cross_entropy_sparse#
- class brainpy.losses.cross_entropy_sparse(predicts, targets)[source]#
Computes the softmax cross-entropy loss.
- Parameters:
predicts – (batch, …, #class) tensor of logits.
targets – (batch, …) integer tensor of label indexes in {0, …,#nclass-1} or just a single integer.
- Returns:
(batch, …) tensor of the cross-entropy for each entry.