mklovin commited on
Commit
bc643dd
·
verified ·
1 Parent(s): 1aa04e1

Upload SmolVLA policy BlockLifting_wb_icub (20k steps)

Browse files
config.json ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "smolvla",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 174
9
+ ]
10
+ },
11
+ "observation.images.left": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 240,
16
+ 320
17
+ ]
18
+ },
19
+ "observation.images.right": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 240,
24
+ 320
25
+ ]
26
+ },
27
+ "observation.images.front": {
28
+ "type": "VISUAL",
29
+ "shape": [
30
+ 3,
31
+ 240,
32
+ 320
33
+ ]
34
+ }
35
+ },
36
+ "output_features": {
37
+ "action": {
38
+ "type": "ACTION",
39
+ "shape": [
40
+ 20
41
+ ]
42
+ }
43
+ },
44
+ "device": "cuda",
45
+ "use_amp": true,
46
+ "use_peft": false,
47
+ "push_to_hub": false,
48
+ "repo_id": null,
49
+ "private": null,
50
+ "tags": null,
51
+ "license": null,
52
+ "pretrained_path": null,
53
+ "pretrained_revision": null,
54
+ "chunk_size": 60,
55
+ "n_action_steps": 60,
56
+ "normalization_mapping": {
57
+ "VISUAL": "IDENTITY",
58
+ "STATE": "MEAN_STD",
59
+ "ACTION": "MEAN_STD"
60
+ },
61
+ "max_state_dim": 181,
62
+ "max_action_dim": 32,
63
+ "resize_imgs_with_padding": [
64
+ 512,
65
+ 512
66
+ ],
67
+ "empty_cameras": 0,
68
+ "adapt_to_pi_aloha": false,
69
+ "use_delta_joint_actions_aloha": false,
70
+ "tokenizer_max_length": 48,
71
+ "num_steps": 10,
72
+ "use_cache": true,
73
+ "freeze_vision_encoder": true,
74
+ "train_expert_only": true,
75
+ "train_state_proj": true,
76
+ "optimizer_lr": 0.0001,
77
+ "optimizer_betas": [
78
+ 0.9,
79
+ 0.95
80
+ ],
81
+ "optimizer_eps": 1e-08,
82
+ "optimizer_weight_decay": 1e-10,
83
+ "optimizer_grad_clip_norm": 10,
84
+ "scheduler_warmup_steps": 1000,
85
+ "scheduler_decay_steps": 30000,
86
+ "scheduler_decay_lr": 2.5e-06,
87
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
88
+ "load_vlm_weights": true,
89
+ "add_image_special_tokens": false,
90
+ "attention_mode": "cross_attn",
91
+ "prefix_length": -1,
92
+ "pad_language_to": "longest",
93
+ "num_expert_layers": -1,
94
+ "num_vlm_layers": 16,
95
+ "self_attn_every_n_layers": 2,
96
+ "expert_width_multiplier": 0.75,
97
+ "min_period": 0.004,
98
+ "max_period": 4.0,
99
+ "rtc_config": null,
100
+ "compile_model": false,
101
+ "compile_mode": "max-autotune"
102
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc52d6c8c7be0e1fe4037fac90e85129c76773e808db4084b9d7328e620cd1b4
3
+ size 907284712
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
+ 20
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
+ }
policy_postprocessor_step_0_unnormalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a46de1683260da14bd1113e0b07c050554e90b505162feb49e4cdf2ad3f1d25
3
+ size 15208
policy_preprocessor.json ADDED
@@ -0,0 +1,90 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "smolvla_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
+ "tokenizer_name": "tokenizer"
27
+ },
28
+ "artifacts": {
29
+ "tokenizer_name": "tokenizer"
30
+ }
31
+ },
32
+ {
33
+ "registry_name": "device_processor",
