search

AngeLouCN / CaraNet

Context Axial Reverse Attention Network for Small Medical Objects Segmentation
462 stars 30 forks source link

After several experiments, the same computer on the test index is different. #21

Closed Progressiveyouth closed 1 year ago

Progressiveyouth commented 2 years ago

Hello, I am very sorry to have occupied your study time. I have a question to ask you! The graphics card of my desktop computer is 3090, I downloaded the source code on your homepage (Github), and conducted several experiments, adding the indexes of mDice and mIOU. After testing and comparing the results of many experiments, I found that for the test dataset Kvasir, the difference of mDice indexes was as high as 1.27 percent. For the other four test sets, the gap may be larger, which is not normal. Why is this? Sorry to trouble you. Sorry again for taking up your time!

AngeLouCN commented 2 years ago

Hi, Thank you for your interests. I guess you use the higher version PyTorch. You can try to add the align_corners = 'True' in Caranet.py where used the F.interpolate functions. It is caused by the PyTorch version change. Also, if you directly use some python library to calculate the mDice or mIOU, it could be difference. That's why I used the matlab code provided by PraNet.