Hi,
Is there any way to specify that I want to perform early stopping based on the validation loss? Moreover, I’d like to specify the “patience” parameter that specifies how long to keep the training after the validation loss starts to increase to see if eventually starts decreasing again.
Thanks