Hugging Face's logo

CSI-Agent
/
groot_visual_expert_tune

Safetensors
Model card Files
xet
Community
HyeonseokE commited on
Commit
4a88873
·
verified ·
1 Parent(s): 40cb42b

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +80 -0
  2. model.safetensors +3 -0
  3. policy_postprocessor.json +20 -0
  4. policy_postprocessor_step_0_groot_action_unpack_unnormalize_v1.safetensors +3 -0
  5. policy_preprocessor.json +52 -0
  6. policy_preprocessor_step_2_groot_pack_inputs_v3.safetensors +3 -0
  7. train_config.json +209 -0
config.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "groot",
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": "your-hf-user/groot-finetune",
32
+ "private": null,
33
+ "tags": null,
34
+ "license": null,
35
+ "pretrained_path": null,
36
+ "chunk_size": 16,
37
+ "n_action_steps": 16,
38
+ "max_state_dim": 64,
39
+ "max_action_dim": 32,
40
+ "normalization_mapping": {
41
+ "VISUAL": "IDENTITY",
42
+ "STATE": "MEAN_STD",
43
+ "ACTION": "MEAN_STD"
44
+ },
45
+ "image_size": [
46
+ 224,
47
+ 224
48
+ ],
49
+ "base_model_path": "nvidia/GR00T-N1.5-3B",
50
+ "tokenizer_assets_repo": "lerobot/eagle2hg-processor-groot-n1p5",
51
+ "embodiment_tag": "new_embodiment",
52
+ "tune_llm": false,
53
+ "tune_visual": true,
54
+ "tune_projector": true,
55
+ "tune_diffusion_model": true,
56
+ "lora_rank": 0,
57
+ "lora_alpha": 16,
58
+ "lora_dropout": 0.1,
59
+ "lora_full_model": false,
60
+ "optimizer_lr": 0.0001,
61
+ "optimizer_betas": [
62
+ 0.95,
63
+ 0.999
64
+ ],
65
+ "optimizer_eps": 1e-08,
66
+ "optimizer_weight_decay": 1e-05,
67
+ "warmup_ratio": 0.05,
68
+ "use_bf16": true,
69
+ "video_backend": "decord",
70
+ "balance_dataset_weights": true,
71
+ "balance_trajectory_weights": true,
72
+ "dataset_paths": null,
73
+ "output_dir": "./tmp/gr00t",
74
+ "save_steps": 1000,
75
+ "max_steps": 10000,
76
+ "batch_size": 32,
77
+ "dataloader_num_workers": 8,
78
+ "report_to": "wandb",
79
+ "resume": false
80
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c3ca9ecb89fbdbfc0c16b7341ba7fc54676602f2e793549dad1706f409545fc0
3
+ size 7586084224
policy_postprocessor.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "policy_postprocessor",
3
+ "steps": [
4
+ {
5
+ "registry_name": "groot_action_unpack_unnormalize_v1",
6
+ "config": {
7
+ "env_action_dim": 18,
8
+ "normalize_min_max": true
9
+ },
10
+ "state_file": "policy_postprocessor_step_0_groot_action_unpack_unnormalize_v1.safetensors"
11
+ },
12
+ {
13
+ "registry_name": "device_processor",
14
+ "config": {
15
+ "device": "cpu",
16
+ "float_dtype": null
17
+ }
18
+ }
19
+ ]
20
+ }
policy_postprocessor_step_0_groot_action_unpack_unnormalize_v1.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e6ee3d991a8593333d78e4d4381910528f2ea4c126bb7953fc47ccf574d48dd
3
+ size 9488
policy_preprocessor.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "groot_pack_inputs_v3",
16
+ "config": {
17
+ "state_horizon": 1,
18
+ "action_horizon": 16,
19
+ "max_state_dim": 64,
20
+ "max_action_dim": 32,
21
+ "language_key": "task",
22
+ "formalize_language": false,
23
+ "embodiment_tag": "new_embodiment",
24
+ "embodiment_mapping": {
25
+ "new_embodiment": 31,
26
+ "oxe_droid": 17,
27
+ "agibot_genie1": 26,
28
+ "gr1": 24,
29
+ "so100": 2,
30
+ "unitree_g1": 3
