fr.lip6.jkernelmachines.classifier

Class LaSVM<T>

java.lang.Object

fr.lip6.jkernelmachines.classifier.LaSVM<T>

All Implemented Interfaces:

Classifier<T>, KernelSVM<T>, OnlineClassifier<T>, java.io.Serializable, java.lang.Cloneable

public final class LaSVM<T>
extends java.lang.Object
implements KernelSVM<T>, java.io.Serializable, OnlineClassifier<T>

Kernel SVM classifier implementing the LaSVM algorithm

Fast Kernel Classifiers with Online and Active Learning
Antoine Bordes, Seyda Ertekin, Jason Weston and Léon Bottou
Journal of Machine Learning Research, 6:1579-1619, September 2005.

Author:

picard

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
LaSVM(Kernel<T> k) Constructor with specific kernel

Method Summary

Modifier and Type	Method and Description
LaSVM<T>	copy() Creates and returns a copy of this object.
double[]	getAlphas() Tells the array of support vector coefficients
double	getB() Tells the bias term
double	getC() Tells the hyperparamter C
int	getE() Tells the number of epochs used for training
Kernel<T>	getKernel() Tells the kernel used by this classifier
double	getTau() Tells the numerical precision
void	onlineTrain(TrainingSampleStream<T> stream) Train the classifier using a stream of TrainingSample sampled from the TrainingSampleStream until no sample can be drawn.
void	retrain()
void	setB(double b) Sets the bias term
void	setC(double c) Sets the hyperparameter C
void	setE(int e) Sets the number of epochs used for training
void	setKernel(Kernel<T> kernel) Sets the kernel used by this classifier
void	train(java.util.List<TrainingSample<T>> l) Replace the current training list and train the classifier
void	train(TrainingSample<T> t) Deprecated.
double	valueOf(T e) Computes the category of the provided example

Methods inherited from class java.lang.Object

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

Constructor Detail

LaSVM

public LaSVM(Kernel<T> k)

Constructor with specific kernel

Parameters:

k - the kernel used by this classifier

Method Detail

train

@Deprecated
public void train(TrainingSample<T> t)

Deprecated.

Description copied from interface: OnlineClassifier

Add a single example to the current training set and train the classifier

Specified by:

train in interface OnlineClassifier<T>

Parameters:

t - the training sample

train

public void train(java.util.List<TrainingSample<T>> l)

Description copied from interface: Classifier

Replace the current training list and train the classifier

Specified by:

train in interface Classifier<T>

Parameters:

l - list of training samples

onlineTrain

public void onlineTrain(TrainingSampleStream<T> stream)

Description copied from interface: OnlineClassifier

Train the classifier using a stream of TrainingSample sampled from the TrainingSampleStream until no sample can be drawn.

Specified by:

onlineTrain in interface OnlineClassifier<T>

Parameters:

stream - the TrainingSampleStream from which the samples are drawn.

retrain

public void retrain()

valueOf

public final double valueOf(T e)

Description copied from interface: Classifier

Computes the category of the provided example

Specified by:

valueOf in interface Classifier<T>

Parameters:

e - example

Returns:

>0. if e belongs to the category, <0. if not.

getC

public double getC()

Tells the hyperparamter C

Specified by:

getC in interface KernelSVM<T>

Returns:

C

setC

public void setC(double c)

Sets the hyperparameter C

Specified by:

setC in interface KernelSVM<T>

Parameters:

c - the hyperparameter C

getE

public int getE()

Tells the number of epochs used for training

Returns:

the number of epochs

setE

public void setE(int e)

Sets the number of epochs used for training

Parameters:

e - the number of epochs

getTau

public double getTau()

Tells the numerical precision

Returns:

the numerical precision

getAlphas

public double[] getAlphas()

Tells the array of support vector coefficients

Specified by:

getAlphas in interface KernelSVM<T>

Returns:

the array of support vector coefficients

getB

public double getB()

Tells the bias term

Returns:

the bias term

setB

public void setB(double b)

Sets the bias term

Parameters:

b - the new bias term

getKernel

public Kernel<T> getKernel()

Tells the kernel used by this classifier

Specified by:

getKernel in interface KernelSVM<T>

Returns:

the kernel used by this classifier

setKernel

public void setKernel(Kernel<T> kernel)

Sets the kernel used by this classifier

Specified by:

setKernel in interface KernelSVM<T>

Parameters:

kernel - the kernel to use

copy

public LaSVM<T> copy()
              throws java.lang.CloneNotSupportedException

Creates and returns a copy of this object.

Specified by:

copy in interface Classifier<T>

Throws:

java.lang.CloneNotSupportedException

See Also:

Object.clone()