Hugging Face's logo

Liyux
/
pusht_diffusion_v5

Safetensors
Model card Files
xet
Community
Liyux commited on
Commit
40edb4c
·
verified ·
1 Parent(s): fcfab27

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +79 -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 +206 -0
config.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "diffusion",
3
+ "n_obs_steps": 2,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 6
9
+ ]
10
+ },
11
+ "observation.images.top": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 240,
16
+ 320
17
+ ]
18
+ }
19
+ },
20
+ "output_features": {
21
+ "action": {
22
+ "type": "ACTION",
23
+ "shape": [
24
+ 6
25
+ ]
26
+ }
27
+ },
28
+ "device": "cuda:0",
29
+ "use_amp": false,
30
+ "push_to_hub": false,
31
+ "repo_id": "yuxianli/pusht_diffusion_v5",
32
+ "private": null,
33
+ "tags": null,
34
+ "license": null,
35
+ "pretrained_path": null,
36
+ "horizon": 16,
37
+ "n_action_steps": 8,
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": null,
46
+ "crop_is_random": true,
47
+ "pretrained_backbone_weights": null,
48
+ "use_group_norm": true,
49
+ "spatial_softmax_num_keypoints": 32,
50
+ "use_separate_rgb_encoder_per_camera": false,
51
+ "down_dims": [
52
+ 512,
53
+ 1024,
54
+ 2048
55
+ ],
56
+ "kernel_size": 5,
57
+ "n_groups": 8,
58
+ "diffusion_step_embed_dim": 128,
59
+ "use_film_scale_modulation": true,
60
+ "noise_scheduler_type": "DDPM",
61
+ "num_train_timesteps": 100,
62
+ "beta_schedule": "squaredcos_cap_v2",
63
+ "beta_start": 0.0001,
64
+ "beta_end": 0.02,
65
+ "prediction_type": "epsilon",
66
+ "clip_sample": true,
67
+ "clip_sample_range": 1.0,
68
+ "num_inference_steps": null,
69
+ "do_mask_loss_for_padding": false,
70
+ "optimizer_lr": 0.0001,
71
+ "optimizer_betas": [
72
+ 0.95,
73
+ 0.999
74
+ ],
75
+ "optimizer_eps": 1e-08,
76
+ "optimizer_weight_decay": 1e-06,
77
+ "scheduler_name": "cosine",
78
+ "scheduler_warmup_steps": 500
79
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27abd660da77675adfb0d4e1415ea27b7ea490e9b1284c1845a75d31833e5331
3
+ size 1051836880
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
+ 6
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:a3c8e29b9072e6106d74adc08b71025f27743113dfc79601b5cf2e8370d1cd49
3
+ size 3256
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.state": {
27
+ "type": "STATE",
28
+ "shape": [
29
+ 6
30
+ ]
31
+ },
32
+ "observation.images.top": {
33
+ "type": "VISUAL",
34
+ "shape": [
35
+ 3,
36
+ 240,
37
+ 320
38
+ ]
39
+ },
40
+ "action": {
41
+ "type": "ACTION",
42
+ "shape": [
43
+ 6
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:a3c8e29b9072e6106d74adc08b71025f27743113dfc79601b5cf2e8370d1cd49
3
+ size 3256
train_config.json ADDED
@@ -0,0 +1,206 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "/home/yuxianli/.cache/huggingface/lerobot/yuxianli/pusht_real_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
+ 6
91
+ ]
92
+ },
93
+ "observation.images.top": {
94
+ "type": "VISUAL",
95
+ "shape": [
96
+ 3,
97
+ 240,
98
+ 320
99
+ ]
100
+ }
101
+ },
102
+ "output_features": {
103
+ "action": {
104
+ "type": "ACTION",
105
+ "shape": [
106
+ 6
107
+ ]
108
+ }
109
+ },
110
+ "device": "cuda:0",
111
+ "use_amp": false,
112
+ "push_to_hub": false,
113
+ "repo_id": "yuxianli/pusht_diffusion_v5",
114
+ "private": null,
115
+ "tags": null,
116
+ "license": null,
117
+ "pretrained_path": null,
118
+ "horizon": 16,
119
+ "n_action_steps": 8,
120
+ "normalization_mapping": {
121
+ "VISUAL": "MEAN_STD",
122
+ "STATE": "MIN_MAX",
123
+ "ACTION": "MIN_MAX"
124
+ },
125
+ "drop_n_last_frames": 7,
126
+ "vision_backbone": "resnet18",
127
+ "crop_shape": null,
128
+ "crop_is_random": true,
129
+ "pretrained_backbone_weights": null,
130
+ "use_group_norm": true,
131
+ "spatial_softmax_num_keypoints": 32,
132
+ "use_separate_rgb_encoder_per_camera": false,
133
+ "down_dims": [
134
+ 512,
135
+ 1024,
136
+ 2048
137
+ ],
138
+ "kernel_size": 5,
139
+ "n_groups": 8,
140
+ "diffusion_step_embed_dim": 128,
141
+ "use_film_scale_modulation": true,
142
+ "noise_scheduler_type": "DDPM",
143
+ "num_train_timesteps": 100,
144
+ "beta_schedule": "squaredcos_cap_v2",
145
+ "beta_start": 0.0001,
146
+ "beta_end": 0.02,
147
+ "prediction_type": "epsilon",
148
+ "clip_sample": true,
149
+ "clip_sample_range": 1.0,
150
+ "num_inference_steps": null,
151
+ "do_mask_loss_for_padding": false,
152
+ "optimizer_lr": 0.0001,
153
+ "optimizer_betas": [
154
+ 0.95,
155
+ 0.999
156
+ ],
157
+ "optimizer_eps": 1e-08,
158
+ "optimizer_weight_decay": 1e-06,
159
+ "scheduler_name": "cosine",
160
+ "scheduler_warmup_steps": 500
161
+ },
162
+ "output_dir": "outputs/pusht_diffusion_v5",
163
+ "job_name": "diffusion",
164
+ "resume": false,
165
+ "seed": 1000,
166
+ "num_workers": 4,
167
+ "batch_size": 32,
168
+ "steps": 50000,
169
+ "eval_freq": 20000,
170
+ "log_freq": 200,
171
+ "save_checkpoint": true,
172
+ "save_freq": 10000,
173
+ "use_policy_training_preset": true,
174
+ "optimizer": {
175
+ "type": "adam",
176
+ "lr": 0.0001,
177
+ "weight_decay": 1e-06,
178
+ "grad_clip_norm": 10.0,
179
+ "betas": [
180
+ 0.95,
181
+ 0.999
182
+ ],
183
+ "eps": 1e-08
184
+ },
185
+ "scheduler": {
186
+ "type": "diffuser",
187
+ "num_warmup_steps": 500,
188
+ "name": "cosine"
189
+ },
190
+ "eval": {
191
+ "n_episodes": 50,
192
+ "batch_size": 50,
193
+ "use_async_envs": false
194
+ },
195
+ "wandb": {
196
+ "enable": true,
197
+ "disable_artifact": false,
198
+ "project": "pusht_fyp",
199
+ "entity": null,
200
+ "notes": null,
201
+ "run_id": "5181okfa",
202
+ "mode": null
203
+ },
204
+ "checkpoint_path": null,
205
+ "rename_map": {}
206
+ }