Package org.neuroph.nnet

Examples of org.neuroph.nnet.MultiLayerPerceptron

24
25
26
27
28
29
30
31
32
33
34
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{0, 0}, new double[]{0}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{0, 1}, new double[]{1}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{1, 0}, new double[]{1}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{1, 1}, new double[]{0}));

        MultiLayerPerceptron nnet = new MultiLayerPerceptron( TransferFunctionType.TANH ,2, 3, 1);
        MatrixMultiLayerPerceptron mnet = new MatrixMultiLayerPerceptron(nnet);

        System.out.println("Training network...");

        mnet.learnInSameThread(trainingSet);
View Full Code Here
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
  public void run() {
   
    // uncomment the following line to use regular Neuroph (non-flat) processing
    Neuroph.getInstance().setFlattenNetworks(false);
   
    NeuralNetwork network = new MultiLayerPerceptron(TransferFunctionType.SIGMOID, WINDOW_SIZE, 10, 1);
     
    normalizeSunspots(0.1, 0.9);
   
    network.getLearningRule().addObserver(this);
   
    TrainingSet training = generateTraining();
    network.learnInSameThread(training);
    predict(network);
   
    Neuroph.getInstance().shutdown();
  }
View Full Code Here
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{0, 1}, new double[]{1}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{1, 0}, new double[]{1}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{1, 1}, new double[]{0}));

        // create multi layer perceptron
        MultiLayerPerceptron myMlPerceptron = new MultiLayerPerceptron(TransferFunctionType.TANH, 2, 3, 1);

        // enable batch if using MomentumBackpropagation
        if( myMlPerceptron.getLearningRule() instanceof MomentumBackpropagation )
          ((MomentumBackpropagation)myMlPerceptron.getLearningRule()).setBatchMode(true);

        // learn the training set
        System.out.println("Training neural network...");
        myMlPerceptron.learnInSameThread(trainingSet);

        // test perceptron
        System.out.println("Testing trained neural network");
        testNeuralNetwork(myMlPerceptron, trainingSet);

        // save trained neural network
        myMlPerceptron.save("myMlPerceptron.nnet");

        // load saved neural network
        NeuralNetwork loadedMlPerceptron = NeuralNetwork.load("myMlPerceptron.nnet");

        // test loaded neural network
View Full Code Here
98
99
100
101
102
103
104
105
106
   * @param transferFunctionType transfer function type for neurons
   * @return instance of Multi Layer Perceptron
   */
  public static MultiLayerPerceptron createMLPerceptron(String layersStr, TransferFunctionType transferFunctionType) {
    ArrayList<Integer> layerSizes = VectorParser.parseInteger(layersStr);
    MultiLayerPerceptron nnet = new MultiLayerPerceptron(layerSizes,
        transferFunctionType);
    return nnet;
  }
View Full Code Here
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
   * @return instance of Multi Layer Perceptron
   */
  public static MultiLayerPerceptron createMLPerceptron(String layersStr, TransferFunctionType transferFunctionType, Class learningRule,  boolean useBias, boolean connectIO) {
    ArrayList<Integer> layerSizes = VectorParser.parseInteger(layersStr);
                NeuronProperties neuronProperties = new NeuronProperties(transferFunctionType, useBias);
    MultiLayerPerceptron nnet = new MultiLayerPerceptron(layerSizes, neuronProperties);
               
                // set learning rule
                if (learningRule.getName().equals(BackPropagation.class.getName()))  {
                    nnet.setLearningRule(new BackPropagation());
                } else if (learningRule.getName().equals(MomentumBackpropagation.class.getName())) {
                    nnet.setLearningRule(new MomentumBackpropagation());
                } else if (learningRule.getName().equals(DynamicBackPropagation.class.getName())) {
                    nnet.setLearningRule(new DynamicBackPropagation());
                }

