/**
* Copyright (C) 2001-2017 by RapidMiner and the contributors
*
* Complete list of developers available at our web site:
*
* http://rapidminer.com
*
* This program is free software: you can redistribute it and/or modify it under the terms of the
* GNU Affero General Public License as published by the Free Software Foundation, either version 3
* of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without
* even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Affero General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License along with this program.
* If not, see http://www.gnu.org/licenses/.
*/
package com.rapidminer.operator.features.selection;
import com.rapidminer.operator.features.Individual;
import com.rapidminer.operator.features.Population;
import com.rapidminer.operator.features.PopulationOperator;
import java.util.ArrayList;
import java.util.LinkedList;
import java.util.List;
import java.util.Random;
/**
* Crossover operator for the used bitlists of example sets. An example set is selected with a given
* fixed propability and a mating partner is determined randomly. Crossover can be either one point,
* uniform or shuffled. Please note that shuffled crossover first uniformly determines the number of
* attributes which should be swapped. Therefore uniform and shuffle crossover are not equivalent. <br>
*
* Only useful if all example sets have the same (number of) attributes.
*
* @author Simon Fischer, Ingo Mierswa Exp $
*/
public class SelectionCrossover implements PopulationOperator {
public static final String[] CROSSOVER_TYPES = { "one_point", "uniform", "shuffle" };
public static final int ONE_POINT = 0;
public static final int UNIFORM = 1;
public static final int SHUFFLE = 2;
private int type;
private double prob;
private Random random;
private int minNumber;
private int maxNumber;
private int exactNumber;
public SelectionCrossover(int type, double prob, Random random, int minNumber, int maxNumber, int exactNumber) {
this.prob = prob;
this.type = type;
this.random = random;
this.minNumber = minNumber;
this.maxNumber = maxNumber;
this.exactNumber = exactNumber;
}
/** The default implementation returns true for every generation. */
@Override
public boolean performOperation(int generation) {
return true;
}
public int getType() {
return type;
}
public void crossover(double[] weights1, double[] weights2) {
switch (type) {
case ONE_POINT:
int n = 1 + random.nextInt(weights1.length - 1);
for (int index = n; index < weights1.length; index++) {
double dummy = weights1[index];
weights1[index] = weights2[index];
weights2[index] = dummy;
}
break;
case UNIFORM:
boolean[] swap = new boolean[weights1.length];
for (int i = 0; i < swap.length; i++) {
swap[i] = random.nextBoolean();
;
}
swapAttributes(weights1, weights2, swap);
break;
case SHUFFLE:
swap = new boolean[weights1.length];
List<Integer> indices = new ArrayList<Integer>();
for (int i = 0; i < swap.length; i++) {
indices.add(i);
}
if (indices.size() > 0) {
int toSwap = random.nextInt(indices.size() - 1) + 1;
for (int i = 0; i < toSwap; i++) {
swap[indices.remove(random.nextInt(indices.size()))] = true;
}
}
swapAttributes(weights1, weights2, swap);
break;
default:
break;
}
}
private void swapAttributes(double[] weights1, double[] weights2, boolean[] swap) {
for (int index = 0; index < weights1.length; index++) {
if (swap[index]) {
double dummy = weights1[index];
weights1[index] = weights2[index];
weights2[index] = dummy;
}
}
}
@Override
public void operate(Population population) {
if (population.getNumberOfIndividuals() < 2) {
return;
}
LinkedList<double[]> matingPool = new LinkedList<double[]>();
for (int i = 0; i < population.getNumberOfIndividuals(); i++) {
matingPool.add(population.get(i).getWeightsClone());
}
List<Individual> l = new LinkedList<Individual>();
while (matingPool.size() > 1) {
double[] p1 = matingPool.remove(random.nextInt(matingPool.size()));
double[] p2 = matingPool.remove(random.nextInt(matingPool.size()));
if (random.nextDouble() < prob) {
crossover(p1, p2);
Individual newIndividual1 = new Individual(p1);
int numberOfFeatures = newIndividual1.getNumberOfUsedAttributes();
if (numberOfFeatures > 0) {
if (exactNumber > 0) {
if (numberOfFeatures == exactNumber) {
l.add(newIndividual1);
}
} else {
if (((maxNumber < 1) || (numberOfFeatures <= maxNumber)) && (numberOfFeatures >= minNumber)) {
l.add(newIndividual1);
}
}
}
Individual newIndividual2 = new Individual(p2);
numberOfFeatures = newIndividual2.getNumberOfUsedAttributes();
if (numberOfFeatures > 0) {
if (exactNumber > 0) {
if (numberOfFeatures == exactNumber) {
l.add(newIndividual2);
}
} else {
if (((maxNumber < 1) || (numberOfFeatures <= maxNumber)) && (numberOfFeatures >= minNumber)) {
l.add(newIndividual2);
}
}
}
}
}
population.addAllIndividuals(l);
}
}