kritokronos commited on
Commit
a135128
·
verified ·
1 Parent(s): b58b951

Upload Fast-WAM FC checkpoint at step 80000

Browse files
config.json ADDED
@@ -0,0 +1,143 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "fastwam",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.images.front": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 224,
10
+ 224
11
+ ]
12
+ },
13
+ "observation.images.overview": {
14
+ "type": "VISUAL",
15
+ "shape": [
16
+ 3,
17
+ 224,
18
+ 224
19
+ ]
20
+ },
21
+ "observation.images.wrist": {
22
+ "type": "VISUAL",
23
+ "shape": [
24
+ 3,
25
+ 224,
26
+ 224
27
+ ]
28
+ },
29
+ "observation.state": {
30
+ "type": "STATE",
31
+ "shape": [
32
+ 10
33
+ ]
34
+ }
35
+ },
36
+ "output_features": {
37
+ "action": {
38
+ "type": "ACTION",
39
+ "shape": [
40
+ 10
41
+ ]
42
+ }
43
+ },
44
+ "device": "cuda:0",
45
+ "use_amp": false,
46
+ "use_peft": false,
47
+ "push_to_hub": false,
48
+ "repo_id": null,
49
+ "private": null,
50
+ "tags": null,
51
+ "license": null,
52
+ "pretrained_path": "/projects/haitian003ssd/ICRA2027-baseline/fast-wam/train_outputs/fastwam_factory_conveyor_level2_110767/checkpoints/040000/pretrained_model",
53
+ "pretrained_revision": null,
54
+ "action_dim": 10,
55
+ "proprio_dim": 10,
56
+ "action_horizon": 32,
57
+ "n_action_steps": 10,
58
+ "num_video_frames": 33,
59
+ "action_video_freq_ratio": 4,
60
+ "image_size": [
61
+ 224,
62
+ 672
63
+ ],
64
+ "context_len": 128,
65
+ "model_id": "Wan-AI/Wan2.2-TI2V-5B",
66
+ "tokenizer_model_id": "google/umt5-xxl",
67
+ "text_encoder_model_id": "Wan-AI/Wan2.2-TI2V-5B-Diffusers",
68
+ "base_model_id": "lerobot/fastwam_base",
69
+ "tokenizer_max_len": 128,
70
+ "load_text_encoder": true,
71
+ "mot_checkpoint_mixed_attn": false,
72
+ "torch_dtype": "bfloat16",
73
+ "prompt_template": "A video recorded from a robot's point of view executing the following instruction: {task}",
74
+ "num_inference_steps": 10,
75
+ "inference_seed": 42,
76
+ "rand_device": "cpu",
77
+ "text_cfg_scale": 1.0,
78
+ "negative_prompt": "",
79
+ "sigma_shift": null,
80
+ "tiled": false,
81
+ "fp32_attention": true,
82
+ "use_gradient_checkpointing": true,
83
+ "freeze_video_expert": false,
84
+ "toggle_action_dimensions": [],
85
+ "video_scheduler": {
86
+ "train_shift": 5.0,
87
+ "infer_shift": 5.0,
88
+ "num_train_timesteps": 1000.0
89
+ },
90
+ "action_scheduler": {
91
+ "train_shift": 5.0,
92
+ "infer_shift": 5.0,
93
+ "num_train_timesteps": 1000.0
94
+ },
95
+ "loss": {
96
+ "lambda_video": 1.0,
97
+ "lambda_action": 1.0
98
+ },
99
+ "video_dit_config": {
100
+ "patch_size": [
101
+ 1,
102
+ 2,
103
+ 2
104
+ ],
105
+ "in_dim": 48,
106
+ "hidden_dim": 3072,
107
+ "ffn_dim": 14336,
108
+ "freq_dim": 256,
109
+ "text_dim": 4096,
110
+ "out_dim": 48,
111
+ "num_heads": 24,
112
+ "attn_head_dim": 128,
113
+ "num_layers": 30,
114
+ "eps": 1e-06,
115
+ "seperated_timestep": true,
116
+ "use_gradient_checkpointing": true,
117
+ "video_attention_mask_mode": "first_frame_causal",
118
+ "action_conditioned": false,
119
+ "action_dim": 10,
120
+ "action_group_causal_mask_mode": "group_diagonal",
121
+ "fp32_attention": true
122
+ },
123
+ "action_dit_config": {
124
+ "action_dim": 10,
125
+ "hidden_dim": 1024,
126
+ "ffn_dim": 4096,
127
+ "num_heads": 24,
128
+ "attn_head_dim": 128,
129
+ "num_layers": 30,
130
+ "text_dim": 4096,
131
+ "freq_dim": 256,
132
+ "eps": 1e-06,
133
+ "use_gradient_checkpointing": true,
134
+ "fp32_attention": true
135
+ },
136
+ "normalization_mapping": {
137
+ "VISUAL": "IDENTITY",
138
+ "STATE": "MEAN_STD",
139
+ "ACTION": "MEAN_STD"
140
+ },
141
+ "optimizer_lr": 0.0001,
