JimChan

如果将新数据集的图片数量删到只剩3450张，就可以正常跑通，请问作者您知道原因吗？

> @chenlanzheng 您好！请问您曾经出现过的问题，训练每个epoch都会报错：NoneType object is not iterable，是怎么解决的呢？ 好像是torch版本那些不匹配

How to get the parsing result？There was no visualization at the end of the test。

> It was a while ago... But from a quick Google - doesn't `scipy.io.loadmat` work? I tried, but why did the visualization result is an all-black picture

How is the Ground-truth kernel obtained ，the second row in Figure 5 and the first row in Figure 6

 and what does the "Convolving all filters of G results with the SR Kernel K" means? The "K" is the Ground-truth kernel? K is another input of Generator?

> It is randomly generated, then we downscale a HR image to obtain the LR version. The details can be found in chapter 5.2 of the paper Could you tell...