Skip to content

Glossary: Deep Learning Basics

Cross-entropy

Also written: CrossEntropyLoss, nn.CrossEntropyLoss

The loss for multi-class problems: −log of the probability given to the true class. `nn.CrossEntropyLoss` applies softmax internally and takes int64 class numbers as targets.

Where the book teaches it

Related terms

SoftmaxLogitBCE (binary cross-entropy)