Hugging Face's logo

nikriz
/
pi05_vl20k

Safetensors
Model card Files
xet
Community
nikriz commited on
Commit
459a7c4
·
verified ·
1 Parent(s): 3b7c3dc

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +94 -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 +223 -0
config.json ADDED
@@ -0,0 +1,94 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "pi_align",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.images.image": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 256,
10
+ 256
11
+ ]
12
+ },
13
+ "observation.state": {
14
+ "type": "STATE",
15
+ "shape": [
16
+ 7
17
+ ]
18
+ }
19
+ },
20
+ "output_features": {
21
+ "action": {
22
+ "type": "ACTION",
23
+ "shape": [
24
+ 7
25
+ ]
26
+ }
27
+ },
28
+ "device": "cuda",
29
+ "use_amp": false,
30
+ "push_to_hub": true,
31
+ "repo_id": "nikriz/lerobot_pi05_align_bridge_oxe",
32
+ "private": null,
33
+ "tags": null,
34
+ "license": null,
35
+ "pretrained_path": "lerobot/pi05_base",
36
+ "base_policy": "pi05",
37
+ "lam_checkpoint_path": "/home/work/robotics/junhachun/dev/vggt_lam/bridge_oxe/checkpoints/last/lam_stage1.ckpt",
38
+ "alignment_loss_weight": 1.0,
39
+ "action_token_layer": 10,
40
+ "paligemma_variant": "gemma_2b",
41
+ "action_expert_variant": "gemma_300m",
42
+ "dtype": "bfloat16",
43
+ "kv_attn": false,
44
+ "kv_projector_type": "linear",
45
+ "sinkhorn": 1,
46
+ "spnorm_gamma": null,
47
+ "eval_nope_front": false,
48
+ "eval_nope_wrist": false,
49
+ "eval_ignore_front": false,
50
+ "eval_ignore_wrist": false,
51
+ "eval_ignore_text": false,
52
+ "eval_vlm_independent": false,
53
+ "eval_vlm_causal": false,
54
+ "chunk_size": 50,
55
+ "n_action_steps": 50,
56
+ "max_state_dim": 32,
57
+ "max_action_dim": 32,
58
+ "num_inference_steps": 10,
59
+ "time_sampling_beta_alpha": 1.5,
60
+ "time_sampling_beta_beta": 1.0,
61
+ "time_sampling_scale": 0.999,
62
+ "time_sampling_offset": 0.001,
63
+ "min_period": 0.004,
64
+ "max_period": 4.0,
65
+ "rtc_config": null,
66
+ "image_resolution": [
67
+ 224,
68
+ 224
69
+ ],
70
+ "empty_cameras": 0,
71
+ "tokenizer_max_length": 48,
72
+ "normalization_mapping": {
73
+ "VISUAL": "IDENTITY",
74
+ "STATE": "QUANTILES",
75
+ "ACTION": "QUANTILES"
76
+ },
77
+ "gradient_checkpointing": true,
78
+ "compile_model": false,
79
+ "compile_mode": "max-autotune",
80
+ "freeze_vision_tower": false,
81
+ "freeze_language_model": false,
82
+ "freeze_expert": false,
83
+ "optimizer_lr": 2.5e-05,
84
+ "optimizer_betas": [
85
+ 0.9,
86
+ 0.95
87
+ ],
88
+ "optimizer_eps": 1e-08,
89
+ "optimizer_weight_decay": 0.01,
90
+ "optimizer_grad_clip_norm": 1.0,
91
+ "scheduler_warmup_steps": 1000,
92
+ "scheduler_decay_steps": 30000,
93
+ "scheduler_decay_lr": 2.5e-06
94
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56f7ac8eece57a5fb9ccc4dfe01c1b75f90baf07d2f04f89afdf32704fded371
3
+ size 13317537248
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
+ 7
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:ac3240d87ee2270596b2d5827ea4a4c68d76b064319b6e205ae64abcf1a85338
3
+ size 6632
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.images.image": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 256,
24
+ 256
25
+ ]
26
+ },
27
+ "observation.state": {
28
+ "type": "STATE",
29
+ "shape": [
30
+ 7
31
+ ]
32
+ },
33
+ "action": {
34
+ "type": "ACTION",
35
+ "shape": [
36
+ 7
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:ac3240d87ee2270596b2d5827ea4a4c68d76b064319b6e205ae64abcf1a85338
3
+ size 6632
train_config.json ADDED
