Catalano.MachineLearning.Classification

Class SparseMulticlassSupportVectorMachine

java.lang.Object

Catalano.MachineLearning.Classification.SparseMulticlassSupportVectorMachine

All Implemented Interfaces:

java.io.Serializable

public class SparseMulticlassSupportVectorMachine
extends java.lang.Object
implements java.io.Serializable

Sparse Multiclass Support Vector Machine.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
SparseMulticlassSupportVectorMachine(IMercerKernel kernel, double c, double[] weight) Initializes a new instance of the SparseMulticlassSupportVectorMachine class.
SparseMulticlassSupportVectorMachine(IMercerKernel kernel, double c, double[] weight, SVM.Strategy strategy) Initializes a new instance of the SparseMulticlassSupportVectorMachine class.
SparseMulticlassSupportVectorMachine(IMercerKernel kernel, double c, int numberOfClasses) Initializes a new instance of the SparseMulticlassSupportVectorMachine class.
SparseMulticlassSupportVectorMachine(IMercerKernel kernel, double c, int numberOfClasses, SVM.Strategy strategy) Initializes a new instance of the SparseMulticlassSupportVectorMachine class.

Method Summary

Modifier and Type	Method and Description
void	Finish() Process support vectors until convergence.
void	Learn(SparseArray[] input, int[] output)
void	Learn(SparseArray input, int output) Online learn.
int	Predict(SparseArray feature)

Methods inherited from class java.lang.Object

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

Constructor Detail

SparseMulticlassSupportVectorMachine

public SparseMulticlassSupportVectorMachine(IMercerKernel kernel,
                                            double c,
                                            int numberOfClasses)

Initializes a new instance of the SparseMulticlassSupportVectorMachine class.

Parameters:

kernel - Mercer kernel.

c - Soft margin penalty parameter.

numberOfClasses - Number of classes.

SparseMulticlassSupportVectorMachine

public SparseMulticlassSupportVectorMachine(IMercerKernel kernel,
                                            double c,
                                            int numberOfClasses,
                                            SVM.Strategy strategy)

Initializes a new instance of the SparseMulticlassSupportVectorMachine class.

Parameters:

kernel - Mercer kernel.

c - Soft margin penalty parameter.

numberOfClasses - Number of classes.

strategy - Strategy.

SparseMulticlassSupportVectorMachine

public SparseMulticlassSupportVectorMachine(IMercerKernel kernel,
                                            double c,
                                            double[] weight)

Initializes a new instance of the SparseMulticlassSupportVectorMachine class.

Parameters:

kernel - Mercer kernel.

c - Soft margin penalty parameter.

weight - Class weight.

SparseMulticlassSupportVectorMachine

public SparseMulticlassSupportVectorMachine(IMercerKernel kernel,
                                            double c,
                                            double[] weight,
                                            SVM.Strategy strategy)

Initializes a new instance of the SparseMulticlassSupportVectorMachine class.

Parameters:

kernel - Mercer kernel.

c - Soft margin penalty parameter.

weight - Class weight.

strategy - Strategy.

Method Detail

Learn

public void Learn(SparseArray[] input,
                  int[] output)

Learn

public void Learn(SparseArray input,
                  int output)

Online learn.

Parameters:

input - Feature.

output - Label.

Predict

public int Predict(SparseArray feature)

Finish

public void Finish()

Process support vectors until convergence.