search

cwmok / DIRAC

This is the official Pytorch implementation of "Unsupervised Deformable Image Registration with Absent Correspondences in Pre-operative and Post-Recurrence Brain Tumor MRI Scans" (MICCAI 2022), written by Tony C. W. Mok and Albert C. S. Chung.
MIT License
37 stars 1 forks source link

Nan loss when training #18

Open sydat2701 opened 6 months ago

sydat2701 commented 6 months ago

Hi, thanks for your great job! Currently, I got a problem when training. My TRE loss always got NAN and the training log is as shown in the figure. Do you have any idea for this? I downloaded the dataset and only resample to your required size in paper (and did not do any other preprocessing steps as I saw that original BraTS dataset was run skull-stripped). Thank for your help! image

cwmok commented 6 months ago

Hi @sydat2701,

I never encounter this issues. Is the resulting warped image looks good after training?