142
+ "optimizer_weight_decay": 0.01
143
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76cf50e7aad070e39e912efdf5a2f902677de729ad3699d8017d860c02034351
3
+ size 24083107528
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
+ 10
13
+ ]
14
+ }
15
+ },
16
+ "norm_map": {
17
+ "VISUAL": "IDENTITY",
18
+ "STATE": "MEAN_STD",
19
+ "ACTION": "MEAN_STD"
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 sha256:2de25635c27e6b78dd81bfb3f1d6e6559500fdd89e48af69fae10955ee61243d
3
+ size 648
policy_preprocessor.json ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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.front": {
27
+ "type": "VISUAL",
28
+ "shape": [
29
+ 3,
30
+ 224,
31
+ 224
32
+ ]
33
+ },
34
+ "observation.images.overview": {
35
+ "type": "VISUAL",
36
+ "shape": [
37
+ 3,
38
+ 224,
39
+ 224
40
+ ]
41
+ },
42
+ "observation.images.wrist": {
43
+ "type": "VISUAL",
44
+ "shape": [
45
+ 3,
46
+ 224,
47
+ 224
48
+ ]
49
+ },
50
+ "observation.state": {
51
+ "type": "STATE",
52
+ "shape": [
53
+ 10
54
+ ]
55
+ },
56
+ "action": {
57
+ "type": "ACTION",
58
+ "shape": [
59
+ 10
60
+ ]
61
+ }
62
+ },
63
+ "norm_map": {
64
+ "VISUAL": "IDENTITY",
65
+ "STATE": "MEAN_STD",
66
+ "ACTION": "MEAN_STD"
67
+ }
68
+ },
69
+ "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors"
70
+ }
71
+ ]
72
+ }
policy_preprocessor_step_3_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf96d1ea1c280728d784bdbfeaa99ae615b32e7af2e66f10869bc54afb43916b
3
+ size 1968
train_config.json ADDED
@@ -0,0 +1,291 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "local/factory_conveyor_level2_seeded",
4
+ "root": "/projects/haitian003ssd/Dataset/factory_conveyor_level2_seeded",
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": "pyav",
80
+ "return_uint8": false,
81
+ "depth_output_unit": "mm",
82
+ "streaming": false,
83
+ "eval_split": 0.0
84
+ },
85
+ "env": null,
86
+ "policy": {
87
+ "type": "fastwam",
88
+ "n_obs_steps": 1,
89
+ "input_features": {
90
+ "observation.images.front": {
91
+ "type": "VISUAL",
92
+ "shape": [
93
+ 3,
94
+ 224,
95
+ 224
96
+ ]
97
+ },
98
+ "observation.images.overview": {
99
+ "type": "VISUAL",
100
+ "shape": [
101
+ 3,
102
+ 224,
103
+ 224
104
+ ]
105
+ },
106
+ "observation.images.wrist": {
107
+ "type": "VISUAL",
108
+ "shape": [
109
+ 3,
110
+ 224,
111
+ 224
112
+ ]
113
+ },
114
+ "observation.state": {
115
+ "type": "STATE",
116
+ "shape": [
117
+ 10
118
+ ]
119
+ }
120
+ },
121
+ "output_features": {
122
+ "action": {
123
+ "type": "ACTION",
124
+ "shape": [
125
+ 10
126
+ ]
127
+ }
128
+ },
129
+ "device": "cuda:0",
130
+ "use_amp": false,
131
+ "use_peft": false,
132
+ "push_to_hub": false,
133
+ "repo_id": null,
134
+ "private": null,
135
+ "tags": null,
136
+ "license": null,
137
+ "pretrained_path": "/projects/haitian003ssd/ICRA2027-baseline/fast-wam/train_outputs/fastwam_factory_conveyor_level2_110767/checkpoints/040000/pretrained_model",
138
+ "pretrained_revision": null,
139
+ "action_dim": 10,
140
+ "proprio_dim": 10,
141
+ "action_horizon": 32,
142
+ "n_action_steps": 10,
143
+ "num_video_frames": 33,
144
+ "action_video_freq_ratio": 4,
145
+ "image_size": [
146
+ 224,
147
+ 672
148
+ ],
149
+ "context_len": 128,
150
+ "model_id": "Wan-AI/Wan2.2-TI2V-5B",
151
+ "tokenizer_model_id": "google/umt5-xxl",
152
+ "text_encoder_model_id": "Wan-AI/Wan2.2-TI2V-5B-Diffusers",
153
+ "base_model_id": "lerobot/fastwam_base",
154
+ "tokenizer_max_len": 128,
155
+ "load_text_encoder": true,
156
+ "mot_checkpoint_mixed_attn": false,
157
+ "torch_dtype": "bfloat16",
158
