masondx commited on
Commit
8ed3843
·
verified ·
1 Parent(s): 6250a0c

Upload 10k EMA checkpoint

Browse files
config.json ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "diffusion",
3
+ "n_obs_steps": 2,
4
+ "input_features": {
5
+ "observation.images.top": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 480,
10
+ 640
11
+ ]
12
+ },
13
+ "observation.state": {
14
+ "type": "STATE",
15
+ "shape": [
16
+ 14
17
+ ]
18
+ }
19
+ },
20
+ "output_features": {
21
+ "action": {
22
+ "type": "ACTION",
23
+ "shape": [
24
+ 14
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
+ "pretrained_revision": null,
38
+ "horizon": 64,
39
+ "n_action_steps": 32,
40
+ "normalization_mapping": {
41
+ "VISUAL": "MEAN_STD",
42
+ "STATE": "MIN_MAX",
43
+ "ACTION": "MIN_MAX"
44
+ },
45
+ "drop_n_last_frames": 7,
46
+ "vision_backbone": "resnet18",
47
+ "resize_shape": null,
48
+ "crop_ratio": 1.0,
49
+ "crop_shape": null,
50
+ "crop_is_random": true,
51
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
52
+ "use_group_norm": false,
53
+ "spatial_softmax_num_keypoints": 32,
54
+ "use_separate_rgb_encoder_per_camera": true,
55
+ "down_dims": [
56
+ 512,
57
+ 1024,
58
+ 2048
59
+ ],
60
+ "kernel_size": 5,
61
+ "n_groups": 8,
62
+ "diffusion_step_embed_dim": 128,
63
+ "use_film_scale_modulation": true,
64
+ "gradient_checkpointing": false,
65
+ "noise_scheduler_type": "DDPM",
66
+ "num_train_timesteps": 100,
67
+ "beta_schedule": "squaredcos_cap_v2",
68
+ "beta_start": 0.0001,
69
+ "beta_end": 0.02,
70
+ "prediction_type": "epsilon",
71
+ "clip_sample": true,
72
+ "clip_sample_range": 1.0,
73
+ "num_inference_steps": null,
74
+ "compile_model": false,
75
+ "compile_mode": "reduce-overhead",
76
+ "do_mask_loss_for_padding": false,
77
+ "optimizer_lr": 0.0001,
78
+ "optimizer_betas": [
79
+ 0.95,
80
+ 0.999
81
+ ],
82
+ "optimizer_eps": 1e-08,
83
+ "optimizer_weight_decay": 1e-06,
84
+ "scheduler_name": "cosine",
85
+ "scheduler_warmup_steps": 500
86
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f687dd0dfb066afa104ed9a9f96f2f8df92c6cae141a32e1904c4a1b5b360c7
3
+ size 1053834144
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
+ 14
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 sha256:797777e6e870b9c70a950b6860aa23d246ee8292a6f10f4bca727ab564e39348
3
+ size 3892
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": {
27
+ "type": "VISUAL",
28
+ "shape": [
29
+ 3,
30
+ 480,
31
+ 640
32
+ ]
33
+ },
34
+ "observation.state": {
35
+ "type": "STATE",
36
+ "shape": [
37
+ 14
38
+ ]
39
+ },
40
+ "action": {
41
+ "type": "ACTION",
42
+ "shape": [
43
+ 14
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:ecd197869abe8329dde6c06838e2c1d5102014e2c107a905f057e678c75651cb
3
+ size 3900
train_config.json ADDED
@@ -0,0 +1,287 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "lerobot/aloha_sim_transfer_cube_human",
4
+ "repo_type": "dataset",
5
+ "root": null,
6
+ "episodes": null,
7
+ "exclude_episodes": null,
8
+ "image_transforms": {
9
+ "enable": false,
10
