Glossary: Deep Learning Basics
Loss function
Also written: loss
Turns the gap between predictions and true answers into one number; lower is better. Use MSELoss for numbers, BCEWithLogitsLoss for yes/no and CrossEntropyLoss for classes.
Glossary: Deep Learning Basics
Also written: loss
Turns the gap between predictions and true answers into one number; lower is better. Use MSELoss for numbers, BCEWithLogitsLoss for yes/no and CrossEntropyLoss for classes.