+ "prompt_template": "A video recorded from a robot's point of view executing the following instruction: {task}",
159
+ "num_inference_steps": 10,
160
+ "inference_seed": 42,
161
+ "rand_device": "cpu",
162
+ "text_cfg_scale": 1.0,
163
+ "negative_prompt": "",
164
+ "sigma_shift": null,
165
+ "tiled": false,
166
+ "fp32_attention": true,
167
+ "use_gradient_checkpointing": true,
168
+ "freeze_video_expert": false,
169
+ "toggle_action_dimensions": [],
170
+ "video_scheduler": {
171
+ "train_shift": 5.0,
172
+ "infer_shift": 5.0,
173
+ "num_train_timesteps": 1000.0
174
+ },
175
+ "action_scheduler": {
176
+ "train_shift": 5.0,
177
+ "infer_shift": 5.0,
178
+ "num_train_timesteps": 1000.0
179
+ },
180
+ "loss": {
181
+ "lambda_video": 1.0,
182
+ "lambda_action": 1.0
183
+ },
184
+ "video_dit_config": {
185
+ "patch_size": [
186
+ 1,
187
+ 2,
188
+ 2
189
+ ],
190
+ "in_dim": 48,
191
+ "hidden_dim": 3072,
192
+ "ffn_dim": 14336,
193
+ "freq_dim": 256,
194
+ "text_dim": 4096,
195
+ "out_dim": 48,
196
+ "num_heads": 24,
197
+ "attn_head_dim": 128,
198
+ "num_layers": 30,
199
+ "eps": 1e-06,
200
+ "seperated_timestep": true,
201
+ "use_gradient_checkpointing": true,
202
+ "video_attention_mask_mode": "first_frame_causal",
203
+ "action_conditioned": false,
204
+ "action_dim": 10,
205
+ "action_group_causal_mask_mode": "group_diagonal",
206
+ "fp32_attention": true
207
+ },
208
+ "action_dit_config": {
209
+ "action_dim": 10,
210
+ "hidden_dim": 1024,
211
+ "ffn_dim": 4096,
212
+ "num_heads": 24,
213
+ "attn_head_dim": 128,
214
+ "num_layers": 30,
215
+ "text_dim": 4096,
216
+ "freq_dim": 256,
217
+ "eps": 1e-06,
218
+ "use_gradient_checkpointing": true,
219
+ "fp32_attention": true
220
+ },
221
+ "normalization_mapping": {
222
+ "VISUAL": "IDENTITY",
223
+ "STATE": "MEAN_STD",
224
+ "ACTION": "MEAN_STD"
225
+ },
226
+ "optimizer_lr": 0.0001,
227
+ "optimizer_weight_decay": 0.01
228
+ },
229
+ "reward_model": null,
230
+ "output_dir": "/projects/haitian003ssd/ICRA2027-baseline/fast-wam/train_outputs/fastwam_resume_80k_132525",
231
+ "job_name": "fastwam_factory_conveyor_level2",
232
+ "resume": true,
233
+ "seed": 1000,
234
+ "cudnn_deterministic": false,
235
+ "num_workers": 0,
236
+ "batch_size": 8,
237
+ "prefetch_factor": 4,
238
+ "persistent_workers": true,
239
+ "steps": 80000,
240
+ "env_eval_freq": 20000,
241
+ "log_freq": 50,
242
+ "eval_steps": 0,
243
+ "max_eval_samples": 0,
244
+ "tolerance_s": 0.0001,
245
+ "save_checkpoint": true,
246
+ "save_training_state": false,
247
+ "save_freq": 40000,
248
+ "use_policy_training_preset": true,
249
+ "optimizer": {
250
+ "type": "adamw",
251
+ "lr": 0.0001,
252
+ "weight_decay": 0.01,
253
+ "grad_clip_norm": 10.0,
254
+ "betas": [
255
+ 0.9,
256
+ 0.999
257
+ ],
258
+ "eps": 1e-08
259
+ },
260
+ "scheduler": null,
261
+ "eval": {
262
+ "n_episodes": 50,
263
+ "batch_size": 50,
264
+ "use_async_envs": true,
265
+ "recording": false,
266
+ "recording_repo_id": null,
267
+ "recording_private": false
268
+ },
269
+ "wandb": {
270
+ "enable": true,
271
+ "disable_artifact": false,
272
+ "project": "fastwam",
273
+ "entity": null,
274
+ "notes": null,
275
+ "run_id": "c27a20qd",
276
+ "mode": null,
277
+ "add_tags": true
278
+ },
279
+ "peft": null,
280
+ "job": {
281
+ "target": null,
282
+ "image": "huggingface/lerobot-gpu:latest",
283
+ "timeout": "2d",
284
+ "detach": false,
285
+ "tags": []
286
+ },
287
+ "save_checkpoint_to_hub": false,
288
+ "sample_weighting": null,
289
+ "rename_map": {},
290
+ "checkpoint_path": "/projects/haitian003ssd/ICRA2027-baseline/fast-wam/train_outputs/fastwam_factory_conveyor_level2_110767/checkpoints/040000"
291
+ }