+ "max_num_transforms": 3,
11
+ "random_order": false,
12
+ "tfs": {
13
+ "brightness": {
14
+ "weight": 1.0,
15
+ "type": "ColorJitter",
16
+ "kwargs": {
17
+ "brightness": [
18
+ 0.8,
19
+ 1.2
20
+ ]
21
+ }
22
+ },
23
+ "contrast": {
24
+ "weight": 1.0,
25
+ "type": "ColorJitter",
26
+ "kwargs": {
27
+ "contrast": [
28
+ 0.8,
29
+ 1.2
30
+ ]
31
+ }
32
+ },
33
+ "saturation": {
34
+ "weight": 1.0,
35
+ "type": "ColorJitter",
36
+ "kwargs": {
37
+ "saturation": [
38
+ 0.5,
39
+ 1.5
40
+ ]
41
+ }
42
+ },
43
+ "hue": {
44
+ "weight": 1.0,
45
+ "type": "ColorJitter",
46
+ "kwargs": {
47
+ "hue": [
48
+ -0.05,
49
+ 0.05
50
+ ]
51
+ }
52
+ },
53
+ "sharpness": {
54
+ "weight": 1.0,
55
+ "type": "SharpnessJitter",
56
+ "kwargs": {
57
+ "sharpness": [
58
+ 0.5,
59
+ 1.5
60
+ ]
61
+ }
62
+ },
63
+ "affine": {
64
+ "weight": 1.0,
65
+ "type": "RandomAffine",
66
+ "kwargs": {
67
+ "degrees": [
68
+ -5.0,
69
+ 5.0
70
+ ],
71
+ "translate": [
72
+ 0.05,
73
+ 0.05
74
+ ]
75
+ }
76
+ }
77
+ }
78
+ },
79
+ "revision": null,
80
+ "use_imagenet_stats": true,
81
+ "video_backend": "pyav",
82
+ "return_uint8": false,
83
+ "depth_output_unit": "mm",
84
+ "streaming": false,
85
+ "eval_split": 0.0
86
+ },
87
+ "env": null,
88
+ "policy": {
89
+ "type": "diffusion",
90
+ "n_obs_steps": 2,
91
+ "input_features": {
92
+ "observation.images.top": {
93
+ "type": "VISUAL",
94
+ "shape": [
95
+ 3,
96
+ 480,
97
+ 640
98
+ ]
99
+ },
100
+ "observation.state": {
101
+ "type": "STATE",
102
+ "shape": [
103
+ 14
104
+ ]
105
+ }
106
+ },
107
+ "output_features": {
108
+ "action": {
109
+ "type": "ACTION",
110
+ "shape": [
111
+ 14
112
+ ]
113
+ }
114
+ },
115
+ "device": "cuda",
116
+ "use_amp": false,
117
+ "use_peft": false,
118
+ "push_to_hub": false,
119
+ "repo_id": null,
120
+ "private": null,
121
+ "tags": null,
122
+ "license": null,
123
+ "pretrained_path": null,
124
+ "pretrained_revision": null,
125
+ "horizon": 64,
126
+ "n_action_steps": 32,
127
+ "normalization_mapping": {
128
+ "VISUAL": "MEAN_STD",
129
+ "STATE": "MIN_MAX",
130
+ "ACTION": "MIN_MAX"
131
+ },
132
+ "drop_n_last_frames": 7,
133
+ "vision_backbone": "resnet18",
134
+ "resize_shape": null,
135
+ "crop_ratio": 1.0,
136
+ "crop_shape": null,
137
+ "crop_is_random": true,
138
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
139
+ "use_group_norm": false,
140
+ "spatial_softmax_num_keypoints": 32,
141
+ "use_separate_rgb_encoder_per_camera": true,
142
+ "down_dims": [
143
+ 512,
144
+ 1024,
145
+ 2048
146
+ ],
147
+ "kernel_size": 5,
148
+ "n_groups": 8,
149
+ "diffusion_step_embed_dim": 128,
150
+ "use_film_scale_modulation": true,
151
+ "gradient_checkpointing": false,
152
+ "noise_scheduler_type": "DDPM",
153
+ "num_train_timesteps": 100,
154
+ "beta_schedule": "squaredcos_cap_v2",
155
+ "beta_start": 0.0001,
156
+ "beta_end": 0.02,
157
+ "prediction_type": "epsilon",
158
+ "clip_sample": true,
159
+ "clip_sample_range": 1.0,
160
+ "num_inference_steps": null,
161
+ "compile_model": false,
162
+ "compile_mode": "reduce-overhead",
163
