Hugging Face's logo

Jiafei1224
/
coffee-making

Safetensors
Model card Files
xet
 Community
Jiafei1224 commited on
Commit
e4be11a
·
verified ·
1 Parent(s): 3f88393

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +92 -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 +87 -0
  6. policy_preprocessor_step_2_normalizer_processor.safetensors +3 -0
  7. train_config.json +224 -0
config.json ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "pi05",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 14
9
+ ]
10
+ },
11
+ "observation.images.camera_front": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 360,
16
+ 640
17
+ ]
18
+ },
19
+ "observation.images.camera_left": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 360,
24
+ 640
25
+ ]
26
+ },
27
+ "observation.images.camera_right": {
28
+ "type": "VISUAL",
29
+ "shape": [
30
+ 3,
31
+ 360,
32
+ 640
33
+ ]
34
+ }
35
+ },
36
+ "output_features": {
37
+ "action": {
38
+ "type": "ACTION",
39
+ "shape": [
40
+ 14
41
+ ]
42
+ }
43
+ },
44
+ "device": "cuda",
45
+ "use_amp": false,
46
+ "push_to_hub": true,
47
+ "repo_id": "Jiafei1224/prepare_the_coffee_0323_posttrain",
48
+ "private": null,
49
+ "tags": null,
50
+ "license": null,
51
+ "pretrained_path": "lerobot/pi05_base",
52
+ "paligemma_variant": "gemma_2b",
53
+ "action_expert_variant": "gemma_300m",
54
+ "dtype": "bfloat16",
55
+ "chunk_size": 50,
56
+ "n_action_steps": 50,
57
+ "max_state_dim": 32,
58
+ "max_action_dim": 32,
59
+ "num_inference_steps": 10,
60
+ "time_sampling_beta_alpha": 1.5,
61
+ "time_sampling_beta_beta": 1.0,
62
+ "time_sampling_scale": 0.999,
63
+ "time_sampling_offset": 0.001,
64
+ "min_period": 0.004,
65
+ "max_period": 4.0,
66
+ "rtc_config": null,
67
+ "image_resolution": [
68
+ 224,
69
+ 224
70
+ ],
71
+ "empty_cameras": 0,
72
+ "tokenizer_max_length": 200,
73
+ "normalization_mapping": {
74
+ "VISUAL": "IDENTITY",
75
+ "STATE": "QUANTILES",
76
+ "ACTION": "QUANTILES"
77
+ },
78
+ "gradient_checkpointing": true,
79
+ "compile_model": false,
80
+ "compile_mode": "max-autotune",
81
+ "optimizer_lr": 2.5e-05,
82
+ "optimizer_betas": [
83
+ 0.9,
84
+ 0.95
85
+ ],
86
+ "optimizer_eps": 1e-08,
87
+ "optimizer_weight_decay": 0.01,
88
+ "optimizer_grad_clip_norm": 1.0,
89
+ "scheduler_warmup_steps": 1000,
90
+ "scheduler_decay_steps": 30000,
91
+ "scheduler_decay_lr": 2.5e-06
92
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da83546a21378d776914774d761022d2f4a520da0ff1e21238c8e14e1bdc86e9
3
+ size 7473096344
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": "IDENTITY",
18
+ "STATE": "QUANTILES",
19
+ "ACTION": "QUANTILES"
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 sha256:115b6a50e9fc8f9ced3eb0f0a60ae51a17e797ca5b653d9a780def45e98c4083
3
+ size 9400
policy_preprocessor.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "normalizer_processor",
16
+ "config": {
17
+ "eps": 1e-08,
18
+ "features": {
19
+ "observation.state": {
20
+ "type": "STATE",
21
+ "shape": [
22
+ 14
23
+ ]
24
+ },
25
+ "observation.images.camera_front": {
26
+ "type": "VISUAL",
27
+ "shape": [
28
+ 3,
29
+ 360,
30
+ 640
31
+ ]
32
+ },
33
+ "observation.images.camera_left": {
34
+ "type": "VISUAL",
35
+ "shape": [
36
+ 3,
37
+ 360,
38
+ 640
39
+ ]
40
+ },
41
+ "observation.images.camera_right": {
42
+ "type": "VISUAL",
43
+ "shape": [
44
+ 3,
45
+ 360,
46
+ 640
47
+ ]
48
+ },
49
+ "action": {
50
+ "type": "ACTION",
51
+ "shape": [
52
+ 14
53
+ ]
54
+ }
55
+ },
56
+ "norm_map": {
57
+ "VISUAL": "IDENTITY",
58
+ "STATE": "QUANTILES",
59
+ "ACTION": "QUANTILES"
60
+ }
61
+ },
62
+ "state_file": "policy_preprocessor_step_2_normalizer_processor.safetensors"
63
+ },
64
+ {
65
+ "registry_name": "pi05_prepare_state_tokenizer_processor_step",
66
+ "config": {}
67
+ },
68
+ {
69
+ "registry_name": "tokenizer_processor",
70
+ "config": {
71
+ "max_length": 200,
72
+ "task_key": "task",
73
+ "padding_side": "right",
74
+ "padding": "max_length",
