[BUG] DINOv3 weight error reported #479
Description
🧠 Describe the Bug
DINOv3 weight error reported
import lightly_train
if __name__ == "__main__":
lightly_train.train_semantic_segmentation(
out="",
model="dinov3/vitl16-eomt",
data={
"train": {
"images": "", # Path to training images
"masks": "", # Path to training masks
},
"val": {
"images": "", # Path to validation images
"masks": "", # Path to validation masks
},
"classes": {
0: {"name": "background", "labels": [(0, 0, 0)]},
1: {"name": "rice", "labels": [(255, 255, 255)]},
},
"ignore_classes": [],
},
transform_args={"ignore_index": 255},
devices=1,
batch_size=8,
steps = 60000
)
This is my code. When I use the dinov3/vitl16-eomt model, the following errors will be reported, and I only use lvd1689m. But no errors will be reported when I use the dinov3/vits16-eomt model or dinov3/vitb16-eomt model. So I think there are problems with the code when using dinov3/vitl16-eomt model.
Traceback (most recent call last):
File "/home/user/cyshi_lx/lightly-train-main/lightly-train-main/train.py", line 4, in <module>
lightly_train.train_semantic_segmentation(
File "/home/user/.conda/envs/seg/lib/python3.11/site-packages/lightly_train/_commands/train_task.py", line 443, in train_semantic_segmentation
return _train_task(config_cls=SemanticSegmentationTrainTaskConfig, **locals())
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/user/.conda/envs/seg/lib/python3.11/site-packages/lightly_train/_commands/train_task.py", line 475, in _train_task
_train_task_from_config(config=config)
File "/home/user/.conda/envs/seg/lib/python3.11/site-packages/lightly_train/_commands/train_task.py", line 656, in _train_task_from_config
train_model = train_model_cls(
^^^^^^^^^^^^^^^^
File "/home/user/.conda/envs/seg/lib/python3.11/site-packages/lightly_train/_task_models/dinov3_eomt_semantic_segmentation/train_model.py", line 190, in __init__
self.model = DINOv3EoMTSemanticSegmentation(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/user/.conda/envs/seg/lib/python3.11/site-packages/lightly_train/_task_models/dinov3_eomt_semantic_segmentation/task_model.py", line 133, in __init__
backbone = DINOV3_PACKAGE.get_model(
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/user/.conda/envs/seg/lib/python3.11/site-packages/lightly_train/_models/dinov3/dinov3_package.py", line 177, in get_model
model = model_builder(**args)
^^^^^^^^^^^^^^^^^^^^^
File "/home/user/.conda/envs/seg/lib/python3.11/site-packages/lightly_train/_models/dinov3/dinov3_src/hub/backbones.py", line 361, in dinov3_vitl16
return _make_dinov3_vit(
^^^^^^^^^^^^^^^^^
File "/home/user/.conda/envs/seg/lib/python3.11/site-packages/lightly_train/_models/dinov3/dinov3_src/hub/backbones.py", line 154, in _make_dinov3_vit
model.load_state_dict(state_dict, strict=True)
File "/home/user/.conda/envs/seg/lib/python3.11/site-packages/torch/nn/modules/module.py", line 2581, in load_state_dict
raise RuntimeError(
RuntimeError: Error(s) in loading state_dict for DinoVisionTransformer:
Unexpected key(s) in state_dict: "local_cls_norm.weight", "local_cls_norm.bias".