Does initializing the position embedding to pure zeros for ViT make it worse? #1669
Unanswered
Phuoc-Hoan-Le
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Has anyone tried initializing the position embedding to pure zeros for ViT?
Does initializing the position embedding to pure zeros make it better or worse than using trunc_normal_()?
Beta Was this translation helpful? Give feedback.
All reactions