Examples of org.encog.neural.som.training.basic.BasicTrainSOM

Package org.encog.neural.som.training.basic

Examples of org.encog.neural.som.training.basic.BasicTrainSOM

org.encog.neural.som.training.basic.BasicTrainSOM
This class implements competitive training, which would be used in a winner-take-all neural network, such as the self organizing map (SOM). This is an unsupervised training method, no ideal data is needed on the training set. If ideal data is provided, it will be ignored. Training is done by looping over all of the training elements and calculating a "best matching unit" (BMU). This BMU output neuron is then adjusted to better "learn" this pattern. Additionally, this training may be applied to other "nearby" output neurons. The degree to which nearby neurons are update is defined by the neighborhood function. A neighborhood function is required to determine the degree to which neighboring neurons (to the winning neuron) are updated by each training iteration. Because this is unsupervised training, calculating an error to measure progress by is difficult. The error is defined to be the "worst", or longest, Euclidean distance of any of the BMU's. This value should be minimized, as learning progresses. Because only the BMU neuron and its close neighbors are updated, you can end up with some output neurons that learn nothing. By default these neurons are not forced to win patterns that are not represented well. This spreads out the workload among all output neurons. This feature is not used by default, but can be enabled by setting the "forceWinner" property. @author jheaton

    this.setDefaultCloseOperation(EXIT_ON_CLOSE);
    this.network = createNetwork();
    this.getContentPane().add(map = new MapPanel(this));
    this.gaussian = new NeighborhoodRBF(RBFEnum.Gaussian,MapPanel.WIDTH,
        MapPanel.HEIGHT);
    this.train = new BasicTrainSOM(this.network, 0.01, null, gaussian);
    train.setForceWinner(false);
    this.thread = new Thread(this);
    thread.start();
  }

View Full Code Here

    
    // Create the neural network.
    SOM network = new SOM(4,2);
    network.reset();
    
    BasicTrainSOM train = new BasicTrainSOM(
        network,
        0.7,
        training,
        new NeighborhoodSingle());
        
    int iteration = 0;
    
    for(iteration = 0;iteration<=10;iteration++)
    {
      train.iteration();
      System.out.println("Iteration: " + iteration + ", Error:" + train.getError());
    }
    
    MLData data1 = new BasicMLData(SOM_INPUT[0]);
    MLData data2 = new BasicMLData(SOM_INPUT[1]);
    System.out.println("Pattern 1 winner: " + network.winner(data1));

View Full Code Here

    }
    if (neighborhoodStr.equalsIgnoreCase("single")) {
      nf = new NeighborhoodSingle();
    }


    final BasicTrainSOM result = new BasicTrainSOM((SOM) method,
        learningRate, training, nf);


    if (args.containsKey(MLTrainFactory.PROPERTY_ITERATIONS)) {
      final int plannedIterations = holder.getInt(
          MLTrainFactory.PROPERTY_ITERATIONS, false, 1000);
      final double startRate = holder.getDouble(
          MLTrainFactory.PROPERTY_START_LEARNING_RATE, false, 0.05);
      final double endRate = holder.getDouble(
          MLTrainFactory.PROPERTY_END_LEARNING_RATE, false, 0.05);
      final double startRadius = holder.getDouble(
          MLTrainFactory.PROPERTY_START_RADIUS, false, 10);
      final double endRadius = holder.getDouble(
          MLTrainFactory.PROPERTY_END_RADIUS, false, 1);
      result.setAutoDecay(plannedIterations, startRate, endRate,
          startRadius, endRadius);
    }


    return result;
  }

View Full Code Here


    if (sel.getSelected() == selectBasicSOM) {
      InputSOM somDialog = new InputSOM();


      if (somDialog.process()) {
        BasicTrainSOM train = new BasicTrainSOM((SOM) file.getObject(),
            somDialog.getLearningRate().getValue(), trainingData,
            somDialog.getNeighborhoodFunction());
        train.setForceWinner(somDialog.getForceWinner().getValue());
        startup(file, train, somDialog.getMaxError().getValue() / 100.0);
      }
    } else if (sel.getSelected() == selectSOMClusterCopy) {
      SOMClusterCopyTraining train = new SOMClusterCopyTraining(
          (SOM) file.getObject(), trainingData);
      train.iteration();
      if (EncogWorkBench.askQuestion("SOM", "Training done, save?")) {
        file.save();
      } else {
        file.revert();
      }

View Full Code Here


    // Create the neural network.
    SOM network = new SOM(4,2);    
    network.setWeights(new Matrix(MATRIX_ARRAY));


    final BasicTrainSOM train = new BasicTrainSOM(network, 0.4,
        training, new NeighborhoodSingle());
    train.setForceWinner(true);
    int iteration = 0;


    for (iteration = 0; iteration <= 100; iteration++) {
      train.iteration();
    }


    final MLData data1 = new BasicMLData(
        TestCompetitive.SOM_INPUT[0]);
    final MLData data2 = new BasicMLData(

View Full Code Here

    }
    if (neighborhoodStr.equalsIgnoreCase("single")) {
      nf = new NeighborhoodSingle();
    }


    final BasicTrainSOM result = new BasicTrainSOM((SOM) method,
        learningRate, training, nf);


    if (args.containsKey(MLTrainFactory.PROPERTY_ITERATIONS)) {
      final int plannedIterations = holder.getInt(
          MLTrainFactory.PROPERTY_ITERATIONS, false, 1000);
      final double startRate = holder.getDouble(
          MLTrainFactory.PROPERTY_START_LEARNING_RATE, false, 0.05);
      final double endRate = holder.getDouble(
          MLTrainFactory.PROPERTY_END_LEARNING_RATE, false, 0.05);
      final double startRadius = holder.getDouble(
          MLTrainFactory.PROPERTY_START_RADIUS, false, 10);
      final double endRadius = holder.getDouble(
          MLTrainFactory.PROPERTY_END_RADIUS, false, 1);
      result.setAutoDecay(plannedIterations, startRate, endRate,
          startRadius, endRadius);
    }


    return result;
  }

View Full Code Here


    // Create the neural network.
    SOM network = new SOM(4,2);    
    network.setWeights(new Matrix(MATRIX_ARRAY));


    final BasicTrainSOM train = new BasicTrainSOM(network, 0.4,
        training, new NeighborhoodSingle());
    train.setForceWinner(true);
    int iteration = 0;


    for (iteration = 0; iteration <= 100; iteration++) {
      train.iteration();
    }


    final MLData data1 = new BasicMLData(
        TestCompetitive.SOM_INPUT[0]);
    final MLData data2 = new BasicMLData(

View Full Code Here

TOP

Related Classes of org.encog.neural.som.training.basic.BasicTrainSOM

org.encog.examples.neural.gui.som.SomColors

org.encog.examples.neural.som.SimpleSOM

org.encog.mathutil.matrices.Matrix

org.encog.ml.data.basic.BasicMLData

org.encog.ml.data.MLData

org.encog.ml.factory.train.NeighborhoodSOMFactory

org.encog.neural.networks.training.competitive.TestCompetitive

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.