Hugging Face's logo

LiamFy
/
diffusion_policy_large

Safetensors
Model card Files
xet
Community
lachenbach commited on
Commit
a22c73d
·
1 Parent(s): 51eecbb

019000

Browse files
Files changed (8) hide show
  1. 019000/pretrained_model/config.json +92 -0
  2. 019000/pretrained_model/model.safetensors +3 -0
  3. 019000/pretrained_model/train_config.json +202 -0
  4. 019000/training_state/optimizer_param_groups.json +232 -0
  5. 019000/training_state/optimizer_state.safetensors +3 -0
  6. 019000/training_state/rng_state.safetensors +3 -0
  7. 019000/training_state/scheduler_state.json +15 -0
  8. 019000/training_state/training_step.json +3 -0
019000/pretrained_model/config.json ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "diffusion",
3
+ "n_obs_steps": 2,
4
+ "normalization_mapping": {
5
+ "VISUAL": "MEAN_STD",
6
+ "STATE": "MIN_MAX",
7
+ "ACTION": "MIN_MAX"
8
+ },
9
+ "input_features": {
10
+ "observation.state": {
11
+ "type": "STATE",
12
+ "shape": [
13
+ 25
14
+ ]
15
+ },
16
+ "observation.images.main": {
17
+ "type": "VISUAL",
18
+ "shape": [
19
+ 3,
20
+ 224,
21
+ 224
22
+ ]
23
+ },
24
+ "observation.images.secondary_0": {
25
+ "type": "VISUAL",
26
+ "shape": [
27
+ 3,
28
+ 224,
29
+ 224
30
+ ]
31
+ },
32
+ "observation.images.secondary_1": {
33
+ "type": "VISUAL",
34
+ "shape": [
35
+ 3,
36
+ 224,
37
+ 224
38
+ ]
39
+ }
40
+ },
41
+ "output_features": {
42
+ "action": {
43
+ "type": "ACTION",
44
+ "shape": [
45
+ 25
46
+ ]
47
+ }
48
+ },
49
+ "device": "cuda",
50
+ "use_amp": false,
51
+ "horizon": 16,
52
+ "n_action_steps": 8,
53
+ "drop_n_last_frames": 7,
54
+ "vision_backbone": "resnet18",
55
+ "crop_shape": [
56
+ 74,
57
+ 74
58
+ ],
59
+ "crop_is_random": true,
60
+ "pretrained_backbone_weights": null,
61
+ "use_group_norm": true,
62
+ "spatial_softmax_num_keypoints": 32,
63
+ "use_separate_rgb_encoder_per_camera": false,
64
+ "down_dims": [
65
+ 512,
66
+ 1024,
67
+ 2048
68
+ ],
69
+ "kernel_size": 5,
70
+ "n_groups": 8,
71
+ "diffusion_step_embed_dim": 128,
72
+ "use_film_scale_modulation": true,
73
+ "noise_scheduler_type": "DDPM",
74
+ "num_train_timesteps": 100,
75
+ "beta_schedule": "squaredcos_cap_v2",
76
+ "beta_start": 0.0001,
77
+ "beta_end": 0.02,
78
+ "prediction_type": "epsilon",
79
+ "clip_sample": true,
80
+ "clip_sample_range": 1.0,
81
+ "num_inference_steps": 18,
82
+ "do_mask_loss_for_padding": false,
83
+ "optimizer_lr": 5e-05,
84
+ "optimizer_betas": [
85
+ 0.95,
86
+ 0.999
87
+ ],
88
+ "optimizer_eps": 1e-08,
89
+ "optimizer_weight_decay": 1e-06,
90
+ "scheduler_name": "cosine",
91
+ "scheduler_warmup_steps": 500
92
+ }
019000/pretrained_model/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cbd7f41a877b87f7be7fc5d96cdaa86cab3a3bd7577b8c2bfa0d401d21f35ef
3
+ size 1085829172
019000/pretrained_model/train_config.json ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "chengkunli/green_cup_pour",
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
+ }
62
+ },
63
+ "revision": null,
64
+ "use_imagenet_stats": true,
65
+ "video_backend": "torchcodec"
66
+ },
67
+ "env": null,
68
+ "policy": {
69
+ "type": "diffusion",
70
