eva2.optimization.operator.fitnessmodifier.FitnessAdaptiveClustering Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of eva2 Show documentation
Gradle plugin that provides tasks for configuring and uploading artifacts to Sonatype Nexus.
The newest version!
package eva2.optimization.operator.fitnessmodifier;

import eva2.optimization.operator.cluster.ClusteringDensityBased;
import eva2.optimization.operator.cluster.InterfaceClustering;
import eva2.optimization.population.Population;
import eva2.util.annotation.Description;

/**
 * The fitness modifier are defunct and are to be moved to
 * the selection operators...
 */
@Description("This is a normation method based on Fitness Sharing. It adds a penalty for too similar individuals on the standard Normation method.")
public class FitnessAdaptiveClustering implements java.io.Serializable, InterfaceFitnessModifier {

    private InterfaceClustering clusteringAlgorithm = new ClusteringDensityBased();

    /**
     * This method allows you to modify the fitness of the individuals
     * of a population. Note that by altering the fitness you may require
     * your problem to store the unaltered fitness somewhere else so that
     * you may still fetch it!
     */
    @Override
    public void modifyFitness(Population population) {
        // prepare the calculation
        double[][] data = new double[population.size()][];
        for (int i = 0; i < data.length; i++) {
            data[i] = population.get(i).getFitness();
        }
        double min = Double.POSITIVE_INFINITY;
        double[] result = new double[data.length];

        for (int x = 0; x < data[0].length; x++) {
            for (int i = 0; i < data.length; i++) {
                data[i][x] = -data[i][x];
            }
            for (int i = 0; i < data.length; i++) {
                if (data[i][x] < min) {
                    min = data[i][x];
                }
            }

            for (int i = 0; i < data.length; i++) {
                // This will cause the worst individual to have no chance of being selected
                // also note that if all individual achieve equal fitness the sum will be zero
                result[i] = data[i][x] - min + 0.1;
            }
            this.clusteringAlgorithm.initClustering(population);
            // Now search for clusters
            Population[] ClusterResult = this.clusteringAlgorithm.cluster(population, population);
            Population cluster;
            for (int i = 1; i < ClusterResult.length; i++) {
                cluster = ClusterResult[i];
                for (int j = 0; j < cluster.size(); j++) {
                    result[i] /= ((double) cluster.size());
                }
            }

            for (int i = 0; i < population.size(); i++) {
                population.get(i).SetFitness(x, result[i]);
            }
        }
    }

    /**
     * This method allows you to set/get the clustering method on which the
     * species convergence is based.
     *
     * @return The current clustering method
     */
    public InterfaceClustering getClusteringAlgorithm() {
        return this.clusteringAlgorithm;
    }

    public void setClusteringAlgorithm(InterfaceClustering b) {
        this.clusteringAlgorithm = b;
    }

    public String clusteringAlgorithmTipText() {
        return "The Cluster Algorithm on which the adaptive fitness sharing is based.";
    }
}