Hugging Face's logo

srik410
/
pi05_book_parallel_grasp

Safetensors
Model card Files
xet
 Community
srik410 commited on
Commit
1be804e
·
verified ·
1 Parent(s): a106414

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +91 -0
  2. model.safetensors +3 -0
  3. policy_postprocessor.json +38 -0
  4. policy_postprocessor_step_0_unnormalizer_processor.safetensors +3 -0
  5. policy_preprocessor.json +81 -0
  6. policy_preprocessor_step_3_normalizer_processor.safetensors +3 -0
  7. train_config.json +229 -0
config.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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.hand_cam": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 480,
16
+ 640
17
+ ]
18
+ }
19
+ },
20
+ "output_features": {
21
+ "action": {
22
+ "type": "ACTION",
23
+ "shape": [
24
+ 6
25
+ ]
26
+ }
27
+ },
28
+ "device": "cuda",
29
+ "use_amp": false,
30
+ "use_peft": false,
31
+ "push_to_hub": true,
32
+ "repo_id": "srik410/pi05_book_parallel_grasp",
33
+ "private": null,
34
+ "tags": null,
35
+ "license": null,
36
+ "pretrained_path": "lerobot/pi05_base",
37
+ "paligemma_variant": "gemma_2b",
38
+ "action_expert_variant": "gemma_300m",
39
+ "dtype": "bfloat16",
40
+ "chunk_size": 50,
41
+ "n_action_steps": 50,
42
+ "max_state_dim": 32,
43
+ "max_action_dim": 32,
44
+ "num_inference_steps": 10,
45
+ "time_sampling_beta_alpha": 1.5,
46
+ "time_sampling_beta_beta": 1.0,
47
+ "time_sampling_scale": 0.999,
48
+ "time_sampling_offset": 0.001,
49
+ "min_period": 0.004,
50
+ "max_period": 4.0,
51
+ "use_relative_actions": false,
52
+ "relative_exclude_joints": [
53
+ "gripper"
54
+ ],
55
+ "action_feature_names": [
56
+ "shoulder_pan.pos",
57
+ "shoulder_lift.pos",
58
+ "elbow_flex.pos",
59
+ "wrist_flex.pos",
60
+ "wrist_roll.pos",
61
+ "gripper.pos"
62
+ ],
63
+ "rtc_config": null,
64
+ "image_resolution": [
65
+ 224,
66
+ 224
67
+ ],
68
+ "empty_cameras": 0,
69
+ "tokenizer_max_length": 200,
70
+ "normalization_mapping": {
71
+ "VISUAL": "IDENTITY",
72
+ "STATE": "QUANTILES",
73
+ "ACTION": "QUANTILES"
74
+ },
75
+ "gradient_checkpointing": true,
76
+ "compile_model": true,
77
+ "compile_mode": "max-autotune",
78
+ "freeze_vision_encoder": false,
79
+ "train_expert_only": true,
80
+ "optimizer_lr": 2.5e-05,
81
+ "optimizer_betas": [
82
+ 0.9,
83
+ 0.95
84
+ ],
85
+ "optimizer_eps": 1e-08,
86
+ "optimizer_weight_decay": 0.01,
87
+ "optimizer_grad_clip_norm": 1.0,
88
+ "scheduler_warmup_steps": 1000,
89
+ "scheduler_decay_steps": 30000,
90
+ "scheduler_decay_lr": 2.5e-06
91
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:36dc6fe6f5d6db95411b38c59a9f14028c8a51ea25d87f71548891ac43feda8e
3
+ size 9354050752
policy_postprocessor.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "absolute_actions_processor",
26
+ "config": {
27
+ "enabled": false
28
+ }
29
+ },
30
+ {
31
+ "registry_name": "device_processor",
32
+ "config": {
33
+ "device": "cpu",
34
+ "float_dtype": null
35
+ }
36
+ }
37
+ ]
38
+ }
policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5949ecfe8ee390a6a37c5188df489eeff1da448a0157aa2c857f871cca86879a
3
+ size 6592
policy_preprocessor.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "relative_actions_processor",
16
+ "config": {
17
+ "enabled": false,
18
+ "exclude_joints": [
19
+ "gripper"
20
+ ],
21
+ "action_names": null
22
+ }
23
+ },
24
+ {
25
+ "registry_name": "normalizer_processor",
26
+ "config": {
27
+ "eps": 1e-08,
28
+ "features": {
29
+ "observation.state": {
30
+ "type": "STATE",
31
+ "shape": [
32
+ 6
33
+ ]
34
+ },
35
+ "observation.images.hand_cam": {
36
+ "type": "VISUAL",
37
+ "shape": [
38
+ 3,
39
+ 480,
40
+ 640
41
+ ]
42
+ },
43
+ "action": {
44
+ "type": "ACTION",
45
+ "shape": [
46
+ 6
47
+ ]
48
+ }
49
+ },
50
+ "norm_map": {
51
+ "VISUAL": "IDENTITY",
52
+ "STATE": "QUANTILES",
53
+ "ACTION": "QUANTILES"
54
+ }
55
+ },
56
+ "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors"
57
+ },
58
+ {
59
+ "registry_name": "pi05_prepare_state_tokenizer_processor_step",
60
+ "config": {}
61
+ },
62
+ {
63
+ "registry_name": "tokenizer_processor",
64
+ "config": {
65
+ "max_length": 200,
66
+ "task_key": "task",
67
+ "padding_side": "right",
68
+ "padding": "max_length",