+ "n_obs_steps": 2,
71
+ "normalization_mapping": {
72
+ "VISUAL": "MEAN_STD",
73
+ "STATE": "MIN_MAX",
74
+ "ACTION": "MIN_MAX"
75
+ },
76
+ "input_features": {
77
+ "observation.state": {
78
+ "type": "STATE",
79
+ "shape": [
80
+ 25
81
+ ]
82
+ },
83
+ "observation.images.main": {
84
+ "type": "VISUAL",
85
+ "shape": [
86
+ 3,
87
+ 224,
88
+ 224
89
+ ]
90
+ },
91
+ "observation.images.secondary_0": {
92
+ "type": "VISUAL",
93
+ "shape": [
94
+ 3,
95
+ 224,
96
+ 224
97
+ ]
98
+ },
99
+ "observation.images.secondary_1": {
100
+ "type": "VISUAL",
101
+ "shape": [
102
+ 3,
103
+ 224,
104
+ 224
105
+ ]
106
+ }
107
+ },
108
+ "output_features": {
109
+ "action": {
110
+ "type": "ACTION",
111
+ "shape": [
112
+ 25
113
+ ]
114
+ }
115
+ },
116
+ "device": "cuda",
117
+ "use_amp": false,
118
+ "horizon": 16,
119
+ "n_action_steps": 8,
120
+ "drop_n_last_frames": 7,
121
+ "vision_backbone": "resnet18",
122
+ "crop_shape": [
123
+ 74,
124
+ 74
125
+ ],
126
+ "crop_is_random": true,
127
+ "pretrained_backbone_weights": null,
128
+ "use_group_norm": true,
129
+ "spatial_softmax_num_keypoints": 32,
130
+ "use_separate_rgb_encoder_per_camera": false,
131
+ "down_dims": [
132
+ 512,
133
+ 1024,
134
+ 2048
135
+ ],
136
+ "kernel_size": 5,
137
+ "n_groups": 8,
138
+ "diffusion_step_embed_dim": 128,
139
+ "use_film_scale_modulation": true,
140
+ "noise_scheduler_type": "DDPM",
141
+ "num_train_timesteps": 100,
142
+ "beta_schedule": "squaredcos_cap_v2",
143
+ "beta_start": 0.0001,
144
+ "beta_end": 0.02,
145
+ "prediction_type": "epsilon",
146
+ "clip_sample": true,
147
+ "clip_sample_range": 1.0,
148
+ "num_inference_steps": 18,
149
+ "do_mask_loss_for_padding": false,
150
+ "optimizer_lr": 5e-05,
151
+ "optimizer_betas": [
152
+ 0.95,
153
+ 0.999
154
+ ],
155
+ "optimizer_eps": 1e-08,
156
+ "optimizer_weight_decay": 1e-06,
157
+ "scheduler_name": "cosine",
158
+ "scheduler_warmup_steps": 500
159
+ },
160
+ "output_dir": "green_cup_pour_big",
161
+ "job_name": "diffusion",
162
+ "resume": false,
163
+ "seed": 1000,
164
+ "num_workers": 4,
165
+ "batch_size": 32,
166
+ "steps": 60000,
167
+ "eval_freq": 20000,
168
+ "log_freq": 100,
169
+ "save_checkpoint": true,
170
+ "save_freq": 1000,
171
+ "use_policy_training_preset": true,
172
+ "optimizer": {
173
+ "type": "adam",
174
+ "lr": 5e-05,
175
+ "weight_decay": 1e-06,
176
+ "grad_clip_norm": 10.0,
177
+ "betas": [
178
+ 0.95,
179
+ 0.999
180
+ ],
181
+ "eps": 1e-08
182
+ },
183
+ "scheduler": {
184
+ "type": "diffuser",
185
+ "num_warmup_steps": 500,
186
+ "name": "cosine"
187
+ },
188
+ "eval": {
189
+ "n_episodes": 50,
190
+ "batch_size": 50,
191
+ "use_async_envs": false
192
+ },
193
+ "wandb": {
194
+ "enable": true,
195
+ "disable_artifact": false,
196
+ "project": "diffusion_policy",
197
+ "entity": null,
198
+ "notes": null,
199
+ "run_id": null,
200
+ "mode": null
201
+ }
202
+ }
019000/training_state/optimizer_param_groups.json ADDED
@@ -0,0 +1,232 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "lr": 3.89919793696077e-05,
4
+ "betas": [
5
+ 0.95,
6
+ 0.999
7
+ ],
8
+ "eps": 1e-08,
9
+ "weight_decay": 1e-06,
10