34
+ "config": {
35
+ "device": "cuda",
36
+ "float_dtype": null
37
+ }
38
+ },
39
+ {
40
+ "registry_name": "normalizer_processor",
41
+ "config": {
42
+ "eps": 1e-08,
43
+ "features": {
44
+ "observation.state": {
45
+ "type": "STATE",
46
+ "shape": [
47
+ 174
48
+ ]
49
+ },
50
+ "observation.images.left": {
51
+ "type": "VISUAL",
52
+ "shape": [
53
+ 3,
54
+ 240,
55
+ 320
56
+ ]
57
+ },
58
+ "observation.images.right": {
59
+ "type": "VISUAL",
60
+ "shape": [
61
+ 3,
62
+ 240,
63
+ 320
64
+ ]
65
+ },
66
+ "observation.images.front": {
67
+ "type": "VISUAL",
68
+ "shape": [
69
+ 3,
70
+ 240,
71
+ 320
72
+ ]
73
+ },
74
+ "action": {
75
+ "type": "ACTION",
76
+ "shape": [
77
+ 20
78
+ ]
79
+ }
80
+ },
81
+ "norm_map": {
82
+ "VISUAL": "IDENTITY",
83
+ "STATE": "MEAN_STD",
84
+ "ACTION": "MEAN_STD"
85
+ }
86
+ },
87
+ "state_file": "policy_preprocessor_step_5_normalizer_processor.safetensors"
88
+ }
89
+ ]
90
+ }
policy_preprocessor_step_5_normalizer_processor.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73baf44a3cfcd00c21e950cffbc4785eae470b43554da44349fc277d061dcfc6
3
+ size 15224
tokenizer/chat_template.jinja ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ <|im_start|>{% for message in messages %}{{message['role'] | capitalize}}{% if message['content'][0]['type'] == 'image' %}{{':'}}{% else %}{{': '}}{% endif %}{% for line in message['content'] %}{% if line['type'] == 'text' %}{{line['text']}}{% elif line['type'] == 'image' %}{{ '<image>' }}{% endif %}{% endfor %}<end_of_utterance>
2
+ {% endfor %}{% if add_generation_prompt %}{{ 'Assistant:' }}{% endif %}
tokenizer/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer/tokenizer_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<|im_start|>",
5
+ "clean_up_tokenization_spaces": false,
6
+ "end_of_utterance_token": "<end_of_utterance>",
7
+ "eos_token": "<end_of_utterance>",
8
+ "errors": "replace",
9
+ "fake_image_token": "<fake_token_around_image>",
10
+ "global_image_token": "<global-img>",
11
+ "image_token": "<image>",
12
+ "is_local": false,
13
+ "legacy": false,
14
+ "model_max_length": 8192,
15
+ "model_specific_special_tokens": {
16
+ "end_of_utterance_token": "<end_of_utterance>",
17
+ "fake_image_token": "<fake_token_around_image>",
18
+ "global_image_token": "<global-img>",
19
+ "image_token": "<image>"
20
+ },
21
+ "pad_token": "<|im_end|>",
22
+ "processor_class": "SmolVLMProcessor",
23
+ "tokenizer_class": "GPT2Tokenizer",
24
+ "truncation_side": "left",
25
+ "unk_token": "<|endoftext|>",
26
+ "vocab_size": 49152
27
+ }
train_config.json ADDED
@@ -0,0 +1,305 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "local/local_BlickliftingPMM_v2_2026-10-29_15-43-20",
4
+ "repo_type": "dataset",
5
+ "root": "/home/icub/Desktop/mkloviiin/RAMA_MIGUEL/PMM_Final/data/local_BlickliftingPMM_v2_2026-10-29_15-43-20",
6
+ "episodes": null,
7
+ "exclude_episodes": null,
8
+ "image_transforms": {
9
+ "enable": false,
10
+ "max_num_transforms": 3,
11
+ "random_order": false,
12
+ "tfs": {
13
+ "brightness": {
14
+ "weight": 1.0,
15
+ "type": "ColorJitter",
16
