weka.classifiers.functions.loss

Interface LossFunction

All Superinterfaces:

java.io.Serializable

All Known Implementing Classes:

ApproximateAbsoluteError, SquaredError

public interface LossFunction
extends java.io.Serializable

Interface implemented by loss functions for MLPRegressor and MLPClassifier.

Version:

$Revision: 10949 $

Author:

Eibe Frank (eibe@cs.waikato.ac.nz)

Method Summary

Modifier and Type	Method and Description
double	derivative(double pred, double actual) The derivative of the loss with respect to the predicted value
double	loss(double pred, double actual) Returns the loss.

Method Detail

loss

double loss(double pred,
            double actual)

Returns the loss.

Parameters:

pred - predicted target value

actual - actual target value

Returns:

the loss

derivative

double derivative(double pred,
                  double actual)

The derivative of the loss with respect to the predicted value

Parameters:

pred - predicted target value

actual - actual target value

Returns:

the value of the derivative