MCPcopy Create free account
hub / github.com/CDTrans/CDTrans / CrossEntropyLabelSmooth

Class CrossEntropyLabelSmooth

loss/softmax_loss.py:4–34  ·  view source on GitHub ↗

Cross entropy loss with label smoothing regularizer. Reference: Szegedy et al. Rethinking the Inception Architecture for Computer Vision. CVPR 2016. Equation: y = (1 - epsilon) * y + epsilon / K. Args: num_classes (int): number of classes. epsilon (float): weight.

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers 1

make_lossFunction · 0.85

Calls

no outgoing calls

Tested by

no test coverage detected