+ "kwargs": {
17
+ "brightness": [
18
+ 0.8,
19
+ 1.2
20
+ ]
21
+ }
22
+ },
23
+ "contrast": {
24
+ "weight": 1.0,
25
+ "type": "ColorJitter",
26
+ "kwargs": {
27
+ "contrast": [
28
+ 0.8,
29
+ 1.2
30
+ ]
31
+ }
32
+ },
33
+ "saturation": {
34
+ "weight": 1.0,
35
+ "type": "ColorJitter",
36
+ "kwargs": {
37
+ "saturation": [
38
+ 0.5,
39
+ 1.5
40
+ ]
41
+ }
42
+ },
43
+ "hue": {
44
+ "weight": 1.0,
45
+ "type": "ColorJitter",
46
+ "kwargs": {
47
+ "hue": [
48
+ -0.05,
49
+ 0.05
50
+ ]
51
+ }
52
+ },
53
+ "sharpness": {
54
+ "weight": 1.0,
55
+ "type": "SharpnessJitter",
56
+ "kwargs": {
57
+ "sharpness": [
58
+ 0.5,
59
+ 1.5
60
+ ]
61
+ }
62
+ },
63
+ "affine": {
64
+ "weight": 1.0,
65
+ "type": "RandomAffine",
66
+ "kwargs": {
67
+ "degrees": [
68
+ -5.0,
69
+ 5.0
70
+ ],
71
+ "translate": [
72
+ 0.05,
73
+ 0.05
74
+ ]
75
+ }
76
+ }
77
+ }
78
+ },
79
+ "revision": null,
80
+ "use_imagenet_stats": true,
81
+ "video_backend": "torchcodec",
82
+ "return_uint8": false,
83
+ "depth_output_unit": "mm",
84
+ "streaming": false,
85
+ "eval_split": 0.0
86
+ },
87
+ "env": null,
88
+ "policy": {
89
+ "type": "smolvla",
90
+ "n_obs_steps": 1,
91
+ "input_features": {
92
+ "observation.state": {
93
+ "type": "STATE",
94
+ "shape": [
95
+ 174
96
+ ]
97
+ },
98
+ "observation.images.left": {
99
+ "type": "VISUAL",
100
+ "shape": [
101
+ 3,
102
+ 240,
103
+ 320
104
+ ]
105
+ },
106
+ "observation.images.right": {
107
+ "type": "VISUAL",
108
+ "shape": [
109
+ 3,
110
+ 240,
111
+ 320
112
+ ]
113
+ },
114
+ "observation.images.front": {
115
+ "type": "VISUAL",
116
+ "shape": [
117
+ 3,
118
+ 240,
119
+ 320
120
+ ]
121
+ }
122
+ },
123
+ "output_features": {
124
+ "action": {
125
+ "type": "ACTION",
126
+ "shape": [
127
+ 20
128
+ ]
129
+ }
130
+ },
131
+ "device": "cuda",
132
+ "use_amp": true,
133
+ "use_peft": false,
134
+ "push_to_hub": false,
135
+ "repo_id": null,
136
+ "private": null,
137
+ "tags": null,
138
+ "license": null,
139
+ "pretrained_path": null,
140
+ "pretrained_revision": null,
141
+ "chunk_size": 60,
142
+ "n_action_steps": 60,
143
+ "normalization_mapping": {
144
+ "VISUAL": "IDENTITY",
145
+ "STATE": "MEAN_STD",
146
+ "ACTION": "MEAN_STD"
147
+ },
148
+ "max_state_dim": 181,
149
+ "max_action_dim": 32,
150
+ "resize_imgs_with_padding": [
151
+ 512,
152
+ 512
153
+ ],
154
+ "empty_cameras": 0,
155
+ "adapt_to_pi_aloha": false,
156
+ "use_delta_joint_actions_aloha": false,
157
+ "tokenizer_max_length": 48,
158
+ "num_steps": 10,
159
+ "use_cache": true,
160
+ "freeze_vision_encoder": true,
161
+ "train_expert_only": true,
162
+ "train_state_proj": true,
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,
171
+ "scheduler_warmup_steps": 1000,
172
+ "scheduler_decay_steps": 30000,
173
+ "scheduler_decay_lr": 2.5e-06,
174
+ "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
175
+ "load_vlm_weights": true,
176
+ "add_image_special_tokens": false,
177
+ "attention_mode": "cross_attn",
178
+ "prefix_length": -1,
179
+ "pad_language_to": "longest",
180
+ "num_expert_layers": -1,
181
+ "num_vlm_layers": 16,
182
+ "self_attn_every_n_layers": 2,
