MHNet
MHNet copied to clipboard
Loss function
Your article has been especially helpful to me,However, it seems that CharbonnierLoss and EdgeLoss that you defined in loss.py are not reflected in train.py. We look forward to hearing from you soon~~
Your article has been especially helpful to me,However, it seems that CharbonnierLoss and EdgeLoss that you defined in loss.py are not reflected in train.py. We look forward to hearing from you soon~~
In this article, we only use PSNRloss. The reasons are as follows: https://github.com/megvii-research/NAFNet/issues/50#issuecomment-1195041449
May I ask how much the psnr loss value of the network generally drops during training?