Hugging Face's logo

thewisp
/
pi05_pick_place_earplug

Robotics
LeRobot
Safetensors
pi05
Model card Files
xet
Community
thewisp commited on
Commit
c1956f0
·
verified ·
1 Parent(s): a8973fe

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +83 -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 +79 -0
  6. policy_preprocessor_step_2_normalizer_processor.safetensors +3 -0
  7. train_config.json +196 -0
config.json ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "pi05",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 6
9
+ ]
10
+ },
11
+ "observation.images.top": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 480,
16
+ 640
17
+ ]
18
+ },
19
+ "observation.images.wrist": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 480,
24
+ 640
25
+ ]
26
+ }
27
+ },
28
+ "output_features": {
29
+ "action": {
30
+ "type": "ACTION",
31
+ "shape": [
32
+ 6
33
+ ]
34
+ }
35
+ },
36
+ "device": "cuda",
37
+ "use_amp": false,
38
+ "push_to_hub": true,
39
+ "repo_id": "thewisp/pi05_pick_place_earplug",
40
+ "private": null,
41
+ "tags": null,
42
+ "license": null,
43
+ "pretrained_path": "lerobot/pi05_base",
44
+ "paligemma_variant": "gemma_2b",
45
+ "action_expert_variant": "gemma_300m",
46
+ "dtype": "bfloat16",
47
+ "chunk_size": 50,
48
+ "n_action_steps": 50,
49
+ "max_state_dim": 32,
50
+ "max_action_dim": 32,
51
+ "num_inference_steps": 10,
52
+ "time_sampling_beta_alpha": 1.5,
53
+ "time_sampling_beta_beta": 1.0,
54
+ "time_sampling_scale": 0.999,
55
+ "time_sampling_offset": 0.001,
56
+ "min_period": 0.004,
57
+ "max_period": 4.0,
58
+ "image_resolution": [
59
+ 224,
60
+ 224
61
+ ],
62
+ "empty_cameras": 0,
63
+ "tokenizer_max_length": 200,
64
+ "normalization_mapping": {
65
+ "VISUAL": "IDENTITY",
66
+ "STATE": "QUANTILES",
67
+ "ACTION": "QUANTILES"
68
+ },
69
+ "gradient_checkpointing": true,
70
+ "compile_model": true,
71
+ "compile_mode": "max-autotune",
72
+ "optimizer_lr": 2.5e-05,
73
+ "optimizer_betas": [
74
+ 0.9,
75
+ 0.95
76
+ ],
77
+ "optimizer_eps": 1e-08,
78
+ "optimizer_weight_decay": 0.01,
79
+ "optimizer_grad_clip_norm": 1.0,
80
+ "scheduler_warmup_steps": 1000,
81
+ "scheduler_decay_steps": 30000,
82
+ "scheduler_decay_lr": 2.5e-06
83
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e462f8aa60e5c5ad14a4698ea925ef13c0a4795fb2bc325d3c0ce18b897fb535
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
+ 6
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:bd77ac0599692304df9bbd183e7bcaf64318b5baf7b66cb36ba36827db0c6b61
3
+ size 7536
policy_preprocessor.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ 6
23
+ ]
24
+ },
25
+ "observation.images.top": {
26
+ "type": "VISUAL",
27
+ "shape": [
28
+ 3,
29
+ 480,
30
+ 640
31
+ ]
32
+ },
33
+ "observation.images.wrist": {
34
+ "type": "VISUAL",
35
+ "shape": [
36
+ 3,
37
+ 480,
38
+ 640
39
+ ]
40
+ },
41
+ "action": {
42
+ "type": "ACTION",
43
+ "shape": [
44
+ 6
45
+ ]
46
+ }
47
+ },
48
+ "norm_map": {
49
+ "VISUAL": "IDENTITY",
50
+ "STATE": "QUANTILES",
51
+ "ACTION": "QUANTILES"
52
+ }
53
+ },
54
+ "state_file": "policy_preprocessor_step_2_normalizer_processor.safetensors"
55
+ },
56
+ {
57
+ "registry_name": "pi05_prepare_state_tokenizer_processor_step",
58
+ "config": {}
59
+ },
60
+ {
61
+ "registry_name": "tokenizer_processor",
62
+ "config": {
63
+ "max_length": 200,
64
+ "task_key": "task",
65
+ "padding_side": "right",
66
+ "padding": "max_length",
67
+ "truncation": true,
68
+ "tokenizer_name": "google/paligemma-3b-pt-224"
