Hugging Face's logo

CSI-Agent
/
pi05_expert_6k

Safetensors
Model card Files
xet
Community
HyeonseokE commited on
Commit
817500f
·
verified ·
1 Parent(s): 4065359

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 +71 -0
  6. policy_preprocessor_step_2_normalizer_processor.safetensors +3 -0
  7. train_config.json +205 -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
+ 18
9
+ ]
10
+ },
11
+ "observation.images.robot2_top": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 240,
16
+ 424
17
+ ]
18
+ }
19
+ },
20
+ "output_features": {
21
+ "action": {
22
+ "type": "ACTION",
23
+ "shape": [
24
+ 18
25
+ ]
26
+ }
27
+ },
28
+ "device": "cuda",
29
+ "use_amp": false,
30
+ "push_to_hub": false,
31
+ "repo_id": null,
32
+ "private": null,
33
+ "tags": null,
34
+ "license": null,
35
+ "pretrained_path": "lerobot/pi05_base",
36
+ "paligemma_variant": "gemma_2b",
37
+ "action_expert_variant": "gemma_300m",
38
+ "dtype": "bfloat16",
39
+ "chunk_size": 15,
40
+ "n_action_steps": 15,
41
+ "max_state_dim": 32,
42
+ "max_action_dim": 32,
43
+ "num_inference_steps": 10,
44
+ "time_sampling_beta_alpha": 1.5,
45
+ "time_sampling_beta_beta": 1.0,
46
+ "time_sampling_scale": 0.999,
47
+ "time_sampling_offset": 0.001,
48
+ "min_period": 0.004,
49
+ "max_period": 4.0,
50
+ "image_resolution": [
51
+ 224,
52
+ 224
53
+ ],
54
+ "empty_cameras": 0,
55
+ "tokenizer_max_length": 200,
56
+ "normalization_mapping": {
57
+ "VISUAL": "IDENTITY",
58
+ "STATE": "QUANTILES",
59
+ "ACTION": "QUANTILES"
60
+ },
61
+ "gradient_checkpointing": true,
62
+ "compile_model": false,
63
+ "compile_mode": "max-autotune",
64
+ "optimizer_lr": 2.5e-05,
65
+ "optimizer_betas": [
66
+ 0.9,
67
+ 0.95
68
+ ],
69
+ "optimizer_eps": 1e-08,
70
+ "optimizer_weight_decay": 0.01,
71
+ "optimizer_grad_clip_norm": 1.0,
72
+ "scheduler_warmup_steps": 1000,
73
+ "scheduler_decay_steps": 30000,
74
+ "scheduler_decay_lr": 2.5e-06,
75
+ "tune_llm": false,
76
+ "tune_visual": false,
77
+ "tune_projector": true,
78
+ "tune_expert": true,
79
+ "lora_rank": 16,
80
+ "lora_alpha": 16,
81
+ "lora_dropout": 0.1,
82
+ "lora_full_model": true
83
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b1ee0b11602b66f9deab203e8d56835ca6f241f9a2e48824aa2ed76bd2e18da
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
+ 18
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 sha256:17f356cf25c5b00e0e5511813a9e64b444f08be30df23d6a03704502929f4b51
3
+ size 7696
policy_preprocessor.json ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ 18
23
+ ]
24
+ },
25
+ "observation.images.robot2_top": {
26
+ "type": "VISUAL",
27
+ "shape": [
28
+ 3,
29
+ 240,
30
+ 424
31
+ ]
32
+ },
33
+ "action": {
34
+ "type": "ACTION",
35
+ "shape": [
36
+ 18
37
+ ]
38
+ }
39
+ },
40
+ "norm_map": {
41
+ "VISUAL": "IDENTITY",
42
+ "STATE": "QUANTILES",
43
+ "ACTION": "QUANTILES"
44
+ }
45
+ },
46
+ "state_file": "policy_preprocessor_step_2_normalizer_processor.safetensors"
47
+ },
48
+ {
49
+ "registry_name": "pi05_prepare_state_tokenizer_processor_step",
50
+ "config": {}
51
+ },
52
+ {
53
+ "registry_name": "tokenizer_processor",
54
+ "config": {
55
+ "max_length": 200,
56
+ "task_key": "task",
57
+ "padding_side": "right",
58
+ "padding": "max_length",
59
+ "truncation": true,
60
+ "tokenizer_name": "google/paligemma-3b-pt-224"
61
+ }
62
+ },
63
+ {
64
+ "registry_name": "device_processor",
65
+ "config": {
66
