Kai Zhang

https://github.com/cszn/KAIR/blob/72e93351bca41d1b1f6a4c3e1957f5bffccc7101/models/network_usrnet_v1.py#L228

Change https://github.com/cszn/KAIR/blob/3eb3cc7776fa8c57e8ed7c71bfa8039beb4c6677/options/train_msrresnet_psnr.json#L65 The training speed of KAIR is slower than BasicSR by xintao, three possible reasons at least: 1, https://github.com/xinntao/BasicSR/blob/master/docs/DatasetPreparation.md 2, https://github.com/xinntao/BasicSR/blob/14bafa5e03468775544f8711d7da7a61dbb3d664/basicsr/train.py#L13 3, https://github.com/xinntao/BasicSR/blob/14bafa5e03468775544f8711d7da7a61dbb3d664/basicsr/train.py#L35

Just replace the path of json file, https://github.com/cszn/KAIR/blob/094ee96e2ad666659983de9669cb8c477cb40371/main_train_msrresnet_psnr.py#L32

Start the learning rate with 1e-4 if you do not use batch norm.

If you get familiar with KAIR, you can modify it to train your own model.

model.load_state_dict(torch.load(model_path), strict=True) # strict=False 再检查一下.

Change batch size https://github.com/cszn/KAIR/blob/6d4ff120435ca2e0f8d0cdfc8aa7e3c59f24d281/options/train_dpsr.json#L28

There is no unique solution, the two settings can give rise to similar results.

See https://github.com/cszn/IRCNN/issues/2#issuecomment-420604117

See https://github.com/cszn/DnCNN/tree/master/TrainingCodes