+ "do_mask_loss_for_padding": false,
164
+ "optimizer_lr": 0.0001,
165
+ "optimizer_betas": [
166
+ 0.95,
167
+ 0.999
168
+ ],
169
+ "optimizer_eps": 1e-08,
170
+ "optimizer_weight_decay": 1e-06,
171
+ "scheduler_name": "cosine",
172
+ "scheduler_warmup_steps": 500
173
+ },
174
+ "reward_model": null,
175
+ "output_dir": "outputs/train/diffusion_aloha_transfer_cube_10k",
176
+ "job_name": "diffusion_aloha_transfer_cube_10k",
177
+ "resume": false,
178
+ "seed": 1000,
179
+ "cudnn_deterministic": false,
180
+ "num_workers": 8,
181
+ "batch_size": 8,
182
+ "prefetch_factor": 4,
183
+ "persistent_workers": true,
184
+ "dataloader_multiprocessing_context": "spawn",
185
+ "steps": 10000,
186
+ "env_eval_freq": 0,
187
+ "log_freq": 200,
188
+ "eval_steps": 0,
189
+ "max_eval_samples": 0,
190
+ "tolerance_s": 0.0001,
191
+ "save_checkpoint": true,
192
+ "save_freq": 10000,
193
+ "checkpoint_format": "safetensors",
194
+ "use_policy_training_preset": true,
195
+ "optimizer": {
196
+ "type": "adam",
197
+ "lr": 0.0001,
198
+ "weight_decay": 1e-06,
199
+ "grad_clip_norm": 10.0,
200
+ "betas": [
201
+ 0.95,
202
+ 0.999
203
+ ],
204
+ "eps": 1e-08
205
+ },
206
+ "scheduler": {
207
+ "type": "diffuser",
208
+ "num_warmup_steps": 500,
209
+ "name": "cosine"
210
+ },
211
+ "parallelism": {
212
+ "dp_replicate": 1,
213
+ "dp_shard": 1,
214
+ "context_parallel": {
215
+ "ring_degree": 1,
216
+ "ulysses_degree": 1
217
+ },
218
+ "cfg_parallel": 1
219
+ },
220
+ "accelerator": {
221
+ "mixed_precision": "no",
222
+ "gradient_accumulation": {
223
+ "steps": 1
224
+ },
225
+ "fsdp": {
226
+ "reshard_after_forward": true,
227
+ "wrap_modules": null,
228
+ "min_num_params": null,
229
+ "cpu_offload": false,
230
+ "ignored_modules": null
231
+ },
232
+ "ddp": {
233
+ "find_unused_parameters": true,
234
+ "gradient_as_bucket_view": false,
235
+ "static_graph": false
236
+ },
237
+ "compile": {
238
+ "enabled": false,
239
+ "backend": "inductor",
240
+ "mode": null,
241
+ "regional": true
242
+ },
243
+ "activation_checkpointing": {
244
+ "mode": "none"
245
+ }
246
+ },
247
+ "eval": {
248
+ "n_episodes": 50,
249
+ "batch_size": 22,
250
+ "use_async_envs": true,
251
+ "recording": false,
252
+ "recording_repo_id": null,
253
+ "recording_private": false
254
+ },
255
+ "ema": {
256
+ "enable": true,
257
+ "decay": null,
258
+ "update_after_step": 0,
259
+ "inv_gamma": 1.0,
260
+ "power": 0.75,
261
+ "min_decay": 0.0,
262
+ "max_decay": 0.9999,
263
+ "use_for_eval": true
264
+ },
265
+ "wandb": {
266
+ "enable": false,
267
+ "disable_artifact": false,
268
+ "project": "lerobot",
269
+ "entity": null,
270
+ "notes": null,
271
+ "run_id": null,
272
+ "mode": null,
273
+ "add_tags": true
274
+ },
275
+ "peft": null,
276
+ "job": {
277
+ "target": null,
278
+ "image": "huggingface/lerobot-gpu:latest",
279
+ "timeout": "2d",
280
+ "detach": false,
281
+ "tags": []
282
+ },
283
+ "save_checkpoint_to_hub": false,
284
+ "sample_weighting": null,
285
+ "rename_map": {},
286
+ "checkpoint_path": null
287
+ }