31
+ },
32
+ "normalize_min_max": true
33
+ },
34
+ "state_file": "policy_preprocessor_step_2_groot_pack_inputs_v3.safetensors"
35
+ },
36
+ {
37
+ "registry_name": "groot_eagle_encode_v3",
38
+ "config": {}
39
+ },
40
+ {
41
+ "registry_name": "groot_eagle_collate_v3",
42
+ "config": {}
43
+ },
44
+ {
45
+ "registry_name": "device_processor",
46
+ "config": {
47
+ "device": "cuda",
48
+ "float_dtype": null
49
+ }
50
+ }
51
+ ]
52
+ }
policy_preprocessor_step_2_groot_pack_inputs_v3.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e6ee3d991a8593333d78e4d4381910528f2ea4c126bb7953fc47ccf574d48dd
3
+ size 9488
train_config.json ADDED
@@ -0,0 +1,209 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "groot",
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": "your-hf-user/groot-finetune",
114
+ "private": null,
115
+ "tags": null,
116
+ "license": null,
117
+ "pretrained_path": null,
118
+ "chunk_size": 16,
119
+ "n_action_steps": 16,
120
+ "max_state_dim": 64,
121
+ "max_action_dim": 32,
122
+ "normalization_mapping": {
123
+ "VISUAL": "IDENTITY",
124
+ "STATE": "MEAN_STD",
125
+ "ACTION": "MEAN_STD"
126
+ },
127
+ "image_size": [
128
+ 224,
129
+ 224
130
+ ],
131
+ "base_model_path": "nvidia/GR00T-N1.5-3B",
132
+ "tokenizer_assets_repo": "lerobot/eagle2hg-processor-groot-n1p5",
133
+ "embodiment_tag": "new_embodiment",
134
+ "tune_llm": false,
135
+ "tune_visual": true,
136
+ "tune_projector": true,
137
+ "tune_diffusion_model": true,
138
+ "lora_rank": 0,
139
+ "lora_alpha": 16,
140
+ "lora_dropout": 0.1,
141
+ "lora_full_model": false,
142
+ "optimizer_lr": 0.0001,
143
+ "optimizer_betas": [
144
+ 0.95,
145
+ 0.999
146
+ ],
147
+ "optimizer_eps": 1e-08,
148
+ "optimizer_weight_decay": 1e-05,
149
+ "warmup_ratio": 0.05,
150
+ "use_bf16": true,
151
+ "video_backend": "decord",
152
+ "balance_dataset_weights": true,
153
+ "balance_trajectory_weights": true,
154
+ "dataset_paths": null,
155
+ "output_dir": "./tmp/gr00t",
156
+ "save_steps": 1000,
157
+ "max_steps": 10000,
158
+ "batch_size": 32,
159
+ "dataloader_num_workers": 8,
160
+ "report_to": "wandb",
161
+ "resume": false
162
+ },
163
+ "output_dir": "outputs/groot_visual_expert_tune_20251120_072554",
164
+ "job_name": "groot_visual_expert_tune",
165
+ "resume": false,
166
+ "seed": 1000,
167
+ "num_workers": 8,
168
+ "batch_size": 32,
169
+ "steps": 16000,
170
+ "eval_freq": 20000,
171
+ "log_freq": 100,
172
+ "save_checkpoint": true,
173
+ "save_freq": 4000,
174
+ "use_policy_training_preset": true,
175
+ "optimizer": {
176
+ "type": "adamw",
177
+ "lr": 0.0001,
178
+ "weight_decay": 1e-05,
179
+ "grad_clip_norm": 10.0,
180
+ "betas": [
181
+ 0.95,
182
+ 0.999
183
+ ],
184
+ "eps": 1e-08
185
+ },
186
+ "scheduler": {
187
+ "type": "cosine_decay_with_warmup",
188
+ "num_warmup_steps": 500,
189
+ "num_decay_steps": 10000,
190
+ "peak_lr": 0.0001,
191
+ "decay_lr": 1e-05
192
+ },
193
+ "eval": {
194
+ "n_episodes": 50,
195
+ "batch_size": 50,
196
+ "use_async_envs": false
197
+ },
198
+ "wandb": {
199
+ "enable": true,
200
+ "disable_artifact": false,
201
+ "project": "lerobot",
202
+ "entity": null,
203
+ "notes": null,
204
+ "run_id": "ty9tx7ax",
205
+ "mode": null
206
+ },
207
+ "checkpoint_path": null,
208
+ "rename_map": {}
209
+ }