Hugging Face's logo

nkmurst
/
dynamicvla_gorgeous_v2

Robotics
LeRobot
Safetensors
dynamicvla
Model card Files
xet
 Community
nkmurst commited on
Commit
e099adc
·
verified ·
1 Parent(s): fce3729

Upload checkpoint 040000

Browse files
Files changed (7) hide show
  1. checkpoints/040000/pretrained_model/config.json +119 -0
  2. checkpoints/040000/pretrained_model/model.safetensors +3 -0
  3. checkpoints/040000/pretrained_model/policy_postprocessor.json +32 -0
  4. checkpoints/040000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors +3 -0
  5. checkpoints/040000/pretrained_model/policy_preprocessor.json +102 -0
  6. checkpoints/040000/pretrained_model/policy_preprocessor_step_6_normalizer_processor.safetensors +3 -0
  7. checkpoints/040000/pretrained_model/train_config.json +255 -0
checkpoints/040000/pretrained_model/config.json ADDED
@@ -0,0 +1,119 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "dynamicvla",
3
+ "n_obs_steps": 2,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 12
9
+ ]
10
+ },
11
+ "observation.images.left_top": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 480,
16
+ 640
17
+ ]
18
+ },
19
+ "observation.images.left_left": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 480,
24
+ 640
25
+ ]
26
+ },
27
+ "observation.images.left_right": {
28
+ "type": "VISUAL",
29
+ "shape": [
30
+ 3,
31
+ 480,
32
+ 640
33
+ ]
34
+ }
35
+ },
36
+ "output_features": {
37
+ "action": {
38
+ "type": "ACTION",
39
+ "shape": [
40
+ 12
41
+ ]
42
+ }
43
+ },
44
+ "device": "cuda",
45
+ "use_amp": false,
46
+ "use_peft": false,
47
+ "push_to_hub": true,
48
+ "repo_id": "nkmurst/dynamicvla_gorgeous_v2",
49
+ "private": null,
50
+ "tags": null,
51
+ "license": null,
52
+ "pretrained_path": "hzxie/dynamic-vla",
53
+ "chunk_size": 20,
54
+ "n_action_steps": 20,
55
+ "normalization_mapping": {
56
+ "VISUAL": "IDENTITY",
57
+ "STATE": "MEAN_STD",
58
+ "ACTION": "MEAN_STD"
59
+ },
60
+ "max_state_dim": 32,
61
+ "max_action_dim": 32,
62
+ "resize_imgs_with_padding": [
63
+ 384,
64
+ 384
65
+ ],
66
+ "empty_cameras": 0,
67
+ "adapt_to_pi_aloha": false,
68
+ "use_delta_joint_actions_aloha": false,
69
+ "use_delta_action": false,
70
+ "enable_streaming": false,
71
+ "temporal_fusion": "attn",
72
+ "tokenizer_max_length": 48,
73
+ "num_steps": 10,
74
+ "use_cache": true,
75
+ "freeze_vision_model": true,
76
+ "freeze_connector": true,
77
+ "freeze_text_model": true,
78
+ "train_state_proj": true,
79
+ "reinitialize_action_expert": false,
80
+ "reinitialize_state_proj": false,
81
+ "optimizer_lr": 0.0001,
82
+ "optimizer_betas": [
83
+ 0.9,
84
+ 0.95
85
+ ],
86
+ "optimizer_eps": 1e-08,
87
+ "optimizer_weight_decay": 1e-10,
88
+ "optimizer_grad_clip_norm": 10.0,
89
+ "scheduler_warmup_steps": 1000,
90
+ "scheduler_decay_steps": 30000,
91
+ "scheduler_decay_lr": 2.5e-06,
92
+ "attention_mode": "cross_attn",
93
+ "prefix_length": -1,
94
+ "pad_language_to": "longest",
95
+ "num_expert_layers": -1,
96
+ "num_expert_skip_layers": 0,
97
+ "vlm_model_name": "HuggingFaceTB/SmolLM2-360M",
98
+ "num_vlm_layers": 16,
99
+ "smolvlm_patch_size": 16,
100
+ "smolvlm_attention_heads": 12,
101
+ "smolvlm_hidden_size": 768,
102
+ "smolvlm_intermediate_size": 3072,
103
+ "fastvlm_inference_mode": true,
104
+ "self_attn_every_n_layers": 2,
105
+ "expert_width_multiplier": 0.75,
106
+ "min_period": 0.004,
107
+ "max_period": 4.0,
108
+ "delta_timestamps": {
109
+ "observation": [
110
+ -2,
111
+ 0
112
+ ]
113
+ },
114
+ "observation_delta_random_range": [
115
+ -6,
116
+ -1
117
+ ],
118
+ "inference_observation_delta": null
119
+ }
checkpoints/040000/pretrained_model/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff46f67169a43ded61b7e5ab0aa5752d9b183ae04624c6119ef1be00c9f900c5
3
+ size 1118677616
checkpoints/040000/pretrained_model/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
+ 12
13
+ ]
14
+ }
15
+ },
16
+ "norm_map": {
17
+ "VISUAL": "IDENTITY",
18
+ "STATE": "MEAN_STD",
19
+ "ACTION": "MEAN_STD"
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
+ }
checkpoints/040000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:536034d07e6f399e3d4c907a6a0a2842ca3363d03a7217c67830afc6e439f2e8
3
+ size 9168
checkpoints/040000/pretrained_model/policy_preprocessor.json ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "dynamicvla_observation_history_processor",
16
+ "config": {
17
+ "feature_keys": [
18
+ "observation.state",
19
+ "observation.images.left_top",
20
+ "observation.images.left_left",
21
+ "observation.images.left_right"
22
+ ],
23
+ "delta_indices": [
24
+ -2,
25
+ 0
26
+ ]
27
+ }
28
+ },
29
+ {
30
+ "registry_name": "dynamicvla_new_line_processor",
31
+ "config": {}
32
+ },
33
+ {
34
+ "registry_name": "tokenizer_processor",
35
+ "config": {
36
+ "max_length": 48,
37
+ "task_key": "task",
38
+ "padding_side": "right",
39
+ "padding": "longest",
40
+ "truncation": true,
41
+ "tokenizer_name": "HuggingFaceTB/SmolLM2-360M"
42
+ }
43
+ },
44
+ {
45
+ "registry_name": "device_processor",
46
+ "config": {
47
+ "device": "cuda",
48
+ "float_dtype": null
49
+ }
50
+ },
51
+ {
52
+ "registry_name": "normalizer_processor",
53
+ "config": {
54
+ "eps": 1e-08,
55
+ "features": {
56
+ "observation.state": {
57
+ "type": "STATE",
58
+ "shape": [
59
+ 12
60
+ ]
61
+ },
62
+ "observation.images.left_top": {
63
+ "type": "VISUAL",
64
+ "shape": [
65
+ 3,
66
+ 480,
67
+ 640
68
+ ]
69
+ },
70
+ "observation.images.left_left": {
71
+ "type": "VISUAL",
72
+ "shape": [
73
+ 3,
74
+ 480,
75
+ 640
76
+ ]
77
+ },
78
+ "observation.images.left_right": {
79
+ "type": "VISUAL",
80
+ "shape": [
81
+ 3,
82
+ 480,
83
+ 640
84
+ ]
85
+ },
86
+ "action": {
87
+ "type": "ACTION",
88
+ "shape": [
89
+ 12
90
+ ]
91
+ }
92
+ },
93
+ "norm_map": {
94
+ "VISUAL": "IDENTITY",
95
+ "STATE": "MEAN_STD",
96
+ "ACTION": "MEAN_STD"
97
+ }
98
+ },
99
+ "state_file": "policy_preprocessor_step_6_normalizer_processor.safetensors"
100
+ }
101
+ ]
102
+ }
checkpoints/040000/pretrained_model/policy_preprocessor_step_6_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:536034d07e6f399e3d4c907a6a0a2842ca3363d03a7217c67830afc6e439f2e8
3
+ size 9168
checkpoints/040000/pretrained_model/train_config.json ADDED
@@ -0,0 +1,255 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "rook86/gorgeous-30hz-merged-v2v3",
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": "dynamicvla",
85
+ "n_obs_steps": 2,
86
+ "input_features": {
87
+ "observation.state": {
88
+ "type": "STATE",
89
+ "shape": [
90
+ 12
91
+ ]
92
+ },
93
+ "observation.images.left_top": {
94
+ "type": "VISUAL",
95
+ "shape": [
96
+ 3,
97
+ 480,
98
+ 640
99
+ ]
100
+ },
101
+ "observation.images.left_left": {
102
+ "type": "VISUAL",
103
+ "shape": [
104
+ 3,
105
+ 480,
106
+ 640
107
+ ]
108
+ },
109
+ "observation.images.left_right": {
110
+ "type": "VISUAL",
111
+ "shape": [
112
+ 3,
113
+ 480,
114
+ 640
115
+ ]
116
+ }
117
+ },
118
+ "output_features": {
119
+ "action": {
120
+ "type": "ACTION",
121
+ "shape": [
122
+ 12
123
+ ]
124
+ }
125
+ },
126
+ "device": "cuda",
127
+ "use_amp": false,
128
+ "use_peft": false,
129
+ "push_to_hub": true,
130
+ "repo_id": "nkmurst/dynamicvla_gorgeous_v2",
131
+ "private": null,
132
+ "tags": null,
133
+ "license": null,
134
+ "pretrained_path": "hzxie/dynamic-vla",
135
+ "chunk_size": 20,
136
+ "n_action_steps": 20,
137
+ "normalization_mapping": {
138
+ "VISUAL": "IDENTITY",
139
+ "STATE": "MEAN_STD",
140
+ "ACTION": "MEAN_STD"
141
+ },
142
+ "max_state_dim": 32,
143
+ "max_action_dim": 32,
144
+ "resize_imgs_with_padding": [
145
+ 384,
146
+ 384
147
+ ],
148
+ "empty_cameras": 0,
149
+ "adapt_to_pi_aloha": false,
150
+ "use_delta_joint_actions_aloha": false,
151
+ "use_delta_action": false,
152
+ "enable_streaming": false,
153
+ "temporal_fusion": "attn",
154
+ "tokenizer_max_length": 48,
155
+ "num_steps": 10,
156
+ "use_cache": true,
157
+ "freeze_vision_model": true,
158
+ "freeze_connector": true,
159
+ "freeze_text_model": true,
160
+ "train_state_proj": true,
161
+ "reinitialize_action_expert": false,
162
+ "reinitialize_state_proj": false,
163
+ "optimizer_lr": 0.0001,
164
+ "optimizer_betas": [
165
+ 0.9,
166
+ 0.95
167
+ ],
168
+ "optimizer_eps": 1e-08,
169
+ "optimizer_weight_decay": 1e-10,
170
+ "optimizer_grad_clip_norm": 10.0,
171
+ "scheduler_warmup_steps": 1000,
172
+ "scheduler_decay_steps": 30000,
173
+ "scheduler_decay_lr": 2.5e-06,
174
+ "attention_mode": "cross_attn",
175
+ "prefix_length": -1,
176
+ "pad_language_to": "longest",
177
+ "num_expert_layers": -1,
178
+ "num_expert_skip_layers": 0,
179
+ "vlm_model_name": "HuggingFaceTB/SmolLM2-360M",
180
+ "num_vlm_layers": 16,
181
+ "smolvlm_patch_size": 16,
182
+ "smolvlm_attention_heads": 12,
183
+ "smolvlm_hidden_size": 768,
184
+ "smolvlm_intermediate_size": 3072,
185
+ "fastvlm_inference_mode": true,
186
+ "self_attn_every_n_layers": 2,
187
+ "expert_width_multiplier": 0.75,
188
+ "min_period": 0.004,
189
+ "max_period": 4.0,
190
+ "delta_timestamps": {
191
+ "observation": [
192
+ -2,
193
+ 0
194
+ ]
195
+ },
196
+ "observation_delta_random_range": [
197
+ -6,
198
+ -1
199
+ ],
200
+ "inference_observation_delta": null
201
+ },
202
+ "output_dir": "outputs/train/dynamicvla_gorgeous_v2",
203
+ "job_name": "dynamicvla_gorgeous_v2",
204
+ "resume": false,
205
+ "seed": 1000,
206
+ "num_workers": 4,
207
+ "batch_size": 32,
208
+ "steps": 50000,
209
+ "eval_freq": 0,
210
+ "log_freq": 50,
211
+ "tolerance_s": 0.0001,
212
+ "save_checkpoint": true,
213
+ "save_freq": 2500,
214
+ "use_policy_training_preset": true,
215
+ "optimizer": {
216
+ "type": "adamw",
217
+ "lr": 0.0001,
218
+ "weight_decay": 1e-10,
219
+ "grad_clip_norm": 10.0,
220
+ "betas": [
221
+ 0.9,
222
+ 0.95
223
+ ],
224
+ "eps": 1e-08
225
+ },
226
+ "scheduler": {
227
+ "type": "cosine_decay_with_warmup",
228
+ "num_warmup_steps": 1000,
229
+ "num_decay_steps": 30000,
230
+ "peak_lr": 0.0001,
231
+ "decay_lr": 2.5e-06
232
+ },
233
+ "eval": {
234
+ "n_episodes": 50,
235
+ "batch_size": 50,
236
+ "use_async_envs": false
237
+ },
238
+ "wandb": {
239
+ "enable": false,
240
+ "disable_artifact": false,
241
+ "project": "lerobot",
242
+ "entity": null,
243
+ "notes": null,
244
+ "run_id": null,
245
+ "mode": null
246
+ },
247
+ "peft": null,
248
+ "use_rabc": false,
249
+ "rabc_progress_path": null,
250
+ "rabc_kappa": 0.01,
251
+ "rabc_epsilon": 1e-06,
252
+ "rabc_head_mode": "sparse",
253
+ "rename_map": {},
254
+ "checkpoint_path": null
255
+ }