Closed PigletOS closed 3 years ago
Still use the loss idea in the PFLD paper, but replace L1Loss with wing loss
Same here, I think there is loss calculation issues, it is always around 30. see this
https://github.com/github-luffy/PFLD_68points_Pytorch/issues/26
What is the final loss you get? I tried to reimplement PFLD using keras. When I was training, the loss saturated around 30 at the early beginning and the loss was still around 30 after 20k iterations.