search

yingkunwu / R-YOLOv4

This is a PyTorch-based R-YOLOv4 implementation which combines YOLOv4 model and loss function from R3Det for arbitrary oriented object detection.
114 stars 20 forks source link

更新后的权重文件 #62

Open liuzhan123456 opened 1 year ago

liuzhan123456 commented 1 year ago

您好，看到您代码更新了，更新后的权重文件可以发给我一份嘛

yingkunwu commented 1 year ago

感謝你的關注！我會在最近有空的時候更新，再麻煩你到時候查看

Boylad commented 9 months ago

Hi, Kunnnnethan I'm encountering a RuntimeError while trying to load a PyTorch model state_dict. Does the architecture of the model being loaded differ? image

yingkunwu commented 9 months ago

Hi, Yes sorry. After refactoring the code, we change the structure of the model a little bit. Therefore the old pretrained model can't be used. However if you don't mind using the previous code, you can use the v1 branch that is still compatible to that pretained model. Alternatively, you can also copy the previous model structure.