Skip to content
This repository has been archived by the owner on Sep 9, 2024. It is now read-only.

why during validation not using model.eval()? #51

Open
Ponytai1j opened this issue Feb 19, 2022 · 1 comment
Open

why during validation not using model.eval()? #51

Ponytai1j opened this issue Feb 19, 2022 · 1 comment

Comments

@Ponytai1j
Copy link

Ponytai1j commented Feb 19, 2022

Since some model have BN,dropout layers, why dont use model.eval() in the val function?

am really confused, hope someone can help me. thx in advance.

@Ponytai1j Ponytai1j changed the title why in the validation not using model.eval()? why during validation not using model.eval()? Apr 1, 2022
@Neymar1011
Copy link

Hi , maybe the val function do not effect the training result.
but i think it's necessary to use model.eval() before validation.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants