This repository has been archived by the owner on Apr 24, 2023. It is now read-only.

Why should lr need to multiply numgpus in make optimizer? #15

Open

pansanity666 opened this issue May 27, 2020 · 0 comments

pansanity666 commented May 27, 2020

i dont understand why during mutiGPUS trainning ,lr should multiply numgpus,thx!

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.