VihaanAgarwal commited on
Commit
f9f1f71
·
verified ·
1 Parent(s): 449b9d8

Upload folder using huggingface_hub

Browse files
checkpoint-002000/config.json ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "pi05",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.images.overhead": {
6
+ "type": "VISUAL",
7
+ "shape": [
8
+ 3,
9
+ 480,
10
+ 640
11
+ ]
12
+ },
13
+ "observation.images.wrist": {
14
+ "type": "VISUAL",
15
+ "shape": [
16
+ 3,
17
+ 480,
18
+ 640
19
+ ]
20
+ },
21
+ "observation.state": {
22
+ "type": "STATE",
23
+ "shape": [
24
+ 6
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
+ "use_peft": false,
39
+ "push_to_hub": true,
40
+ "repo_id": "Lab2A/pi05_pickplace_all",
41
+ "private": null,
42
+ "tags": null,
43
+ "license": null,
44
+ "pretrained_path": "lerobot/pi05_base",
45
+ "pretrained_revision": null,
46
+ "paligemma_variant": "gemma_2b",
47
+ "action_expert_variant": "gemma_300m",
48
+ "dtype": "bfloat16",
49
+ "chunk_size": 50,
50
+ "n_action_steps": 50,
51
+ "max_state_dim": 32,
52
+ "max_action_dim": 32,
53
+ "num_inference_steps": 10,
54
+ "time_sampling_beta_alpha": 1.5,
55
+ "time_sampling_beta_beta": 1.0,
56
+ "time_sampling_scale": 0.999,
57
+ "time_sampling_offset": 0.001,
58
+ "min_period": 0.004,
59
+ "max_period": 4.0,
60
+ "use_relative_actions": false,
61
+ "relative_exclude_joints": [
62
+ "gripper"
63
+ ],
64
+ "action_feature_names": null,
65
+ "rtc_config": null,
66
+ "image_resolution": [
67
+ 224,
68
+ 224
69
+ ],
70
+ "empty_cameras": 0,
71
+ "tokenizer_max_length": 200,
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_encoder": false,
81
+ "train_expert_only": false,
82
+ "optimizer_lr": 2.5e-05,
83
+ "optimizer_betas": [
84
+ 0.9,
85
+ 0.95
86
+ ],
87
+ "optimizer_eps": 1e-08,
88
+ "optimizer_weight_decay": 0.01,
89
+ "optimizer_grad_clip_norm": 1.0,
90
+ "scheduler_warmup_steps": 1000,
91
+ "scheduler_decay_steps": 30000,
92
+ "scheduler_decay_lr": 2.5e-06
93
+ }
checkpoint-002000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ba08137e2cf1e071f58262c5bf18df17a7bc9976faaa44921b9325d3fbc70ff
3
+ size 9354050752
checkpoint-002000/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
+ }
checkpoint-002000/policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7f133d79bece44d35d01ec8af9ca510be2f2cd5ba5d87c3fff8c7644c4dbd946
3
+ size 7584
checkpoint-002000/policy_preprocessor.json ADDED
@@ -0,0 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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.images.overhead": {
30
+ "type": "VISUAL",
31
+ "shape": [
32
+ 3,
33
+ 480,
34
+ 640
35
+ ]
36
+ },
37
+ "observation.images.wrist": {
38
+ "type": "VISUAL",
39
+ "shape": [
40
+ 3,
41
+ 480,
42
+ 640
43
+ ]
44
+ },
45
+ "observation.state": {
46
+ "type": "STATE",
47
+ "shape": [
48
+ 6
49
+ ]
50
+ },
51
+ "action": {
52
+ "type": "ACTION",
53
+ "shape": [
54
+ 6
55
+ ]
56
+ }
57
+ },
58
+ "norm_map": {
59
+ "VISUAL": "IDENTITY",
60
+ "STATE": "QUANTILES",
61
+ "ACTION": "QUANTILES"
62
+ }
63
+ },
64
+ "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors"
65
+ },
66
+ {
67
+ "registry_name": "pi05_prepare_state_tokenizer_processor_step",
68
+ "config": {}
69
+ },
70
+ {
71
+ "registry_name": "tokenizer_processor",
72
+ "config": {
73
+ "max_length": 200,
74
+ "task_key": "task",
75
+ "padding_side": "right",
76
+ "padding": "max_length",
77
+ "truncation": true,
78
+ "tokenizer_name": "google/paligemma-3b-pt-224"
79
+ }
80
+ },
81
+ {
82
+ "registry_name": "device_processor",
83
+ "config": {
84
+ "device": "cuda",
85
+ "float_dtype": null
86
+ }
87
+ }
88
+ ]
89
+ }
checkpoint-002000/policy_preprocessor_step_3_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fedf7dba4db70d326189adb3febf4fd0eb3a39ffe98a9f4132b16a8da75d13be
3
+ size 7592
checkpoint-002000/train_config.json ADDED
@@ -0,0 +1,246 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "Lab2A/so101_pickplace_all",
