org.openimaj.ml.linear.learner.perceptron

Class KernelPerceptron<INDEPENDANT,DEPENDANT>

java.lang.Object

org.openimaj.ml.linear.learner.perceptron.KernelPerceptron<INDEPENDANT,DEPENDANT>

Type Parameters:

INDEPENDANT -

DEPENDANT -

All Implemented Interfaces:

OnlineLearner<INDEPENDANT,DEPENDANT>

Direct Known Subclasses:

MatrixKernelPerceptron

public abstract class KernelPerceptron<INDEPENDANT,DEPENDANT>
extends Object
implements OnlineLearner<INDEPENDANT,DEPENDANT>

Author:

Sina Samangooei (ss@ecs.soton.ac.uk)

Field Summary

Fields

Modifier and Type	Field and Description
protected int	errors

Constructor Summary

Constructors

Constructor and Description
KernelPerceptron()
KernelPerceptron(Kernel<INDEPENDANT> kernel)

Method Summary

Methods

Modifier and Type	Method and Description
abstract double	getBias()
int	getErrors()
abstract List<INDEPENDANT>	getSupports()
abstract List<Double>	getWeights()
void	process(INDEPENDANT xt, DEPENDANT yt)
abstract void	update(INDEPENDANT xt, DEPENDANT yt, DEPENDANT yt_prime) When there is an error in prediction, update somehow

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.openimaj.ml.linear.learner.OnlineLearner

predict

Field Detail

errors

protected int errors

Constructor Detail

KernelPerceptron

public KernelPerceptron()

KernelPerceptron

public KernelPerceptron(Kernel<INDEPENDANT> kernel)

Parameters:

kernel -

Method Detail

process

public void process(INDEPENDANT xt,
           DEPENDANT yt)

Specified by:

process in interface OnlineLearner<INDEPENDANT,DEPENDANT>

update

public abstract void update(INDEPENDANT xt,
          DEPENDANT yt,
          DEPENDANT yt_prime)

When there is an error in prediction, update somehow

Parameters:

xt -

yt -

yt_prime -

getSupports

public abstract List<INDEPENDANT> getSupports()

Returns:

the vectors that form the support

getWeights

public abstract List<Double> getWeights()

Returns:

the weights of the support vectors

getBias

public abstract double getBias()

Returns:

the bias

getErrors

public int getErrors()

Returns:

number of errors made