keel.Algorithms.Instance_Generation.LVQ

Class LVQGenerator

java.lang.Object

keel.Algorithms.Instance_Generation.Basic.PrototypeGenerator

keel.Algorithms.Instance_Generation.LVQ.LVQGenerator

Direct Known Subclasses:

LVQ1

public abstract class LVQGenerator
extends PrototypeGenerator

LVQGenerator

Abstract class parent of every LVQGenerator-type algorithm.

Version:

keel0.1

Author:

Diego J. Romero López

See Also:

Instance, InstanceSet

Field Summary

Fields

Modifier and Type	Field and Description
protected PrototypeSet	initial Initial Prototype set.
protected boolean	initialset If the initialset is set.
protected int	iterations Number of iterations of the algorithm.
protected int	numberOfPrototypesGenerated Number of prototypes of the resulting set.

Fields inherited from class keel.Algorithms.Instance_Generation.Basic.PrototypeGenerator

algorithmName, generatedDataSet, SEED, seedDefaultValueList, trainingDataSet

Constructor Summary

Constructors

Constructor and Description
LVQGenerator(PrototypeSet _trainingDataSet, int iterations, double percRed) Constructs a new LVQGenerator algorithm.
LVQGenerator(PrototypeSet _trainingDataSet, int iterations, int n) Constructs a new LVQGenerator algorithm.
LVQGenerator(PrototypeSet _trainingDataSet, Parameters parameters) Constructs a new LVQGenerator algorithm.
LVQGenerator(PrototypeSet InitialSet, PrototypeSet _trainingDataSet, int iterations, int n) WITH INITIAL CODE-BOOKS Constructs a new LVQGenerator algorithm.

Method Summary

Modifier and Type	Method and Description
protected abstract void	penalize(Prototype m, Prototype x) Applies a penalization to prototype m
PrototypeSet	reduceSet() Execute the method and returns the condensed prototype set.
protected abstract void	reward(Prototype m, Prototype x) Applies a reward to prototype m

Methods inherited from class keel.Algorithms.Instance_Generation.Basic.PrototypeGenerator

absoluteAccuracy, absoluteAccuracyAndError, absoluteAccuracyKNN, accuracy, accuracy2, desordenar_vector_sin, desordenar_vector, execute, generateReducedDataSet, getResultingAccuracy, getResultingAccuracy, getResults, getResultsOfAccuracy, getResultsOfAccuracy, getSeed, getSetSizeFromPercentage, getSetSizeFromPercentage, getTime, inic_vector_sin, inic_vector, main, saveResultsOfAccuracyIn, saveResultsOfAccuracyIn, selecRandomSet, setSeed, showResultsOfAccuracy, showResultsOfAccuracy

Methods inherited from class java.lang.Object

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

Field Detail

iterations

protected int iterations

Number of iterations of the algorithm.

numberOfPrototypesGenerated

protected int numberOfPrototypesGenerated

Number of prototypes of the resulting set.

initialset

protected boolean initialset

If the initialset is set.

initial

protected PrototypeSet initial

Initial Prototype set.

Constructor Detail

LVQGenerator

public LVQGenerator(PrototypeSet _trainingDataSet,
                    int iterations,
                    int n)

Constructs a new LVQGenerator algorithm. NOTE: this class is abstract.

Parameters:

_trainingDataSet -

iterations - Iterations that will execute the algorithm.

n - Number of prototypes to be returned.

LVQGenerator

public LVQGenerator(PrototypeSet InitialSet,
                    PrototypeSet _trainingDataSet,
                    int iterations,
                    int n)

WITH INITIAL CODE-BOOKS Constructs a new LVQGenerator algorithm. NOTE: this class is abstract.

Parameters:

InitialSet - Initial datset

_trainingDataSet - training dataset

iterations - Iterations that will execute the algorithm.

n - Number of prototypes to be returned.

LVQGenerator

public LVQGenerator(PrototypeSet _trainingDataSet,
                    int iterations,
                    double percRed)

Constructs a new LVQGenerator algorithm. NOTE: this class is abstract.

Parameters:

_trainingDataSet -

iterations - Iterations that will execute the algorithm.

percRed - Percentage of the training size that will have the reduced set.

LVQGenerator

public LVQGenerator(PrototypeSet _trainingDataSet,
                    Parameters parameters)

Constructs a new LVQGenerator algorithm. NOTE: this class is abstract.

Parameters:

_trainingDataSet -

parameters - Parameters of the algorithm (number of iterations [integer], % of reduced size [double in (0-100]])

Method Detail

reward

protected abstract void reward(Prototype m,
                               Prototype x)

Applies a reward to prototype m

Parameters:

m - Rewarded prototype. IT IS MODIFIED.

x - Nearest prototype to m.

penalize

protected abstract void penalize(Prototype m,
                                 Prototype x)

Applies a penalization to prototype m

Parameters:

m - Penalized prototype. IT IS MODIFIED.

x - Nearest prototype to m.

reduceSet

public PrototypeSet reduceSet()

Execute the method and returns the condensed prototype set.

Overrides:

reduceSet in class PrototypeGenerator

Returns:

Prototype set modified from the training instance set by a LVQGenerator method.