Pytorch Implementation Loss function Line 163 (main_train.py)

cszn / DnCNN

Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising (TIP, 2017)

https://cszn.github.io/

1.42k stars 534 forks source link

Open dagarshali opened 4 years ago

dagarshali commented 4 years ago

On line 163 we have (loss = criterion(model(batch_y), batch_x))

batch_y = clean + noise model(batch_y) = x - output of the network.

If so, shouldn't the loss function be loss = criterion(model(batch_y),noise)?

angry77cat commented 4 years ago

I think the model is designed to predict noise