pytorch-lanenet
pytorch-lanenet copied to clipboard
trainining got NAN
HI, when I trained the network on the examples, i got loss as NAN after first step, how can i fix it up?
I will see if I can get it fixed. I will not be able to get to it until November though, in the meantime maybe try this repo: https://github.com/harryhan618/LaneNet , it seems to be regularly updated.