+ "amsgrad": false,
11
+ "maximize": false,
12
+ "foreach": null,
13
+ "capturable": false,
14
+ "differentiable": false,
15
+ "fused": null,
16
+ "initial_lr": 5e-05,
17
+ "params": [
18
+ 0,
19
+ 1,
20
+ 2,
21
+ 3,
22
+ 4,
23
+ 5,
24
+ 6,
25
+ 7,
26
+ 8,
27
+ 9,
28
+ 10,
29
+ 11,
30
+ 12,
31
+ 13,
32
+ 14,
33
+ 15,
34
+ 16,
35
+ 17,
36
+ 18,
37
+ 19,
38
+ 20,
39
+ 21,
40
+ 22,
41
+ 23,
42
+ 24,
43
+ 25,
44
+ 26,
45
+ 27,
46
+ 28,
47
+ 29,
48
+ 30,
49
+ 31,
50
+ 32,
51
+ 33,
52
+ 34,
53
+ 35,
54
+ 36,
55
+ 37,
56
+ 38,
57
+ 39,
58
+ 40,
59
+ 41,
60
+ 42,
61
+ 43,
62
+ 44,
63
+ 45,
64
+ 46,
65
+ 47,
66
+ 48,
67
+ 49,
68
+ 50,
69
+ 51,
70
+ 52,
71
+ 53,
72
+ 54,
73
+ 55,
74
+ 56,
75
+ 57,
76
+ 58,
77
+ 59,
78
+ 60,
79
+ 61,
80
+ 62,
81
+ 63,
82
+ 64,
83
+ 65,
84
+ 66,
85
+ 67,
86
+ 68,
87
+ 69,
88
+ 70,
89
+ 71,
90
+ 72,
91
+ 73,
92
+ 74,
93
+ 75,
94
+ 76,
95
+ 77,
96
+ 78,
97
+ 79,
98
+ 80,
99
+ 81,
100
+ 82,
101
+ 83,
102
+ 84,
103
+ 85,
104
+ 86,
105
+ 87,
106
+ 88,
107
+ 89,
108
+ 90,
109
+ 91,
110
+ 92,
111
+ 93,
112
+ 94,
113
+ 95,
114
+ 96,
115
+ 97,
116
+ 98,
117
+ 99,
118
+ 100,
119
+ 101,
120
+ 102,
121
+ 103,
122
+ 104,
123
+ 105,
124
+ 106,
125
+ 107,
126
+ 108,
127
+ 109,
128
+ 110,
129
+ 111,
130
+ 112,
131
+ 113,
132
+ 114,
133
+ 115,
134
+ 116,
135
+ 117,
136
+ 118,
137
+ 119,
138
+ 120,
139
+ 121,
140
+ 122,
141
+ 123,
142
+ 124,
143
+ 125,
144
+ 126,
145
+ 127,
146
+ 128,
147
+ 129,
148
+ 130,
149
+ 131,
150
+ 132,
151
+ 133,
152
+ 134,
153
+ 135,
154
+ 136,
155
+ 137,
156
+ 138,
157
+ 139,
158
+ 140,
159
+ 141,
160
+ 142,
161
+ 143,
162
+ 144,
163
+ 145,
164
+ 146,
165
+ 147,
166
+ 148,
167
+ 149,
168
+ 150,
169
+ 151,
170
+ 152,
171
+ 153,
172
+ 154,
173
+ 155,
174
+ 156,
175
+ 157,
176
+ 158,
177
+ 159,
178
+ 160,
179
+ 161,
180
+ 162,
181
+ 163,
182
+ 164,
183
+ 165,
184
+ 166,
185
+ 167,
186
+ 168,
187
+ 169,
188
+ 170,
189
+ 171,
190
+ 172,
191
+ 173,
192
+ 174,
193
+ 175,
194
+ 176,
195
+ 177,
196
+ 178,
197
+ 179,
198
+ 180,
199
+ 181,
200
+ 182,
201
+ 183,
202
+ 184,
203
+ 185,
204
+ 186,
205
+ 187,
206
+ 188,
207
+ 189,
208
+ 190,
209
+ 191,
210
+ 192,
211
+ 193,
212
+ 194,
213
+ 195,
214
+ 196,
215
+ 197,
216
+ 198,
217
+ 199,
218
+ 200,
219
+ 201,
220
+ 202,
221
+ 203,
222
+ 204,
223
+ 205,
224
+ 206,
225
+ 207,
226
+ 208,
227
+ 209,
228
+ 210,
229
+ 211
230
+ ]
231
+ }
232
+ ]
019000/training_state/optimizer_state.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ce407afcaf57970c51f9398ef33887750e5b8de6dc63eb47f8aa34d18e65bcf
3
+ size 2171660424
019000/training_state/rng_state.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2f422ec0a620956996c6ff7bf1cf1c169d75f4871841304257fa4b78acba8ac
3
+ size 15708
019000/training_state/scheduler_state.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_lrs": [
3
+ 5e-05
4
+ ],
5
+ "last_epoch": 19000,
6
+ "verbose": false,
7
+ "_step_count": 19001,
8
+ "_get_lr_called_within_step": false,
9
+ "_last_lr": [
10
+ 3.89919793696077e-05
11
+ ],
12
+ "lr_lambdas": [
13
+ null
14
+ ]
15
+ }
019000/training_state/training_step.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "step": 19000
3
+ }