69
+ "truncation": true,
70
+ "tokenizer_name": "google/paligemma-3b-pt-224"
71
+ }
72
+ },
73
+ {
74
+ "registry_name": "device_processor",
75
+ "config": {
76
+ "device": "cuda",
77
+ "float_dtype": null
78
+ }
79
+ }
80
+ ]
81
+ }
policy_preprocessor_step_3_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4d87ddc0240c06f9f1adaaf37d9c859f1e5d19553fcf71ddbcd5895e2101064
3
+ size 6600
train_config.json ADDED
@@ -0,0 +1,229 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "srik410/makermods_pick_book_parallel_grasp_merged",
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
+ "return_uint8": false,
81
+ "streaming": false
82
+ },
83
+ "env": null,
84
+ "policy": {
85
+ "type": "pi05",
86
+ "n_obs_steps": 1,
87
+ "input_features": {
88
+ "observation.state": {
89
+ "type": "STATE",
90
+ "shape": [
91
+ 6
92
+ ]
93
+ },
94
+ "observation.images.hand_cam": {
95
+ "type": "VISUAL",
96
+ "shape": [
97
+ 3,
98
+ 480,
99
+ 640
100
+ ]
101
+ }
102
+ },
103
+ "output_features": {
104
+ "action": {
105
+ "type": "ACTION",
106
+ "shape": [
107
+ 6
108
+ ]
109
+ }
110
+ },
111
+ "device": "cuda",
112
+ "use_amp": false,
113
+ "use_peft": false,
114
+ "push_to_hub": true,
115
+ "repo_id": "srik410/pi05_book_parallel_grasp",
116
+ "private": null,
117
+ "tags": null,
118
+ "license": null,
119
+ "pretrained_path": "lerobot/pi05_base",
120
+ "paligemma_variant": "gemma_2b",
121
+ "action_expert_variant": "gemma_300m",
122
+ "dtype": "bfloat16",
123
+ "chunk_size": 50,
124
+ "n_action_steps": 50,
125
+ "max_state_dim": 32,
126
+ "max_action_dim": 32,
127
+ "num_inference_steps": 10,
128
+ "time_sampling_beta_alpha": 1.5,
129
+ "time_sampling_beta_beta": 1.0,
130
+ "time_sampling_scale": 0.999,
131
+ "time_sampling_offset": 0.001,
132
+ "min_period": 0.004,
133
+ "max_period": 4.0,
134
+ "use_relative_actions": false,
135
+ "relative_exclude_joints": [
136
+ "gripper"
137
+ ],
138
+ "action_feature_names": [
139
+ "shoulder_pan.pos",
140
+ "shoulder_lift.pos",
141
+ "elbow_flex.pos",
142
+ "wrist_flex.pos",
143
+ "wrist_roll.pos",
144
+ "gripper.pos"
145
+ ],
146
+ "rtc_config": null,
147
+ "image_resolution": [
148
+ 224,
149
+ 224
150
+ ],
151
+ "empty_cameras": 0,
152
+ "tokenizer_max_length": 200,
153
+ "normalization_mapping": {
154
+ "VISUAL": "IDENTITY",
155
+ "STATE": "QUANTILES",
156
+ "ACTION": "QUANTILES"
157
+ },
158
+ "gradient_checkpointing": true,
159
+ "compile_model": true,
160
+ "compile_mode": "max-autotune",
161
+ "freeze_vision_encoder": false,
162
+ "train_expert_only": true,
163
+ "optimizer_lr": 2.5e-05,
164
+ "optimizer_betas": [
165
+ 0.9,
166
+ 0.95
167
+ ],
168
+ "optimizer_eps": 1e-08,
169
+ "optimizer_weight_decay": 0.01,
170
+ "optimizer_grad_clip_norm": 1.0,
171
+ "scheduler_warmup_steps": 1000,
172
+ "scheduler_decay_steps": 30000,
173
+ "scheduler_decay_lr": 2.5e-06
174
+ },
175
+ "reward_model": null,
176
+ "output_dir": "outputs/train/pi05_book_parallel_grasp",
177
+ "job_name": "pi05_book_parallel_grasp",
178
+ "resume": false,
179
+ "seed": 1000,
180
+ "cudnn_deterministic": false,
181
+ "num_workers": 8,
182
+ "batch_size": 32,
183
+ "prefetch_factor": 4,
184
+ "persistent_workers": true,
185
+ "steps": 10000,
186
+ "eval_freq": 20000,
187
+ "log_freq": 100,
188
+ "tolerance_s": 0.0001,
189
+ "save_checkpoint": true,
190
+ "save_freq": 2000,
191
+ "use_policy_training_preset": true,
192
+ "optimizer": {
193
+ "type": "adamw",
194
+ "lr": 2.5e-05,
195
+ "weight_decay": 0.01,
196
+ "grad_clip_norm": 1.0,
197
+ "betas": [
198
+ 0.9,
199
+ 0.95
200
+ ],
201
+ "eps": 1e-08
202
+ },
203
+ "scheduler": {
204
+ "type": "cosine_decay_with_warmup",
205
+ "num_warmup_steps": 1000,
206
+ "num_decay_steps": 30000,
207
+ "peak_lr": 2.5e-05,
208
+ "decay_lr": 2.5e-06
209
+ },
210
+ "eval": {
211
+ "n_episodes": 50,
212
+ "batch_size": 19,
213
+ "use_async_envs": true
214
+ },
215
+ "wandb": {
216
+ "enable": false,
217
+ "disable_artifact": false,
218
+ "project": "lerobot",
219
+ "entity": null,
220
+ "notes": null,
221
+ "run_id": null,
222
+ "mode": null,
223
+ "add_tags": true
224
+ },
225
+ "peft": null,
226
+ "sample_weighting": null,
227
+ "rename_map": {},
228
+ "checkpoint_path": null
229
+ }