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": "main",
78
+ "use_imagenet_stats": true,
79
+ "video_backend": "pyav",
80
+ "return_uint8": false,
81
+ "depth_output_unit": "mm",
82
+ "streaming": false,
83
+ "eval_split": 0.0
84
+ },
85
+ "env": null,
86
+ "policy": {
87
+ "type": "pi05",
88
+ "n_obs_steps": 1,
89
+ "input_features": {
90
+ "observation.images.overhead": {
91
+ "type": "VISUAL",
92
+ "shape": [
93
+ 3,
94
+ 480,
95
+ 640
96
+ ]
97
+ },
98
+ "observation.images.wrist": {
99
+ "type": "VISUAL",
100
+ "shape": [
101
+ 3,
102
+ 480,
103
+ 640
104
+ ]
105
+ },
106
+ "observation.state": {
107
+ "type": "STATE",
108
+ "shape": [
109
+ 6
110
+ ]
111
+ }
112
+ },
113
+ "output_features": {
114
+ "action": {
115
+ "type": "ACTION",
116
+ "shape": [
117
+ 6
118
+ ]
119
+ }
120
+ },
121
+ "device": "cuda",
122
+ "use_amp": false,
123
+ "use_peft": false,
124
+ "push_to_hub": true,
125
+ "repo_id": "Lab2A/pi05_pickplace_all",
126
+ "private": null,
127
+ "tags": null,
128
+ "license": null,
129
+ "pretrained_path": "lerobot/pi05_base",
130
+ "pretrained_revision": null,
131
+ "paligemma_variant": "gemma_2b",
132
+ "action_expert_variant": "gemma_300m",
133
+ "dtype": "bfloat16",
134
+ "chunk_size": 50,
135
+ "n_action_steps": 50,
136
+ "max_state_dim": 32,
137
+ "max_action_dim": 32,
138
+ "num_inference_steps": 10,
139
+ "time_sampling_beta_alpha": 1.5,
140
+ "time_sampling_beta_beta": 1.0,
141
+ "time_sampling_scale": 0.999,
142
+ "time_sampling_offset": 0.001,
143
+ "min_period": 0.004,
144
+ "max_period": 4.0,
145
+ "use_relative_actions": false,
146
+ "relative_exclude_joints": [
147
+ "gripper"
148
+ ],
149
+ "action_feature_names": null,
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
+ "freeze_vision_encoder": false,
166
+ "train_expert_only": false,
167
+ "optimizer_lr": 2.5e-05,
168
+ "optimizer_betas": [
169
+ 0.9,
170
+ 0.95
171
+ ],
172
+ "optimizer_eps": 1e-08,
173
+ "optimizer_weight_decay": 0.01,
174
+ "optimizer_grad_clip_norm": 1.0,
175
+ "scheduler_warmup_steps": 1000,
176
+ "scheduler_decay_steps": 30000,
177
+ "scheduler_decay_lr": 2.5e-06
178
+ },
179
+ "reward_model": null,
180
+ "output_dir": "/workspace/outputs/pi05_pickplace_all",
181
+ "job_name": "pi05_pickplace_all",
182
+ "resume": false,
183
+ "seed": 1000,
184
+ "cudnn_deterministic": false,
185
+ "num_workers": 4,
186
+ "batch_size": 16,
187
+ "prefetch_factor": 4,
188
+ "persistent_workers": true,
189
+ "steps": 6000,
190
+ "env_eval_freq": 20000,
191
+ "log_freq": 100,
192
+ "eval_steps": 0,
193
+ "max_eval_samples": 0,
194
+ "tolerance_s": 0.0001,
195
+ "save_checkpoint": true,
196
+ "save_freq": 1000,
197
+ "use_policy_training_preset": true,
198
+ "optimizer": {
199
+ "type": "adamw",
200
+ "lr": 2.5e-05,
201
+ "weight_decay": 0.01,
202
+ "grad_clip_norm": 1.0,
203
+ "betas": [
204
+ 0.9,
205
+ 0.95
206
+ ],
207
+ "eps": 1e-08
208
+ },
209
+ "scheduler": {
210
+ "type": "cosine_decay_with_warmup",
211
+ "num_warmup_steps": 1000,
212
+ "num_decay_steps": 30000,
213
+ "peak_lr": 2.5e-05,
214
+ "decay_lr": 2.5e-06
215
+ },
216
+ "eval": {
217
+ "n_episodes": 50,
218
+ "batch_size": 50,
219
+ "use_async_envs": true,
220
+ "recording": false,
221
+ "recording_repo_id": null,
222
+ "recording_private": false
223
+ },
224
+ "wandb": {
225
+ "enable": true,
226
+ "disable_artifact": false,
227
+ "project": "pi05-so101-generalist",
228
+ "entity": null,
229
+ "notes": null,
230
+ "run_id": "mwz04glh",
231
+ "mode": null,
232
+ "add_tags": true
233
+ },
234
+ "peft": null,
235
+ "job": {
236
+ "target": null,
237
+ "image": "huggingface/lerobot-gpu:latest",
238
+ "timeout": "2d",
239
+ "detach": false,
240
+ "tags": []
241
+ },
242
+ "save_checkpoint_to_hub": false,
243
+ "sample_weighting": null,
244
+ "rename_map": {},
245
+ "checkpoint_path": null
246
+ }