jsat.regression

Class KernelRidgeRegression

java.lang.Object

jsat.regression.KernelRidgeRegression

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, Parameterized, Regressor

public class KernelRidgeRegression
extends java.lang.Object
implements Regressor, Parameterized

A kernelized implementation of Ridge Regression. Ridge Regression is equivalent to MultipleLinearRegression with an added L₂ penalty for the weight vector.

This algorithm is very expensive to compute O(n³), where n is the number of training points.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
KernelRidgeRegression() Creates a new Kernel Ridge Regression learner that uses an RBF kernel
KernelRidgeRegression(double lambda, KernelTrick kernel) Creates a new Kernel Ridge Regression learner

Method Summary

Modifier and Type	Method and Description
KernelRidgeRegression	clone()
KernelTrick	getKernel() Returns the kernel in use
double	getLambda() Returns the regularization constant in use
static Distribution	guessLambda(DataSet d) Guesses the distribution to use for the λ parameter
double	regress(DataPoint data)
void	setKernel(KernelTrick k) Sets the kernel trick to use
void	setLambda(double lambda) Sets the regularization parameter used.
boolean	supportsWeightedData()
void	train(RegressionDataSet dataSet, boolean parallel)

Methods inherited from class java.lang.Object

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

Methods inherited from interface jsat.regression.Regressor

train

Methods inherited from interface jsat.parameters.Parameterized

getParameter, getParameters

Constructor Detail

KernelRidgeRegression

public KernelRidgeRegression()

Creates a new Kernel Ridge Regression learner that uses an RBF kernel

KernelRidgeRegression

public KernelRidgeRegression(double lambda,
                             KernelTrick kernel)

Creates a new Kernel Ridge Regression learner

Parameters:

lambda - the regularization parameter

kernel - the kernel to use

See Also:

setLambda(double)

Method Detail

guessLambda

public static Distribution guessLambda(DataSet d)

Guesses the distribution to use for the λ parameter

Parameters:

d - the dataset to get the guess for

Returns:

the guess for the λ parameter

setLambda

public void setLambda(double lambda)

Sets the regularization parameter used. The value of lambda depends on the data set and kernel used, with easier problems using smaller lambdas.

Parameters:

lambda - the positive regularization constant in (0, Inf)

getLambda

public double getLambda()

Returns the regularization constant in use

Returns:

the regularization constant in use

setKernel

public void setKernel(KernelTrick k)

Sets the kernel trick to use

Parameters:

k - the kernel to use

getKernel

public KernelTrick getKernel()

Returns the kernel in use

Returns:

the kernel in use

regress

public double regress(DataPoint data)

Specified by:

regress in interface Regressor

train

public void train(RegressionDataSet dataSet,
                  boolean parallel)

Specified by:

train in interface Regressor

supportsWeightedData

public boolean supportsWeightedData()

Specified by:

supportsWeightedData in interface Regressor

clone

public KernelRidgeRegression clone()

Specified by:

clone in interface Regressor

Overrides:

clone in class java.lang.Object