org.apache.ignite.ml.composition.boosting.loss

Class SquaredError

java.lang.Object

org.apache.ignite.ml.composition.boosting.loss.SquaredError

All Implemented Interfaces:

Serializable, Loss

public class SquaredError
extends Object
implements Loss

Represent error function as E(label, modelAnswer) = 1/N * (label - prediction)^2

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
SquaredError()

Method Summary

Modifier and Type	Method and Description
double	error(long sampleSize, double lb, double prediction) Error value for model answer.
double	gradient(long sampleSize, double lb, double prediction) Error gradient value for model answer.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

SquaredError

public SquaredError()

Method Detail

error

public double error(long sampleSize,
                    double lb,
                    double prediction)

Error value for model answer.

Specified by:

error in interface Loss

Parameters:

sampleSize - Sample size.

lb - Label.

prediction - Model answer.

Returns:

Error value.

gradient

public double gradient(long sampleSize,
                       double lb,
                       double prediction)

Error gradient value for model answer.

Specified by:

gradient in interface Loss

Parameters:

sampleSize - Sample size.

lb - Label.

prediction - Model answer.

Returns:

Error value.