Closed shamuiscoding closed 3 years ago
I think I'm supposed to train on main_train.py first then move onto SR.py. Will close if this works
It is confirmed that issue 114's solution is indeed correct. The reason that I was experiencing the problem was because I ran the code on input image with torch 1.8.0 and torchvision 0.7.0. Even though I reinstalled the libraries with the correct version it had already jammed. When I tried training with SR.py on another image (with the correct versions installed) it worked like a charm.
Firstly, thank you for the upload of your code. It's a very interesting piece of research. I'm trying to use SinGAN for the SR task. As far as I understood, python SR.py ~~ is the code for training on the SR task (according to the description below)![image](https://user-images.githubusercontent.com/75746908/110567061-3fc5c280-8194-11eb-9b43-a2121f9c1b3d.png)
However, I'm encountering this error.
According to #114 (the only issue with the same problem as mine), it is a version issue. But, in my case, I'm using Python 3.6.13, Torch 1.4.0, and Torchvision 0.5.0 which are supposedly the correct version according to issue 114.
![image](https://user-images.githubusercontent.com/75746908/110567236-85828b00-8194-11eb-97bf-20ecb239c026.png)
Any ideas on how to fix this problem? Thanks in advance.