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
How do you evaluate the speed? The speed may be the same if the framework merges bn layer automatically. But there is no reason that the model with bn is faster than merged.
model with batch normalization runs faster than merged
The text was updated successfully, but these errors were encountered: