Examples of org.encog.neural.networks.BasicNetwork

org.encog.neural.networks.BasicNetwork
This class implements a neural network. This class works in conjunction the Layer classes. Layers are added to the BasicNetwork to specify the structure of the neural network. The first layer added is the input layer, the final layer added is the output layer. Any layers added between these two layers are the hidden layers. The network structure is stored in the structure member. It is important to call: network.getStructure().finalizeStructure(); Once the neural network has been completely constructed.

    
    public NeuralNetworkWrapper(String filePath,int inputNeurons,int outputNeurons){
        this.filePath=filePath;
        File file=new File(filePath);
        if(file==null || !file.exists()){
            neuralNetwork=new BasicNetwork();
            neuralNetwork.addLayer(new BasicLayer(null,true,inputNeurons));
            neuralNetwork.addLayer(new BasicLayer(new ActivationSigmoid(),true,ProgramConfig.NEURAL_HIDDEN_LAYER_LENGTH));
            neuralNetwork.addLayer(new BasicLayer(new ActivationSigmoid(),false,outputNeurons));
            neuralNetwork.getStructure().finalizeStructure();
            neuralNetwork.reset();

View Full Code Here

    }
    if (hidden2 > 0) {
      pattern.addHiddenLayer(hidden2);
    }


    final BasicNetwork network = (BasicNetwork)pattern.generate();
    network.reset();
    return network;
  }

View Full Code Here

    double outputLow = -1;
    double outputHigh = 1;
    
    // Create a basic neural network, just to examine activation functions.
    MLMethodFactory methodFactory = new MLMethodFactory();    
    BasicNetwork network = (BasicNetwork)methodFactory.create(getMethodName(), architecture, 1, 1);
    
    if( network.getLayerCount()<1 ) {
      throw new EncogError("Neural network does not have an output layer.");
    }
    
    ActivationFunction outputFunction = network.getActivation(network.getLayerCount()-1);
    
    double[] d = { -1000, -100, -50 };
    outputFunction.activationFunction(d, 0, d.length);
    
    if( d[0]>0 && d[1]>0 && d[2]>0 ) {

View Full Code Here

   * 
   * @return A cloned copy of the neural network.
   */
  @Override
  public Object clone() {
    final BasicNetwork result = (BasicNetwork) ObjectCloner.deepCopy(this);
    return result;
  }

View Full Code Here

   *            Contains information about the job unit.
   */
  @Override
  public void performJobUnit(final JobUnitContext context) {


    final BasicNetwork network = (BasicNetwork) context.getJobUnit();
    BufferedMLDataSet buffer = null;
    MLDataSet useTraining = this.training;


    if (this.training instanceof BufferedMLDataSet) {
      buffer = (BufferedMLDataSet) this.training;
      useTraining = buffer.openAdditional();
    }


    // train the neural network


    double error = Double.POSITIVE_INFINITY;
    for (int z = 0; z < this.weightTries; z++) {
      network.reset();
      final Propagation train = new ResilientPropagation(network,
          useTraining);
      final StopTrainingStrategy strat = new StopTrainingStrategy(0.001,
          5);


      train.addStrategy(strat);
      train.setThreadCount(1); // force single thread mode


      for (int i = 0; (i < this.iterations) && !getShouldStop()
          && !strat.shouldStop(); i++) {
        train.iteration();
      }


      error = Math.min(error, train.getError());
    }


    if (buffer != null) {
      buffer.close();
    }


    if (!getShouldStop()) {
      // update min and max


      this.high = Math.max(this.high, error);
      this.low = Math.min(this.low, error);


      if (this.hidden1Size > 0) {
        int networkHidden1Count;
        int networkHidden2Count;


        if (network.getLayerCount() > 3) {
          networkHidden2Count = network.getLayerNeuronCount(2);
          networkHidden1Count = network.getLayerNeuronCount(1);
        } else {
          networkHidden2Count = 0;
          networkHidden1Count = network.getLayerNeuronCount(1);
        }


        int row, col;


        if (this.hidden2Size == 0) {

View Full Code Here

  public Object requestNextTask() {
    if (this.done || getShouldStop()) {
      return null;
    }


    final BasicNetwork network = generateNetwork();


    if (!increaseHiddenCounts()) {
      this.done = true;
    }

View Full Code Here

    }


    // now select the best network, which is the most simple of the
    // top networks.


    BasicNetwork choice = null;


    for (final BasicNetwork n : this.topNetworks) {
      if (n == null) {
        continue;
      }


      if (choice == null) {
        choice = n;
      } else {
        if (n.getStructure().calculateSize() < choice.getStructure()
            .calculateSize()) {
          choice = n;
        }
      }
    }

View Full Code Here

   */
  @Override
  public MLMethod generate() {
    BasicLayer hidden, input;


    final BasicNetwork network = new BasicNetwork();
    network.addLayer(input = new BasicLayer(this.activation, true,
        this.inputNeurons));
    network.addLayer(hidden = new BasicLayer(this.activation, true,
        this.hiddenNeurons));
    network.addLayer(new BasicLayer(null, false, this.outputNeurons));
    input.setContextFedBy(hidden);
    network.getStructure().finalizeStructure();
    network.reset();
    return network;
  }

View Full Code Here

  @Override
  public MLMethod generate() {


    BasicLayer hidden, output;


    final BasicNetwork network = new BasicNetwork();
    network.addLayer(new BasicLayer(null, true,
        this.inputNeurons));
    network.addLayer(hidden = new BasicLayer(this.activation, true,
        this.hiddenNeurons));
    network.addLayer(output = new BasicLayer(this.activation, false,
        this.outputNeurons));
    hidden.setContextFedBy(output);
    network.getStructure().finalizeStructure();
    network.reset();
    return network;
  }

View Full Code Here

   */
  @Override
  public void randomize(final MLMethod method) {


    if (method instanceof BasicNetwork) {
      final BasicNetwork network = (BasicNetwork) method;
      for (int i = 0; i < network.getLayerCount() - 1; i++) {
        randomize(network, i);
      }
    } else if (method instanceof MLEncodable) {
      final MLEncodable encode = (MLEncodable) method;
      final double[] encoded = new double[encode.encodedArrayLength()];

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of org.encog.neural.networks.BasicNetwork

com.wiieditor.neuralnetwork.NeuralNetworkWrapper

mechanics.NeuralNetwork

org.encog.engine.network.activation.ActivationFunction

org.encog.ensemble.ml.mlp.factory.MultiLayerPerceptronFactory

org.encog.examples.neural.activation.CustomActivation

org.encog.examples.neural.adaline.AdalineDigits

org.encog.examples.neural.benchmark.MultiBench

org.encog.examples.neural.benchmark.SimpleBenchmark

org.encog.examples.neural.benchmark.TestConverge

org.encog.examples.neural.benchmark.ThreadCount

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.