183
+ "expert_width_multiplier": 0.75,
184
+ "min_period": 0.004,
185
+ "max_period": 4.0,
186
+ "rtc_config": null,
187
+ "compile_model": false,
188
+ "compile_mode": "max-autotune"
189
+ },
190
+ "reward_model": null,
191
+ "output_dir": "/home/icub/Desktop/mkloviiin/RAMA_MIGUEL/PMM_Final/output/train/BlockLifting_wb_icub",
192
+ "job_name": "icub_smolvla_mujoco_blocklifting_wb",
193
+ "resume": false,
194
+ "seed": 1000,
195
+ "cudnn_deterministic": false,
196
+ "num_workers": 32,
197
+ "batch_size": 22,
198
+ "prefetch_factor": 4,
199
+ "persistent_workers": true,
200
+ "dataloader_multiprocessing_context": "spawn",
201
+ "steps": 20000,
202
+ "env_eval_freq": 20000,
203
+ "log_freq": 30,
204
+ "eval_steps": 0,
205
+ "max_eval_samples": 0,
206
+ "tolerance_s": 0.0001,
207
+ "save_checkpoint": true,
208
+ "save_freq": 20000,
209
+ "checkpoint_format": "safetensors",
210
+ "use_policy_training_preset": true,
211
+ "optimizer": {
212
+ "type": "adamw",
213
+ "lr": 0.0001,
214
+ "weight_decay": 1e-10,
215
+ "grad_clip_norm": 10,
216
+ "betas": [
217
+ 0.9,
218
+ 0.95
219
+ ],
220
+ "eps": 1e-08
221
+ },
222
+ "scheduler": {
223
+ "type": "cosine_decay_with_warmup",
224
+ "num_warmup_steps": 1000,
225
+ "num_decay_steps": 30000,
226
+ "peak_lr": 0.0001,
227
+ "decay_lr": 2.5e-06
228
+ },
229
+ "parallelism": {
230
+ "dp_replicate": 1,
231
+ "dp_shard": 1,
232
+ "context_parallel": {
233
+ "ring_degree": 1,
234
+ "ulysses_degree": 1
235
+ },
236
+ "cfg_parallel": 1
237
+ },
238
+ "accelerator": {
239
+ "mixed_precision": "no",
240
+ "gradient_accumulation": {
241
+ "steps": 1
242
+ },
243
+ "fsdp": {
244
+ "reshard_after_forward": true,
245
+ "wrap_modules": null,
246
+ "min_num_params": null,
247
+ "cpu_offload": false,
248
+ "ignored_modules": null
249
+ },
250
+ "ddp": {
251
+ "find_unused_parameters": true,
252
+ "gradient_as_bucket_view": false,
253
+ "static_graph": false
254
+ },
255
+ "compile": {
256
+ "enabled": false,
257
+ "backend": "inductor",
258
+ "mode": null,
259
+ "regional": true
260
+ },
261
+ "activation_checkpointing": {
262
+ "mode": "none"
263
+ }
264
+ },
265
+ "eval": {
266
+ "n_episodes": 50,
267
+ "batch_size": 22,
268
+ "use_async_envs": true,
269
+ "recording": false,
270
+ "recording_repo_id": null,
271
+ "recording_private": false
272
+ },
273
+ "ema": {
274
+ "enable": false,
275
+ "decay": null,
276
+ "update_after_step": 0,
277
+ "inv_gamma": 1.0,
278
+ "power": 0.75,
279
+ "min_decay": 0.0,
280
+ "max_decay": 0.9999,
281
+ "use_for_eval": true
282
+ },
283
+ "wandb": {
284
+ "enable": true,
285
+ "disable_artifact": false,
286
+ "project": "icub_smolvla_blocklifting_wb",
287
+ "entity": null,
288
+ "notes": null,
289
+ "run_id": "z6sfbaby",
290
+ "mode": null,
291
+ "add_tags": true
292
+ },
293
+ "peft": null,
294
+ "job": {
295
+ "target": null,
296
+ "image": "huggingface/lerobot-gpu:latest",
297
+ "timeout": "2d",
298
+ "detach": false,
299
+ "tags": []
300
+ },
301
+ "save_checkpoint_to_hub": false,
302
+ "sample_weighting": null,
303
+ "rename_map": {},
304
+ "checkpoint_path": null
305
+ }