69
+ }
70
+ },
71
+ {
72
+ "registry_name": "device_processor",
73
+ "config": {
74
+ "device": "cuda",
75
+ "float_dtype": null
76
+ }
77
+ }
78
+ ]
79
+ }
policy_preprocessor_step_2_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd77ac0599692304df9bbd183e7bcaf64318b5baf7b66cb36ba36827db0c6b61
3
+ size 7536
train_config.json ADDED
@@ -0,0 +1,196 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "thewisp/pick_place_earplug",
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
+ "streaming": false
67
+ },
68
+ "env": null,
69
+ "policy": {
70
+ "type": "pi05",
71
+ "n_obs_steps": 1,
72
+ "input_features": {
73
+ "observation.state": {
74
+ "type": "STATE",
75
+ "shape": [
76
+ 6
77
+ ]
78
+ },
79
+ "observation.images.top": {
80
+ "type": "VISUAL",
81
+ "shape": [
82
+ 3,
83
+ 480,
84
+ 640
85
+ ]
86
+ },
87
+ "observation.images.wrist": {
88
+ "type": "VISUAL",
89
+ "shape": [
90
+ 3,
91
+ 480,
92
+ 640
93
+ ]
94
+ }
95
+ },
96
+ "output_features": {
97
+ "action": {
98
+ "type": "ACTION",
99
+ "shape": [
100
+ 6
101
+ ]
102
+ }
103
+ },
104
+ "device": "cuda",
105
+ "use_amp": false,
106
+ "push_to_hub": true,
107
+ "repo_id": "thewisp/pi05_pick_place_earplug",
108
+ "private": null,
109
+ "tags": null,
110
+ "license": null,
111
+ "pretrained_path": "lerobot/pi05_base",
112
+ "paligemma_variant": "gemma_2b",
113
+ "action_expert_variant": "gemma_300m",
114
+ "dtype": "bfloat16",
115
+ "chunk_size": 50,
116
+ "n_action_steps": 50,
117
+ "max_state_dim": 32,
118
+ "max_action_dim": 32,
119
+ "num_inference_steps": 10,
120
+ "time_sampling_beta_alpha": 1.5,
121
+ "time_sampling_beta_beta": 1.0,
122
+ "time_sampling_scale": 0.999,
123
+ "time_sampling_offset": 0.001,
124
+ "min_period": 0.004,
125
+ "max_period": 4.0,
126
+ "image_resolution": [
127
+ 224,
128
+ 224
129
+ ],
130
+ "empty_cameras": 0,
131
+ "tokenizer_max_length": 200,
132
+ "normalization_mapping": {
133
+ "VISUAL": "IDENTITY",
134
+ "STATE": "QUANTILES",
135
+ "ACTION": "QUANTILES"
136
+ },
137
+ "gradient_checkpointing": true,
138
+ "compile_model": true,
139
+ "compile_mode": "max-autotune",
140
+ "optimizer_lr": 2.5e-05,
141
+ "optimizer_betas": [
142
+ 0.9,
143
+ 0.95
144
+ ],
145
+ "optimizer_eps": 1e-08,
146
+ "optimizer_weight_decay": 0.01,
147
+ "optimizer_grad_clip_norm": 1.0,
148
+ "scheduler_warmup_steps": 1000,
149
+ "scheduler_decay_steps": 30000,
150
+ "scheduler_decay_lr": 2.5e-06
151
+ },
152
+ "output_dir": "outputs/pi05_training",
153
+ "job_name": "pi05_training",
154
+ "resume": false,
155
+ "seed": 1000,
156
+ "num_workers": 4,
157
+ "batch_size": 2,
158
+ "steps": 48000,
159
+ "eval_freq": 20000,
160
+ "log_freq": 200,
161
+ "save_checkpoint": true,
162
+ "save_freq": 8000,
163
+ "use_policy_training_preset": true,
164
+ "optimizer": {
165
+ "type": "adamw",
166
+ "lr": 2.5e-05,
167
+ "weight_decay": 0.01,
168
+ "grad_clip_norm": 1.0,
169
+ "betas": [
170
+ 0.9,
171
+ 0.95
172
+ ],
173
+ "eps": 1e-08
174
+ },
175
+ "scheduler": {
176
+ "type": "cosine_decay_with_warmup",
177
+ "num_warmup_steps": 1000,
178
+ "num_decay_steps": 30000,
179
+ "peak_lr": 2.5e-05,
180
+ "decay_lr": 2.5e-06
181
+ },
182
+ "eval": {
183
+ "n_episodes": 50,
184
+ "batch_size": 50,
185
+ "use_async_envs": false
186
+ },
187
+ "wandb": {
188
+ "enable": true,
189
+ "disable_artifact": false,
190
+ "project": "lerobot",
191
+ "entity": null,
192
+ "notes": null,
193
+ "run_id": "sfljyr4v",
194
+ "mode": null
195
+ }
196
+ }