eva2/src/main/java/eva2/optimization/operator/selection/SelectParticleWheel.java

package eva2.optimization.operator.selection;

import eva2.optimization.individuals.AbstractEAIndividual;
import eva2.optimization.operator.selection.probability.InterfaceSelectionProbability;
import eva2.optimization.operator.selection.probability.SelectionProbabilityStandard;
import eva2.optimization.operator.selection.probability.SelectionProbabilityStandardScaling;
import eva2.optimization.population.Population;
import eva2.tools.math.RNG;
import eva2.util.annotation.Description;


/**
 * This method implements the roulette wheel selection for
 * a particle filter. In case of multiple fitness values the selection
 * criteria should be selected randomly for each selection event.
 */
@Description("This method chooses individuals similar to the static roulette wheel. The chance for each individual to be selected depends on the selection probability. The selection probability is 1 for all Individuals with a fitness that is bigger than the midean fitness." +
        "This is a single objective selecting method, it will select in respect to a random criterion.")
public class SelectParticleWheel implements InterfaceSelection, java.io.Serializable {

    private boolean obeyDebsConstViolationPrinciple = true;
    /**
     * Comment for {@code serialVersionUID}
     */
    private static final long serialVersionUID = 1L;
    private InterfaceSelectionProbability selProbCalculator = new SelectionProbabilityStandard();
    private boolean selectFixedSteps = false;

    public SelectParticleWheel() {
    }

    public SelectParticleWheel(double scalingProb) {
        selProbCalculator = new SelectionProbabilityStandardScaling(scalingProb);
    }

    public SelectParticleWheel(InterfaceSelectionProbability selProb) {
        selProbCalculator = selProb;
    }

    public SelectParticleWheel(SelectParticleWheel a) {
        this.selProbCalculator = (InterfaceSelectionProbability) a.selProbCalculator.clone();
        this.obeyDebsConstViolationPrinciple = a.obeyDebsConstViolationPrinciple;
    }

    @Override
    public Object clone() {
        return new SelectParticleWheel(this);
    }

    @Override
    public void prepareSelection(Population population) {
        selProbCalculator.computeSelectionProbability(population, "Fitness", obeyDebsConstViolationPrinciple);
    }

    /**
     * This method will select individuals from the given Population with respect to their
     * selection propability. This implements a fixed segment roulette wheel selection which ensures
     * that every individual which has a selection probability p >= (k/size) is selected k or k+1 times.
     *
     * @param population The source population where to select from
     * @param size       The number of Individuals to select
     * @return The selected population.
     */
    @Override
    public Population selectFrom(Population population, int size) {
        Population result = new Population();
        result.setTargetSize(size);

        if (selectFixedSteps) {
            selectFixed(population, size, result);
        } else {
            selectDrawIndependent(population, size, result);
        }

////	Unfortunately, this was really problem specific (mk)
//
//        double overallFit=0.0;
//    
//        for (int i = 0; i < size; i++) {
//        	overallFit += 33.0-((AbstractEAIndividual)population.get(i)).getFitness(0);
//        }
//        overallFit/=(double)size;
//
//        double momMark=0.0;
//        double momFitSum=0.0;
//        
//        for (int i = 0; i < size; i++) {
//        	momFitSum += 33.0-((AbstractEAIndividual)population.get(i)).getFitness(0);
//        	
//        	while (momFitSum > momMark) {
//        		result.add(((AbstractEAIndividual)population.get(i)));
//        		momMark+=overallFit;        		
//        	}
//        	
//        }
        return result;
    }

    private void selectDrawIndependent(Population population, int size,
                                       Population result) {
        double sum = 0, selPoint = 0;
        int selIndex;
        for (int i = 0; i < size; i++) {
            selPoint = RNG.randomDouble();
            selIndex = 0;
            sum = ((AbstractEAIndividual) population.getIndividual(0)).getSelectionProbability(0);
            while (selPoint >= sum) {
                selIndex++;
                sum += ((AbstractEAIndividual) population.getIndividual(selIndex)).getSelectionProbability(0);
            }
            result.add((AbstractEAIndividual) (population.get(selIndex)).clone());
            ((AbstractEAIndividual) result.getIndividual(i)).setAge(0);
        }
    }

    private void selectFixed(Population population, int size, Population result) {
        // use a fixed segment roulette wheel selection
        double segment = 1. / (size + 1);
        double selPoint = RNG.randomDouble(0., segment);

        int selIndex = 0;
        double selFitSum = ((AbstractEAIndividual) population.getIndividual(selIndex)).getSelectionProbability(0);

        for (int i = 0; i < size; i++) {
            while (selFitSum < selPoint) {
                selIndex++;
                selFitSum += ((AbstractEAIndividual) population.getIndividual(selIndex)).getSelectionProbability(0);
            }
            result.add((AbstractEAIndividual) (population.get(selIndex)).clone());
            ((AbstractEAIndividual) result.getIndividual(i)).setAge(0);
            selPoint += segment;
        }
    }

    /**
     * This method allows you to select partners for a given Individual
     *
     * @param dad              The already seleceted parent
     * @param availablePartners The mating pool.
     * @param size             The number of partners needed.
     * @return The selected partners.
     */
    @Override
    public Population findPartnerFor(AbstractEAIndividual dad, Population availablePartners, int size) {
        return this.selectFrom(availablePartners, size);
    }

    /**
     * This method allows the CommonJavaObjectEditorPanel to read the
     * name to the current object.
     *
     * @return The name.
     */
    public String getName() {
        return "Particle Wheel Selection";
    }

    /**
     * Toggle the use of obeying the constraint violation principle
     * of Deb
     *
     * @param b The new state
     */
    @Override
    public void setObeyDebsConstViolationPrinciple(boolean b) {
        this.obeyDebsConstViolationPrinciple = b;
    }

    public boolean getObeyDebsConstViolationPrinciple() {
        return this.obeyDebsConstViolationPrinciple;
    }

    public String obeyDebsConstViolationPrincipleToolTip() {
        return "Toggle the use of Deb's coonstraint violation principle(todo).";
    }

    /**
     * @return the selectFixedSteps
     */
    public boolean isSelectFixedSteps() {
        return selectFixedSteps;
    }

    /**
     * @param selectFixedSteps the selectFixedSteps to set
     */
    public void setSelectFixedSteps(boolean selectFixedSteps) {
        this.selectFixedSteps = selectFixedSteps;
    }

    public String selectFixedStepsTipText() {
        return "Use fixed segment wheel for selection if marked or independent draws if not.";
    }

    /**
     * @return the selProbCalculator
     */
    public InterfaceSelectionProbability getSelProbCalculator() {
        return selProbCalculator;
    }

    /**
     * @param selProbCalculator the selProbCalculator to set
     */
    public void setSelProbCalculator(
            InterfaceSelectionProbability selProbCalculator) {
        this.selProbCalculator = selProbCalculator;
    }

    public String selProbCalculatorTipText() {
        return "The method for calculating selection probability from the fitness.";
    }
}