losses (pmrf.losses)

Stateful metric modules for frequentist optimization.

These classes wrap pure mathematical loss functions into a :class:pmrf.Metric. All losses take the true and predict arrays as inputs, and return the loss value when called.

Classes

LogMSELoss([multioutput, name])	Log of Mean Squared Error (RMSE) metric.
MSELoss([multioutput, name])	Mean Squared Error (MSE) metric.
RMSELoss([multioutput, name])	Root Mean Squared Error (RMSE) metric.
L1Loss([multioutput, name])	L1 Loss (Mean Absolute Error) metric.
MAPELoss([multioutput, name])	Mean Absolute Percentage Error (MAPE) metric.
HuberLoss([delta, multioutput, name])	Huber loss metric.
HingeLoss([operator, weight, mask, ...])	Applies a one-sided constraint (hinge) before evaluating a base metric.