sanghyun-son / EDSR-PyTorch

PyTorch version of the paper 'Enhanced Deep Residual Networks for Single Image Super-Resolution' (CVPRW 2017)
MIT License
2.42k stars 668 forks source link

How to reproduce Track2(unknown downscaling) result? #262

Open ElonZhang0328 opened 4 years ago

ElonZhang0328 commented 4 years ago

Hi,

Thank you for your work!

It seems like that pre-trained models are all only for bicubic track. How can I reproduce the result on Track2(unknown) of paper? According the paper, the training method and network architecture of Track2 are different with those of Track1, but I don't find the difference in the code.

Could you please give me a little guidance? Thank you very much!