search

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

把损失函数换成mse #70

Closed ssywzzc closed 4 years ago

ssywzzc commented 4 years ago

我把损失函数换成mse，代码main_train部分163行DLoader = DataLoader(dataset=DDataset, num_workers=4, drop_last=True, batch_size=batch_size, shuffle=True)，把num_workers=4改成0，训练时loss值显示0.0001甚至更小，和我用keras版本训练的loss值（0.0028左右）差距太大了