@@ -0,0 +1,223 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "Jackie1/bridge_data_v2_convert_to_lerobot",
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": "pi_align",
85
+ "n_obs_steps": 1,
86
+ "input_features": {
87
+ "observation.images.image": {
88
+ "type": "VISUAL",
89
+ "shape": [
90
+ 3,
91
+ 256,
92
+ 256
93
+ ]
94
+ },
95
+ "observation.state": {
96
+ "type": "STATE",
97
+ "shape": [
98
+ 7
99
+ ]
100
+ }
101
+ },
102
+ "output_features": {
103
+ "action": {
104
+ "type": "ACTION",
105
+ "shape": [
106
+ 7
107
+ ]
108
+ }
109
+ },
110
+ "device": "cuda",
111
+ "use_amp": false,
112
+ "push_to_hub": true,
113
+ "repo_id": "nikriz/lerobot_pi05_align_bridge_oxe",
114
+ "private": null,
115
+ "tags": null,
116
+ "license": null,
117
+ "pretrained_path": "lerobot/pi05_base",
118
+ "base_policy": "pi05",
119
+ "lam_checkpoint_path": "/home/work/robotics/junhachun/dev/vggt_lam/bridge_oxe/checkpoints/last/lam_stage1.ckpt",
120
+ "alignment_loss_weight": 1.0,
121
+ "action_token_layer": 10,
122
+ "paligemma_variant": "gemma_2b",
123
+ "action_expert_variant": "gemma_300m",
124
+ "dtype": "bfloat16",
125
+ "kv_attn": false,
126
+ "kv_projector_type": "linear",
127
+ "sinkhorn": 1,
128
+ "spnorm_gamma": null,
129
+ "eval_nope_front": false,
130
+ "eval_nope_wrist": false,
131
+ "eval_ignore_front": false,
132
+ "eval_ignore_wrist": false,
133
+ "eval_ignore_text": false,
134
+ "eval_vlm_independent": false,
135
+ "eval_vlm_causal": false,
136
+ "chunk_size": 50,
137
+ "n_action_steps": 50,
138
+ "max_state_dim": 32,
139
+ "max_action_dim": 32,
140
+ "num_inference_steps": 10,
141
+ "time_sampling_beta_alpha": 1.5,
142
+ "time_sampling_beta_beta": 1.0,
143
+ "time_sampling_scale": 0.999,
144
+ "time_sampling_offset": 0.001,
145
+ "min_period": 0.004,
146
+ "max_period": 4.0,
147
+ "rtc_config": null,
148
+ "image_resolution": [
149
+ 224,
150
+ 224
151
+ ],
152
+ "empty_cameras": 0,
153
+ "tokenizer_max_length": 48,
154
+ "normalization_mapping": {
155
+ "VISUAL": "IDENTITY",
156
+ "STATE": "QUANTILES",
157
+ "ACTION": "QUANTILES"
158
+ },
159
+ "gradient_checkpointing": true,
160
+ "compile_model": false,
161
+ "compile_mode": "max-autotune",
162
+ "freeze_vision_tower": false,
163
+ "freeze_language_model": false,
164
+ "freeze_expert": false,
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": "/home/work/robotics/junhachun/dev/vggt_lam_pi05/bridge_oxe",
178
+ "job_name": "lerobot_pi05_align_bridge_oxe",
179
+ "resume": false,
180
+ "seed": 1000,
181
+ "num_workers": 4,
182
+ "batch_size": 32,
183
+ "steps": 50000,
184
+ "eval_freq": 5000,
185
+ "log_freq": 200,
186
+ "save_checkpoint": true,
187
+ "save_freq": 10000,
188
+ "use_policy_training_preset": true,
189
+ "optimizer": {
190
+ "type": "adamw",
191
+ "lr": 2.5e-05,
192
+ "weight_decay": 0.01,
193
+ "grad_clip_norm": 1.0,
194
+ "betas": [
195
+ 0.9,
196
+ 0.95
197
+ ],
198
+ "eps": 1e-08
199
+ },
200
+ "scheduler": {
201
+ "type": "cosine_decay_with_warmup",
202
+ "num_warmup_steps": 1000,
203
+ "num_decay_steps": 30000,
204
+ "peak_lr": 2.5e-05,
205
+ "decay_lr": 2.5e-06
206
+ },
207
+ "eval": {
208
+ "n_episodes": 50,
209
+ "batch_size": 50,
210
+ "use_async_envs": false
211
+ },
212
+ "wandb": {
213
+ "enable": true,
214
+ "disable_artifact": false,
215
+ "project": "lerobot",
216
+ "entity": null,
217
+ "notes": null,
218
+ "run_id": "8qwbpyej",
219
+ "mode": null
220
+ },
221
+ "checkpoint_path": null,
222
+ "rename_map": {}
223
+ }