jsat.regression

Class RidgeRegression

java.lang.Object

jsat.regression.RidgeRegression

All Implemented Interfaces:

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

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

An implementation of Ridge Regression that finds the exact solution. Ridge Regression is equivalent to MultipleLinearRegression with an added L₂ penalty for the weight vector.

Two different methods of finding the solution can be used. This algorithm should be used only for small dimensions problems with a reasonable number of example points.
For large dimension sparse problems, or dense problems with many data points (or both), use the StochasticRidgeRegression. For small data sets that pose non-linear problems, you can also use KernelRidgeRegression

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	RidgeRegression.SolverMode Sets which solver to use

Constructor Summary

Constructors

Constructor and Description
RidgeRegression()
RidgeRegression(double regularization)
RidgeRegression(double regularization, RidgeRegression.SolverMode mode)

Method Summary

Modifier and Type	Method and Description
RidgeRegression	clone()
double	getLambda() Returns the regularization constant in use
RidgeRegression.SolverMode	getSolverMode() Returns the solver in use
double	regress(DataPoint data)
void	setLambda(double lambda) Sets the regularization parameter used.
void	setSolverMode(RidgeRegression.SolverMode mode) Sets which solver is to be 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

RidgeRegression

public RidgeRegression()

RidgeRegression

public RidgeRegression(double regularization)

RidgeRegression

public RidgeRegression(double regularization,
                       RidgeRegression.SolverMode mode)

Method Detail

setLambda

public void setLambda(double lambda)

Sets the regularization parameter used.

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

setSolverMode

public void setSolverMode(RidgeRegression.SolverMode mode)

Sets which solver is to be used

Parameters:

mode - the solver mode to use

getSolverMode

public RidgeRegression.SolverMode getSolverMode()

Returns the solver in use

Returns:

the solver to 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 RidgeRegression clone()

Specified by:

clone in interface Regressor

Overrides:

clone in class java.lang.Object