+ "device": "cuda",
67
+ "float_dtype": null
68
+ }
69
+ }
70
+ ]
71
+ }
policy_preprocessor_step_2_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17f356cf25c5b00e0e5511813a9e64b444f08be30df23d6a03704502929f4b51
3
+ size 7696
train_config.json ADDED
@@ -0,0 +1,205 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "CSI-Agent/merged_robot_imit_",
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
+ "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": "torchcodec",
80
+ "streaming": false
81
+ },
82
+ "env": null,
83
+ "policy": {
84
+ "type": "pi05",
85
+ "n_obs_steps": 1,
86
+ "input_features": {
87
+ "observation.state": {
88
+ "type": "STATE",
89
+ "shape": [
90
+ 18
91
+ ]
92
+ },
93
+ "observation.images.robot2_top": {
94
+ "type": "VISUAL",
95
+ "shape": [
96
+ 3,
97
+ 240,
98
+ 424
99
+ ]
100
+ }
101
+ },
102
+ "output_features": {
103
+ "action": {
104
+ "type": "ACTION",
105
+ "shape": [
106
+ 18
107
+ ]
108
+ }
109
+ },
110
+ "device": "cuda",
111
+ "use_amp": false,
112
+ "push_to_hub": false,
113
+ "repo_id": null,
114
+ "private": null,
115
+ "tags": null,
116
+ "license": null,
117
+ "pretrained_path": "lerobot/pi05_base",
118
+ "paligemma_variant": "gemma_2b",
119
+ "action_expert_variant": "gemma_300m",
120
+ "dtype": "bfloat16",
121
+ "chunk_size": 15,
122
+ "n_action_steps": 15,
123
+ "max_state_dim": 32,
124
+ "max_action_dim": 32,
125
+ "num_inference_steps": 10,
126
+ "time_sampling_beta_alpha": 1.5,
127
+ "time_sampling_beta_beta": 1.0,
128
+ "time_sampling_scale": 0.999,
129
+ "time_sampling_offset": 0.001,
130
+ "min_period": 0.004,
131
+ "max_period": 4.0,
132
+ "image_resolution": [
133
+ 224,
134
+ 224
135
+ ],
136
+ "empty_cameras": 0,
137
+ "tokenizer_max_length": 200,
138
+ "normalization_mapping": {
139
+ "VISUAL": "IDENTITY",
140
+ "STATE": "QUANTILES",
141
+ "ACTION": "QUANTILES"
142
+ },
143
+ "gradient_checkpointing": true,
144
+ "compile_model": false,
145
+ "compile_mode": "max-autotune",
146
+ "optimizer_lr": 2.5e-05,
147
+ "optimizer_betas": [
148
+ 0.9,
149
+ 0.95
150
+ ],
151
+ "optimizer_eps": 1e-08,
152
+ "optimizer_weight_decay": 0.01,
153
+ "optimizer_grad_clip_norm": 1.0,
154
+ "scheduler_warmup_steps": 1000,
155
+ "scheduler_decay_steps": 30000,
156
+ "scheduler_decay_lr": 2.5e-06,
157
+ "train_expert_only": true
158
+ },
159
+ "output_dir": "outputs/pi05_multi",
160
+ "job_name": "pi05_multi",
161
+ "resume": false,
162
+ "seed": 1000,
163
+ "num_workers": 8,
164
+ "batch_size": 32,
165
+ "steps": 15000,
166
+ "eval_freq": 20000,
167
+ "log_freq": 100,
168
+ "save_checkpoint": true,
169
+ "save_freq": 3000,
170
+ "use_policy_training_preset": true,
171
+ "optimizer": {
172
+ "type": "adamw",
173
+ "lr": 2.5e-05,
174
+ "weight_decay": 0.01,
175
+ "grad_clip_norm": 1.0,
176
+ "betas": [
177
+ 0.9,
178
+ 0.95
179
+ ],
180
+ "eps": 1e-08
181
+ },
182
+ "scheduler": {
183
+ "type": "cosine_decay_with_warmup",
184
+ "num_warmup_steps": 1000,
185
+ "num_decay_steps": 30000,
186
+ "peak_lr": 2.5e-05,
187
+ "decay_lr": 2.5e-06
188
+ },
189
+ "eval": {
190
+ "n_episodes": 50,
191
+ "batch_size": 50,
192
+ "use_async_envs": false
193
+ },
194
+ "wandb": {
195
+ "enable": true,
196
+ "disable_artifact": false,
197
+ "project": "lerobot",
198
+ "entity": null,
199
+ "notes": null,
200
+ "run_id": "oddaksxk",
201
+ "mode": null
202
+ },
203
+ "checkpoint_path": null,
204
+ "rename_map": {}
205
+ }