75
+ "truncation": true,
76
+ "tokenizer_name": "google/paligemma-3b-pt-224"
77
+ }
78
+ },
79
+ {
80
+ "registry_name": "device_processor",
81
+ "config": {
82
+ "device": "cuda",
83
+ "float_dtype": null
84
+ }
85
+ }
86
+ ]
87
+ }
policy_preprocessor_step_2_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:115b6a50e9fc8f9ced3eb0f0a60ae51a17e797ca5b653d9a780def45e98c4083
3
+ size 9400
train_config.json ADDED
@@ -0,0 +1,224 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": [
4
+ "williamtsai726/prepare_the_coffee_0323"
5
+ ],
6
+ "root": null,
7
+ "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": "torchcodec",
82
+ "streaming": false
83
+ },
84
+ "env": null,
85
+ "policy": {
86
+ "type": "pi05",
87
+ "n_obs_steps": 1,
88
+ "input_features": {
89
+ "observation.state": {
90
+ "type": "STATE",
91
+ "shape": [
92
+ 14
93
+ ]
94
+ },
95
+ "observation.images.camera_front": {
96
+ "type": "VISUAL",
97
+ "shape": [
98
+ 3,
99
+ 360,
100
+ 640
101
+ ]
102
+ },
103
+ "observation.images.camera_left": {
104
+ "type": "VISUAL",
105
+ "shape": [
106
+ 3,
107
+ 360,
108
+ 640
109
+ ]
110
+ },
111
+ "observation.images.camera_right": {
112
+ "type": "VISUAL",
113
+ "shape": [
114
+ 3,
115
+ 360,
116
+ 640
117
+ ]
118
+ }
119
+ },
120
+ "output_features": {
121
+ "action": {
122
+ "type": "ACTION",
123
+ "shape": [
124
+ 14
125
+ ]
126
+ }
127
+ },
128
+ "device": "cuda",
129
+ "use_amp": false,
130
+ "push_to_hub": true,
131
+ "repo_id": "Jiafei1224/prepare_the_coffee_0323_posttrain",
132
+ "private": null,
133
+ "tags": null,
134
+ "license": null,
135
+ "pretrained_path": "lerobot/pi05_base",
136
+ "paligemma_variant": "gemma_2b",
137
+ "action_expert_variant": "gemma_300m",
138
+ "dtype": "bfloat16",
139
+ "chunk_size": 50,
140
+ "n_action_steps": 50,
141
+ "max_state_dim": 32,
142
+ "max_action_dim": 32,
143
+ "num_inference_steps": 10,
144
+ "time_sampling_beta_alpha": 1.5,
145
+ "time_sampling_beta_beta": 1.0,
146
+ "time_sampling_scale": 0.999,
147
+ "time_sampling_offset": 0.001,
148
+ "min_period": 0.004,
149
+ "max_period": 4.0,
150
+ "rtc_config": null,
151
+ "image_resolution": [
152
+ 224,
153
+ 224
154
+ ],
155
+ "empty_cameras": 0,
156
+ "tokenizer_max_length": 200,
157
+ "normalization_mapping": {
158
+ "VISUAL": "IDENTITY",
159
+ "STATE": "QUANTILES",
160
+ "ACTION": "QUANTILES"
161
+ },
162
+ "gradient_checkpointing": true,
163
+ "compile_model": false,
164
+ "compile_mode": "max-autotune",
165
+ "optimizer_lr": 2.5e-05,
166
+ "optimizer_betas": [
167
+ 0.9,
168
+ 0.95
169
+ ],
170
+ "optimizer_eps": 1e-08,
171
+ "optimizer_weight_decay": 0.01,
172
+ "optimizer_grad_clip_norm": 1.0,
173
+ "scheduler_warmup_steps": 1000,
174
+ "scheduler_decay_steps": 30000,
175
+ "scheduler_decay_lr": 2.5e-06
176
+ },
177
+ "output_dir": "outputs/pi05_multi-steermix-Jiafei1224_pi-05-prepare_the_coffee_0323",
178
+ "job_name": "pi05_training",
179
+ "resume": false,
180
+ "seed": 1000,
181
+ "num_workers": 4,
182
+ "batch_size": 32,
183
+ "steps": 10000,
184
+ "eval_freq": 20000,
185
+ "log_freq": 200,
186
+ "tolerance_s": 0.0001,
187
+ "save_checkpoint": true,
188
+ "save_freq": 1000,
189
+ "use_policy_training_preset": true,
190
+ "optimizer": {
191
+ "type": "adamw",
192
+ "lr": 2.5e-05,
193
+ "weight_decay": 0.01,
194
+ "grad_clip_norm": 1.0,
195
+ "betas": [
196
+ 0.9,
197
+ 0.95
198
+ ],
199
+ "eps": 1e-08
200
+ },
201
+ "scheduler": {
202
+ "type": "cosine_decay_with_warmup",
203
+ "num_warmup_steps": 1000,
204
+ "num_decay_steps": 30000,
205
+ "peak_lr": 2.5e-05,
206
+ "decay_lr": 2.5e-06
207
+ },
208
+ "eval": {
209
+ "n_episodes": 50,
210
+ "batch_size": 50,
211
+ "use_async_envs": false
212
+ },
213
+ "wandb": {
214
+ "enable": false,
215
+ "disable_artifact": false,
216
+ "project": "GroundedVLA",
217
+ "entity": null,
218
+ "notes": null,
219
+ "run_id": null,
220
+ "mode": null
221
+ },
222
+ "checkpoint_path": null,
223
+ "rename_map": {}
224
+ }