klyn.math.nn.losses - Klyn API

Package klyn.math.nn.losses

Contains 6 type(s).

Types

Interfaces

ClassificationLoss - Mean classification loss with integer target labels.
RegressionLoss - Mean regression objective consuming targets in the model's floating precision.
SupervisedLoss - Typed mean supervised objective.

Classes

LossResult - A mean batch loss and its derivative with respect to model outputs.
MeanSquaredError - Mean squared residual over all samples and output coordinates.
SparseCategoricalCrossentropy - Mean multiclass cross-entropy from logits or normalized probabilities and integer labels.