eugenelet/Meta-rPPG

Duplicate few-shot updates of the encoder

Closed this issue · 1 comments

Thanks so much for releasing the code for the great paper!

At the moment the encoder seems to get updated twice during validation, judging from the following two loops:

for i in range(self.opt.fewshots):

for i in range(self.opt.fewshots):

Is it a bug? Am I missing anything?

Sorry for my misunderstanding, I have missed the generator term from the paper:
Screen Shot 2020-07-29 at 19 55 12