Super-SlowMo icon indicating copy to clipboard operation
Super-SlowMo copied to clipboard

Published 20 hours ago • verified publisher icon MayankSingal

Network not converge ......

Open ChenxiTU opened this issue 5 years ago • 2 comments

Thank you for sharing your implementation. I test the code on Adobe 240-fps data with pytorch 0.4.1. It seems that network can not converge ...... Sorry, I'm a beginner, after checking the code, I could not find the bug causing this (maybe in dataloader.populateTrainList2 12 should be replaced by 10? But this also not help.)
If it is possible, could you give me some hints why network can not converge?

ChenxiTU avatar Oct 22 '18 07:10 ChenxiTU

Hello there! In my experience, the convergence of the training was extremely slow. The only way to know if the network is learning anything was to generate the frames for some random online video and see if the network is learning anything at all or not. Hope that helped.

Cheers.

MayankSingal avatar Oct 22 '18 08:10 MayankSingal

Thank you very much! I see.

ChenxiTU avatar Oct 22 '18 08:10 ChenxiTU