search

LiJunnan1992 / DivideMix

Code for paper: DivideMix: Learning with Noisy Labels as Semi-supervised Learning
MIT License
543 stars 84 forks source link

Warmup dataloader batch size double #4

Closed lfb-1 closed 4 years ago

lfb-1 commented 4 years ago

Hi, I wonder why in the warmup process, the dataloader batch size has to be doubled? Can I use the normal batch size?

Cheers

LiJunnan1992 commented 4 years ago

Yes you can totally use normal batch size. I double it just to speed up training.