Examples of org.uncommons.watchmaker.framework.selection.RouletteWheelSelection

Package org.uncommons.watchmaker.framework.selection

Examples of org.uncommons.watchmaker.framework.selection.RouletteWheelSelection

org.uncommons.watchmaker.framework.selection.RouletteWheelSelection

Implements selection of n candidates from a population by selecting n candidates at random where the probability of each candidate getting selected is proportional to its fitness score. This is analogous to each candidate being assigned an area on a roulette wheel proportionate to its fitness and the wheel being spun i times. Candidates may be selected more than once.

In some instances, particularly with small population sizes, the randomness of selection may result in excessively high occurrences of particular candidates. If this is a problem, {@link StochasticUniversalSampling} provides an alternativefitness-proportionate strategy for selection.
@author Daniel Dyer

    DatasetSplit split = new DatasetSplit(0.75);
    
    // Fitness Evaluator (defaults to training)
    FitnessEvaluator<? super CDRule> evaluator = new CDFitnessEvaluator(dataset, target, split);
    // Selection Strategy
    SelectionStrategy<? super CDRule> selection = new RouletteWheelSelection();
    
    EvolutionEngine<CDRule> engine = new SequentialEvolutionEngine<CDRule>(factory, pipeline, evaluator,
        selection, RandomUtils.getRandom());
    
    engine.addEvolutionObserver(new EvolutionObserver<CDRule>() {

View Full Code Here

      innerPanel.add(generationsSpinner);
      
      selectionLabel = new JLabel("Selection Strategy: ");
      innerPanel.add(selectionLabel);
      
      SelectionStrategy<?>[] selectionStrategies = {new RankSelection(), new RouletteWheelSelection(),
                                                    new StochasticUniversalSampling(),
                                                    new TournamentSelection(new Probability(0.95)),
                                                    new TruncationSelection(0.5)};
      
      selectionCombo = new JComboBox(selectionStrategies);

View Full Code Here

    DatasetSplit split = new DatasetSplit(0.75);


    // Fitness Evaluator (defaults to training)
    FitnessEvaluator<? super CDRule> evaluator = new CDFitnessEvaluator(dataset, target, split);
    // Selection Strategy
    SelectionStrategy<? super CDRule> selection = new RouletteWheelSelection();


    EvolutionEngine<CDRule> engine =
        new SequentialEvolutionEngine<CDRule>(factory, pipeline, evaluator, selection, RandomUtils.getRandom());


    engine.addEvolutionObserver(new EvolutionObserver<CDRule>() {

View Full Code Here

      innerPanel.add(generationsSpinner);
      
      selectionLabel = new JLabel("Selection Strategy: ");
      innerPanel.add(selectionLabel);
      
      SelectionStrategy<?>[] selectionStrategies = {new RankSelection(), new RouletteWheelSelection(),
                                                    new StochasticUniversalSampling(),
                                                    new TournamentSelection(new Probability(0.95)),
                                                    new TruncationSelection(0.5)};
      
      selectionCombo = new JComboBox(selectionStrategies);

View Full Code Here

    DatasetSplit split = new DatasetSplit(0.75);


    // Fitness Evaluator (defaults to training)
    FitnessEvaluator<? super CDRule> evaluator = new CDFitnessEvaluator(dataset, target, split);
    // Selection Strategy
    SelectionStrategy<? super CDRule> selection = new RouletteWheelSelection();


    EvolutionEngine<CDRule> engine =
        new SequentialEvolutionEngine<CDRule>(factory, pipeline, evaluator, selection, RandomUtils.getRandom());


    engine.addEvolutionObserver(new EvolutionObserver<CDRule>() {

View Full Code Here


    // Fitness Evaluator (defaults to training)
    FitnessEvaluator<? super CDRule> evaluator = new CDFitnessEvaluator(
        dataset, target, split);
    // Selection Strategy
    SelectionStrategy<? super CDRule> selection = new RouletteWheelSelection();


    EvolutionEngine<CDRule> engine = new SequentialEvolutionEngine<CDRule>(factory,
        pipeline, evaluator, selection, new MersenneTwisterRNG());


    engine.addEvolutionObserver(new EvolutionObserver<CDRule>() {

View Full Code Here


      selectionLabel = new JLabel("Selection Strategy: ");
      innerPanel.add(selectionLabel);


      SelectionStrategy<?>[] selectionStrategies = {
          new RankSelection(), new RouletteWheelSelection(),
          new StochasticUniversalSampling(), new TournamentSelection(new Probability(0.95d)),
          new TruncationSelection(0.5d)};


      selectionCombo = new JComboBox(selectionStrategies);
      selectionCombo.setRenderer(new DefaultListCellRenderer() {

View Full Code Here


    // Fitness Evaluator (defaults to training)
    FitnessEvaluator<? super CDRule> evaluator = new CDFitnessEvaluator(
        dataset, target, split);
    // Selection Strategy
    SelectionStrategy<? super CDRule> selection = new RouletteWheelSelection();


    EvolutionEngine<CDRule> engine = new SequentialEvolutionEngine<CDRule>(factory,
        pipeline, evaluator, selection, RandomUtils.getRandom());


    engine.addEvolutionObserver(new EvolutionObserver<CDRule>() {

View Full Code Here


      selectionLabel = new JLabel("Selection Strategy: ");
      innerPanel.add(selectionLabel);


      SelectionStrategy<?>[] selectionStrategies = {
          new RankSelection(), new RouletteWheelSelection(),
          new StochasticUniversalSampling(), new TournamentSelection(new Probability(0.95)),
          new TruncationSelection(0.5)};


      selectionCombo = new JComboBox(selectionStrategies);
      selectionCombo.setRenderer(new DefaultListCellRenderer() {

View Full Code Here

        operators.add(new StringCrossover());
        EvolutionaryOperator<String> pipeline = new EvolutionPipeline<String>(operators);
        EvolutionEngine<String> engine = new GenerationalEvolutionEngine<String>(factory,
                                                                                 pipeline,
                                                                                 new StringEvaluator(target),
                                                                                 new RouletteWheelSelection(),
                                                                                 new MersenneTwisterRNG());
        engine.addEvolutionObserver(new EvolutionLogger());
        return engine.evolve(100, // 100 individuals in the population.
                             5, // 5% elitism.
                             new TargetFitness(0, false));

View Full Code Here

0 1 2

TOP

Related Classes of org.uncommons.watchmaker.framework.selection.RouletteWheelSelection

org.apache.mahout.ga.watchmaker.cd.CDGA

org.apache.mahout.ga.watchmaker.travellingsalesman.StrategyPanel$EvolutionPanel

org.uncommons.watchmaker.examples.bits.BitsExample

org.uncommons.watchmaker.examples.geneticprogramming.GeneticProgrammingExample

org.uncommons.watchmaker.examples.strings.StringsExample

org.uncommons.watchmaker.framework.GenerationalEvolutionEngineTest

org.uncommons.watchmaker.framework.islands.IslandEvolutionTest

org.uncommons.watchmaker.framework.SteadyStateEvolutionEngineTest

org.uncommons.watchmaker.swing.SelectionStrategyControl

org.uncommons.watchmaker.swing.SelectionStrategyControlTest

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.