xmindflow / deformableLKA

[WACV 2024] Beyond Self-Attention: Deformable Large Kernel Attention for Medical Image Segmentation
https://arxiv.org/abs/2309.00121
186 stars 14 forks source link

It seems that Deformable Large Kernel Attention is not used in the train_pancreas.py code, but EPA is used. #6

Closed liaochuanlin closed 1 year ago

Leonngm commented 1 year ago

Hi, you can select the attention block for the network by specifying it during the creation of the network. Please have a look at this file:

https://github.com/xmindflow/deformableLKA/blob/main/3D/pancreas_code/train_pancreas.py#L114

The default attention mechanism is deformable LKA.