mlquantify.losses

Loss Functions

BaseLoss	Base class for optimization losses.
DistanceLoss	Generic distance-based loss between two probability distributions.
LeastSquaresLoss	Squared Euclidean (least-squares) loss.
HellingerSurrogateLoss	Optimization surrogate for the squared Hellinger distance.
EnergyLoss	Energy-distance loss for distribution matching.
NegativeLogLikelihoodLoss	Negative log-likelihood loss for mixture likelihoods.
MixtureNegativeLogLikelihoodLoss	Negative log-likelihood for class likelihood mixtures.
RegularizedMixtureNLLLoss	Mixture NLL with optional ordinal-smoothness regularization.
normalize_distribution	Normalize an array to a valid probability distribution.
get_loss	Instantiate a loss object from a string identifier or return a callable.