You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
dear author, the transformer use the layer norm in stead of batch norm, is it possible to apply inplace abn to transformer-based models? or is there any way to lower those models' gpu memory? thanks.
The text was updated successfully, but these errors were encountered:
dear author, the transformer use the layer norm in stead of batch norm, is it possible to apply inplace abn to transformer-based models? or is there any way to lower those models' gpu memory? thanks.
The text was updated successfully, but these errors were encountered: