Hugging Face's logo

wlinhugging
/
pushT_diffusion_DDIM

Safetensors
Model card Files
xet
 Community
wlinhugging commited on
Commit
974ea2a
·
verified ·
1 Parent(s): 4a8bc8a

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +84 -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 +220 -0
config.json ADDED
@@ -0,0 +1,84 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "diffusion",
3
+ "n_obs_steps": 2,
4
+ "input_features": {
5
+ "observation.image": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 96,
10
+ 96
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": true,
30
+ "use_peft": false,
31
+ "push_to_hub": true,
32
+ "repo_id": "wlinhugging/pushT_diffusion_policy_DDIM",
33
+ "private": null,
34
+ "tags": null,
35
+ "license": null,
36
+ "pretrained_path": null,
37
+ "horizon": 16,
38
+ "n_action_steps": 8,
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
+ "resize_shape": null,
47
+ "crop_ratio": 1.0,
48
+ "crop_shape": null,
49
+ "crop_is_random": true,
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": "DDIM",
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
+ "compile_model": false,
73
+ "compile_mode": "reduce-overhead",
74
+ "do_mask_loss_for_padding": false,
75
+ "optimizer_lr": 0.0001,
76
+ "optimizer_betas": [
77
+ 0.95,
78
+ 0.999
79
+ ],
80
+ "optimizer_eps": 1e-08,
81
+ "optimizer_weight_decay": 1e-06,
82
+ "scheduler_name": "cosine",
83
+ "scheduler_warmup_steps": 500
84
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0cb6bced003591476a56b0dd110bea74980047de40327d8b0f06d4de19fb192
3
+ size 1050861448
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:bc174a7436ff5075efebe7be7cece5e5d8430658cc9b5314fb8a652eab912d96
3
+ size 4204
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.image": {
27
+ "type": "VISUAL",
28
+ "shape": [
29
+ 3,
30
+ 96,
31
+ 96
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:bc174a7436ff5075efebe7be7cece5e5d8430658cc9b5314fb8a652eab912d96
3
+ size 4204
train_config.json ADDED
@@ -0,0 +1,220 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "lerobot/pusht",
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": null,
83
+ "policy": {
84
+ "type": "diffusion",
85
+ "n_obs_steps": 2,
86
+ "input_features": {
87
+ "observation.image": {
88
+ "type": "VISUAL",
89
+ "shape": [
90
+ 3,
91
+ 96,
92
+ 96
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": true,
112
+ "use_peft": false,
113
+ "push_to_hub": true,
114
+ "repo_id": "wlinhugging/pushT_diffusion_policy_DDIM",
115
+ "private": null,
116
+ "tags": null,
117
+ "license": null,
118
+ "pretrained_path": null,
119
+ "horizon": 16,
120
+ "n_action_steps": 8,
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
+ "resize_shape": null,
129
+ "crop_ratio": 1.0,
130
+ "crop_shape": null,
131
+ "crop_is_random": true,
132
+ "pretrained_backbone_weights": null,
133
+ "use_group_norm": true,
134
+ "spatial_softmax_num_keypoints": 32,
135
+ "use_separate_rgb_encoder_per_camera": false,
136
+ "down_dims": [
137
+ 512,
138
+ 1024,
139
+ 2048
140
+ ],
141
+ "kernel_size": 5,
142
+ "n_groups": 8,
143
+ "diffusion_step_embed_dim": 128,
144
+ "use_film_scale_modulation": true,
145
+ "noise_scheduler_type": "DDIM",
146
+ "num_train_timesteps": 100,
147
+ "beta_schedule": "squaredcos_cap_v2",
148
+ "beta_start": 0.0001,
149
+ "beta_end": 0.02,
150
+ "prediction_type": "epsilon",
151
+ "clip_sample": true,
152
+ "clip_sample_range": 1.0,
153
+ "num_inference_steps": null,
154
+ "compile_model": false,
155
+ "compile_mode": "reduce-overhead",
156
+ "do_mask_loss_for_padding": false,
157
+ "optimizer_lr": 0.0001,
158
+ "optimizer_betas": [
159
+ 0.95,
160
+ 0.999
161
+ ],
162
+ "optimizer_eps": 1e-08,
163
+ "optimizer_weight_decay": 1e-06,
164
+ "scheduler_name": "cosine",
165
+ "scheduler_warmup_steps": 500
166
+ },
167
+ "output_dir": "outputs/train/pushT_diffusion_DDIM",
168
+ "job_name": "pushT_diffusion",
169
+ "resume": false,
170
+ "seed": 1000,
171
+ "cudnn_deterministic": false,
172
+ "num_workers": 2,
173
+ "batch_size": 64,
174
+ "steps": 50000,
175
+ "eval_freq": 5000,
176
+ "log_freq": 200,
177
+ "tolerance_s": 0.0001,
178
+ "save_checkpoint": true,
179
+ "save_freq": 10000,
180
+ "use_policy_training_preset": true,
181
+ "optimizer": {
182
+ "type": "adam",
183
+ "lr": 0.0001,
184
+ "weight_decay": 1e-06,
185
+ "grad_clip_norm": 10.0,
186
+ "betas": [
187
+ 0.95,
188
+ 0.999
189
+ ],
190
+ "eps": 1e-08
191
+ },
192
+ "scheduler": {
193
+ "type": "diffuser",
194
+ "num_warmup_steps": 500,
195
+ "name": "cosine"
196
+ },
197
+ "eval": {
198
+ "n_episodes": 50,
199
+ "batch_size": 50,
200
+ "use_async_envs": false
201
+ },
202
+ "wandb": {
203
+ "enable": false,
204
+ "disable_artifact": false,
205
+ "project": "lerobot",
206
+ "entity": null,
207
+ "notes": null,
208
+ "run_id": null,
209
+ "mode": null,
210
+ "add_tags": true
211
+ },
212
+ "peft": null,
213
+ "use_rabc": false,
214
+ "rabc_progress_path": null,
215
+ "rabc_kappa": 0.01,
216
+ "rabc_epsilon": 1e-06,
217
+ "rabc_head_mode": "sparse",
218
+ "rename_map": {},
219
+ "checkpoint_path": null
220
+ }