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
In the function ln.train_model the score of a fold is returned as
score=np.min(history[return_metric])
This means that is monitor != return_metric we early stop according to monitor, but still take the minimum wrt return_metric. So early stopping doesn't make much sense.
In the function
ln.train_model
the score of a fold is returned asThis means that is
monitor != return_metric
we early stop according tomonitor
, but still take the minimum wrtreturn_metric
. So early stopping doesn't make much sense.A better solution would be
This may complicate matters with
optimal_chekpoint
, as there to have consistency one should havemetric == monitor
instead ofmetric == return_metric
The text was updated successfully, but these errors were encountered: