Hugging Face's logo

nkmurst
/
dynamicvla_gorgeous_v1

Robotics
LeRobot
Safetensors
dynamicvla
Model card Files
xet
 Community
nkmurst commited on
Commit
ab84fd1
·
verified ·
1 Parent(s): 5c578ac

Upload checkpoint 005000

Browse files
Files changed (7) hide show
  1. checkpoints/005000/pretrained_model/config.json +116 -0
  2. checkpoints/005000/pretrained_model/model.safetensors +3 -0
  3. checkpoints/005000/pretrained_model/policy_postprocessor.json +32 -0
  4. checkpoints/005000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors +3 -0
  5. checkpoints/005000/pretrained_model/policy_preprocessor.json +86 -0
  6. checkpoints/005000/pretrained_model/policy_preprocessor_step_5_normalizer_processor.safetensors +3 -0
  7. checkpoints/005000/pretrained_model/train_config.json +252 -0
checkpoints/005000/pretrained_model/config.json ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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_v1",
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": true,
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": false,
76
+ "freeze_connector": false,
77
+ "freeze_text_model": false,
78
+ "train_state_proj": true,
79
+ "optimizer_lr": 0.0001,
80
+ "optimizer_betas": [
81
+ 0.9,
82
+ 0.95
83
+ ],
84
+ "optimizer_eps": 1e-08,
85
+ "optimizer_weight_decay": 1e-10,
86
+ "optimizer_grad_clip_norm": 10.0,
87
+ "scheduler_warmup_steps": 1000,
88
+ "scheduler_decay_steps": 30000,
89
+ "scheduler_decay_lr": 2.5e-06,
90
+ "attention_mode": "cross_attn",
91
+ "prefix_length": -1,
92
+ "pad_language_to": "longest",
93
+ "num_expert_layers": -1,
94
+ "num_expert_skip_layers": 0,
95
+ "vlm_model_name": "HuggingFaceTB/SmolLM2-360M",
96
+ "num_vlm_layers": 16,
97
+ "smolvlm_patch_size": 16,
98
+ "smolvlm_attention_heads": 12,
99
+ "smolvlm_hidden_size": 768,
100
+ "smolvlm_intermediate_size": 3072,
101
+ "fastvlm_inference_mode": true,
102
+ "self_attn_every_n_layers": 2,
103
+ "expert_width_multiplier": 0.75,
104
+ "min_period": 0.004,
105
+ "max_period": 4.0,
106
+ "delta_timestamps": {
107
+ "observation": [
108
+ -2,
109
+ 0
110
+ ]
111
+ },
112
+ "observation_delta_random_range": [
113
+ -6,
114
+ -1
115
+ ]
116
+ }
checkpoints/005000/pretrained_model/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a22eea1fc18bf76cde2fd6dcfdc4a25992345676b4e78ddfca3a9e5767bb1c21
3
+ size 1118677616
checkpoints/005000/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/005000/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/005000/pretrained_model/policy_preprocessor.json ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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_new_line_processor",
16
+ "config": {}
17
+ },
18
+ {
19
+ "registry_name": "tokenizer_processor",
20
+ "config": {
21
+ "max_length": 48,
22
+ "task_key": "task",
23
+ "padding_side": "right",
24
+ "padding": "longest",
25
+ "truncation": true
26
+ }
27
+ },
28
+ {
29
+ "registry_name": "device_processor",
30
+ "config": {
31
+ "device": "cuda",
32
+ "float_dtype": null
33
+ }
34
+ },
35
+ {
36
+ "registry_name": "normalizer_processor",
37
+ "config": {
38
+ "eps": 1e-08,
39
+ "features": {
40
+ "observation.state": {
41
+ "type": "STATE",
42
+ "shape": [
43
+ 12
44
+ ]
45
+ },
46
+ "observation.images.left_top": {
47
+ "type": "VISUAL",
48
+ "shape": [
49
+ 3,
50
+ 480,
51
+ 640
52
+ ]
53
+ },
54
+ "observation.images.left_left": {
55
+ "type": "VISUAL",
56
+ "shape": [
57
+ 3,
58
+ 480,
59
+ 640
60
+ ]
61
+ },
62
+ "observation.images.left_right": {
63
+ "type": "VISUAL",
64
+ "shape": [
65
+ 3,
66
+ 480,
67
+ 640
68
+ ]
69
+ },
70
+ "action": {
71
+ "type": "ACTION",
72
+ "shape": [
73
+ 12
74
+ ]
75
+ }
76
+ },
77
+ "norm_map": {
78
+ "VISUAL": "IDENTITY",
79
+ "STATE": "MEAN_STD",
80
+ "ACTION": "MEAN_STD"
81
+ }
82
+ },
83
+ "state_file": "policy_preprocessor_step_5_normalizer_processor.safetensors"
84
+ }
85
+ ]
86
+ }
checkpoints/005000/pretrained_model/policy_preprocessor_step_5_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/005000/pretrained_model/train_config.json ADDED
@@ -0,0 +1,252 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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_v1",
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": true,
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": false,
158
+ "freeze_connector": false,
159
+ "freeze_text_model": false,
160
+ "train_state_proj": true,
161
+ "optimizer_lr": 0.0001,
162
+ "optimizer_betas": [
163
+ 0.9,
164
+ 0.95
165
+ ],
166
+ "optimizer_eps": 1e-08,
167
+ "optimizer_weight_decay": 1e-10,
168
+ "optimizer_grad_clip_norm": 10.0,
169
+ "scheduler_warmup_steps": 1000,
170
+ "scheduler_decay_steps": 30000,
171
+ "scheduler_decay_lr": 2.5e-06,
172
+ "attention_mode": "cross_attn",
173
+ "prefix_length": -1,
174
+ "pad_language_to": "longest",
175
+ "num_expert_layers": -1,
176
+ "num_expert_skip_layers": 0,
177
+ "vlm_model_name": "HuggingFaceTB/SmolLM2-360M",
178
+ "num_vlm_layers": 16,
179
+ "smolvlm_patch_size": 16,
180
+ "smolvlm_attention_heads": 12,
181
+ "smolvlm_hidden_size": 768,
182
+ "smolvlm_intermediate_size": 3072,
183
+ "fastvlm_inference_mode": true,
184
+ "self_attn_every_n_layers": 2,
185
+ "expert_width_multiplier": 0.75,
186
+ "min_period": 0.004,
187
+ "max_period": 4.0,
188
+ "delta_timestamps": {
189
+ "observation": [
190
+ -2,
191
+ 0
192
+ ]
193
+ },
194
+ "observation_delta_random_range": [
195
+ -6,
196
+ -1
197
+ ]
198
+ },
199
+ "output_dir": "outputs/train/dynamicvla_gorgeous_v1",
200
+ "job_name": "dynamicvla_gorgeous_v1",
201
+ "resume": false,
202
+ "seed": 1000,
203
+ "num_workers": 4,
204
+ "batch_size": 8,
205
+ "steps": 50000,
206
+ "eval_freq": 0,
207
+ "log_freq": 50,
208
+ "tolerance_s": 0.0001,
209
+ "save_checkpoint": true,
210
+ "save_freq": 2500,
211
+ "use_policy_training_preset": true,
212
+ "optimizer": {
213
+ "type": "adamw",
214
+ "lr": 0.0001,
215
+ "weight_decay": 1e-10,
216
+ "grad_clip_norm": 10.0,
217
+ "betas": [
218
+ 0.9,
219
+ 0.95
220
+ ],
221
+ "eps": 1e-08
222
+ },
223
+ "scheduler": {
224
+ "type": "cosine_decay_with_warmup",
225
+ "num_warmup_steps": 1000,
226
+ "num_decay_steps": 30000,
227
+ "peak_lr": 0.0001,
228
+ "decay_lr": 2.5e-06
229
+ },
230
+ "eval": {
231
+ "n_episodes": 50,
232
+ "batch_size": 50,
233
+ "use_async_envs": false
234
+ },
235
+ "wandb": {
236
+ "enable": false,
237
+ "disable_artifact": false,
238
+ "project": "lerobot",
239
+ "entity": null,
240
+ "notes": null,
241
+ "run_id": null,
242
+ "mode": null
243
+ },
244
+ "peft": null,
245
+ "use_rabc": false,
246
+ "rabc_progress_path": null,
247
+ "rabc_kappa": 0.01,
248
+ "rabc_epsilon": 1e-06,
249
+ "rabc_head_mode": "sparse",
250
+ "rename_map": {},
251
+ "checkpoint_path": null
252
+ }