keel.Algorithms.Decision_Trees.DT_GA

Class Poblacion

java.lang.Object

keel.Algorithms.Decision_Trees.DT_GA.Poblacion

public class Poblacion
extends java.lang.Object

Title: Poblacion (Population).

Description: This class implements the population of chromosomes used to perform the genetic algorithm

Version:

1.0

Author:

not attributable

Constructor Summary

Constructors

Constructor and Description
Poblacion() Default Constructor.
Poblacion(boolean[] ejemplosTr, int nGenerations, int popSize, double crossProb, double mutProb, myDataset train, double[] norm_acc) Paramenter constructor.
Poblacion(int codigoRegla, Regla r, int nGenerations, int popSize, double crossProb, double mutProb, myDataset train, java.lang.String clase) Paramenter constructor.

Method Summary

Modifier and Type	Method and Description
boolean	BETTER(double a, double b) Checks if the double a is greater than b.
java.util.ArrayList<Regla>	dameReglas() Returns the rules generated by the GA and stored on the chromosomes of the population.
void	escogerEjemplos(boolean[] ejemplos) Chooses the examples to be used as training whoses boolean value is true.
void	GA_Large() Performs the large GA to generate the different rules for the decision tree.
void	GA_Small() Performs the small GA to generate the different rules for the decision tree.
int	numEjemplos(int clase) Obtains the number of examples for the i-th class.
java.lang.String	printString() Returns a String representation of the population.

Methods inherited from class java.lang.Object

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

Constructor Detail

Poblacion

public Poblacion()

Default Constructor. Basic structures will be initialized.

Poblacion

public Poblacion(int codigoRegla,
                 Regla r,
                 int nGenerations,
                 int popSize,
                 double crossProb,
                 double mutProb,
                 myDataset train,
                 java.lang.String clase)

Paramenter constructor. The population structures will be initialized with the parameters given.

Parameters:

codigoRegla - rule's code.

r - rule used to initialize the different chromosomes.

nGenerations - maximum number for generations.

popSize - population size.

crossProb - crossover probability.

mutProb - mutation probability.

clase - class for the population.

train - training dataset.

Poblacion

public Poblacion(boolean[] ejemplosTr,
                 int nGenerations,
                 int popSize,
                 double crossProb,
                 double mutProb,
                 myDataset train,
                 double[] norm_acc)

Paramenter constructor. The population structures will be initialized with the parameters given.

Parameters:

ejemplosTr - examples considered to be used as training.

nGenerations - maximum number for generations.

popSize - population size.

crossProb - crossover probability.

mutProb - mutation probability.

train - training dataset.

norm_acc - initial accuracies.

Method Detail

BETTER

public boolean BETTER(double a,
                      double b)

Checks if the double a is greater than b.

Parameters:

a - first given number.

b - second given number.

Returns:

True if the double a is greater than b.

printString

public java.lang.String printString()

Returns a String representation of the population.

Returns:

a String representation of the population.

GA_Small

public void GA_Small()

Performs the small GA to generate the different rules for the decision tree.

numEjemplos

public int numEjemplos(int clase)

Obtains the number of examples for the i-th class.

Parameters:

clase - int class position.

Returns:

int the number of examples for the i-th class.

escogerEjemplos

public void escogerEjemplos(boolean[] ejemplos)

Chooses the examples to be used as training whoses boolean value is true.

Parameters:

ejemplos - given boolean vector containing the information of the selected examples.

GA_Large

public void GA_Large()

Performs the large GA to generate the different rules for the decision tree.

dameReglas

public java.util.ArrayList<Regla> dameReglas()

Returns the rules generated by the GA and stored on the chromosomes of the population.

Returns:

Array of rules that is used to build the tree.