warmup_steps": 500,
190
+ "name": "cosine"
191
+ },
192
+ "eval": {
193
+ "n_episodes": 50,
194
+ "batch_size": 50,
195
+ "use_async_envs": false
196
+ },
197
+ "wandb": {
198
+ "enable": false,
199
+ "disable_artifact": false,
200
+ "project": "lerobot",
201
+ "entity": null,
202
+ "notes": null,
203
+ "run_id": null,
204
+ "mode": null
205
+ },
206
+ "peft": null,
207
+ "use_rabc": false,
208
+ "rabc_progress_path": null,
209
+ "rabc_kappa": 0.01,
210
+ "rabc_epsilon": 1e-06,
211
+ "rabc_head_mode": "sparse",
212
+ "rename_map": {},
213
+ "checkpoint_path": null
214
+ }