Examples of org.encog.neural.networks.training.genetic.NeuralGeneticAlgorithm

Package org.encog.neural.networks.training.genetic

Examples of org.encog.neural.networks.training.genetic.NeuralGeneticAlgorithm

org.encog.neural.networks.training.genetic.NeuralGeneticAlgorithm
Implements a genetic algorithm that allows a feedforward or simple recurrent neural network to be trained using a genetic algorithm. There are essentially two ways you can make use of this class. Either way, you will need a score object. The score object tells the genetic algorithm how well suited a neural network is. If you would like to use genetic algorithms with a training set you should make use TrainingSetScore class. This score object uses a training set to score your neural network. If you would like to be more abstract, and not use a training set, you can create your own implementation of the CalculateScore method. This class can then score the networks any way that you like.

  public void testGenetic() throws Throwable
  {
    MLDataSet trainingData = new BasicMLDataSet(XOR.XOR_INPUT,XOR.XOR_IDEAL);    
    BasicNetwork network = NetworkUtil.createXORNetworkUntrained();
    CalculateScore score = new TrainingSetScore(trainingData);
    NeuralGeneticAlgorithm genetic = new NeuralGeneticAlgorithm(network, new RangeRandomizer(-1,1), score, 500,0.1,0.25);
    NetworkUtil.testTraining(genetic,0.00001);
  }

View Full Code Here

      train = new NeuralSimulatedAnnealing(
          network, new PilotScore(), 10, 2, 100);
    }
    else
    {
      train = new NeuralGeneticAlgorithm(
          network, new FanInRandomizer(), 
          new PilotScore(),500, 0.1, 0.25);
    }
    
    int epoch = 1;

View Full Code Here

      final double mutationPercent = dialog.getMutationPercent()
          .getValue();
      final double percentToMate = dialog.getPercentToMate().getValue();


      CalculateScore score = new TrainingSetScore(trainingData);
      final NeuralGeneticAlgorithm train = new NeuralGeneticAlgorithm(
          (BasicNetwork) file.getObject(),
          new RangeRandomizer(-1, 1), score, populationSize,
          mutationPercent, percentToMate);
      train.setTraining(trainingData);
      startup(file, train, dialog.getMaxError().getValue() / 100.0);
    }


  }

View Full Code Here

    final double mutation = holder.getDouble(
        MLTrainFactory.PROPERTY_MUTATION, false, 0.1);
    final double mate = holder.getDouble(MLTrainFactory.PROPERTY_MATE,
        false, 0.25);


    final MLTrain train = new NeuralGeneticAlgorithm((BasicNetwork) method,
        new RangeRandomizer(-1, 1), score, populationSize, mutation,
        mate);


    return train;
  }

View Full Code Here

TOP

Related Classes of org.encog.neural.networks.training.genetic.NeuralGeneticAlgorithm

org.encog.examples.neural.lunar.LunarLander

org.encog.ml.factory.train.GeneticFactory

org.encog.ml.genetic.crossover.Splice

org.encog.ml.genetic.genome.Genome

org.encog.ml.genetic.mutate.MutatePerturb

org.encog.ml.genetic.population.BasicPopulation

org.encog.ml.genetic.population.Population

org.encog.neural.networks.BasicNetwork

org.encog.neural.networks.training.TestTraining

org.encog.workbench.process.TrainBasicNetwork

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.