                // connect io
                if (connectIO) {
                    nnet.connectInputsToOutputs();
                }

    return nnet;
  }
View Full Code Here
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
    public static void main(String[] args) {
        System.out.println("Time stamp N1:" + new SimpleDateFormat("dd-MMM-yyyy HH:mm:ss:MM").format(new Date()));

        int maxIterations = 10000;
        NeuralNetwork neuralNet = new MultiLayerPerceptron(4, 9, 1);
        ((LMS) neuralNet.getLearningRule()).setMaxError(0.001);//0-1
        ((LMS) neuralNet.getLearningRule()).setLearningRate(0.7);//0-1
        ((LMS) neuralNet.getLearningRule()).setMaxIterations(maxIterations);//0-1
        TrainingSet trainingSet = new TrainingSet();

        double daxmax = 10000.0D;
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{3710.0D / daxmax, 3690.0D / daxmax, 3890.0D / daxmax, 3695.0D / daxmax}, new double[]{3666.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{3690.0D / daxmax, 3890.0D / daxmax, 3695.0D / daxmax, 3666.0D / daxmax}, new double[]{3692.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{3890.0D / daxmax, 3695.0D / daxmax, 3666.0D / daxmax, 3692.0D / daxmax}, new double[]{3886.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{3695.0D / daxmax, 3666.0D / daxmax, 3692.0D / daxmax, 3886.0D / daxmax}, new double[]{3914.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{3666.0D / daxmax, 3692.0D / daxmax, 3886.0D / daxmax, 3914.0D / daxmax}, new double[]{3956.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{3692.0D / daxmax, 3886.0D / daxmax, 3914.0D / daxmax, 3956.0D / daxmax}, new double[]{3953.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{3886.0D / daxmax, 3914.0D / daxmax, 3956.0D / daxmax, 3953.0D / daxmax}, new double[]{4044.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{3914.0D / daxmax, 3956.0D / daxmax, 3953.0D / daxmax, 4044.0D / daxmax}, new double[]{3987.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{3956.0D / daxmax, 3953.0D / daxmax, 4044.0D / daxmax, 3987.0D / daxmax}, new double[]{3996.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{3953.0D / daxmax, 4044.0D / daxmax, 3987.0D / daxmax, 3996.0D / daxmax}, new double[]{4043.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{4044.0D / daxmax, 3987.0D / daxmax, 3996.0D / daxmax, 4043.0D / daxmax}, new double[]{4068.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{3987.0D / daxmax, 3996.0D / daxmax, 4043.0D / daxmax, 4068.0D / daxmax}, new double[]{4176.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{3996.0D / daxmax, 4043.0D / daxmax, 4068.0D / daxmax, 4176.0D / daxmax}, new double[]{4187.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{4043.0D / daxmax, 4068.0D / daxmax, 4176.0D / daxmax, 4187.0D / daxmax}, new double[]{4223.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{4068.0D / daxmax, 4176.0D / daxmax, 4187.0D / daxmax, 4223.0D / daxmax}, new double[]{4259.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{4176.0D / daxmax, 4187.0D / daxmax, 4223.0D / daxmax, 4259.0D / daxmax}, new double[]{4203.0D / daxmax}));
        trainingSet.addElement(new SupervisedTrainingElement(new double[]{4187.0D / daxmax, 4223.0D / daxmax, 4259.0D / daxmax, 4203.0D / daxmax}, new double[]{3989.0D / daxmax}));
        neuralNet.learnInSameThread(trainingSet);
        System.out.println("Time stamp N2:" + new SimpleDateFormat("dd-MMM-yyyy HH:mm:ss:MM").format(new Date()));

        TrainingSet testSet = new TrainingSet();
        testSet.addElement(new TrainingElement(new double[]{4223.0D / daxmax, 4259.0D / daxmax, 4203.0D / daxmax, 3989.0D / daxmax}));

        for (TrainingElement testElement : testSet.trainingElements()) {
            neuralNet.setInput(testElement.getInput());
            neuralNet.calculate();
            double[] networkOutput = neuralNet.getOutput();
            System.out.print("Input: " + Arrays.toString(testElement.getInput()) );
            System.out.println(" Output: " + Arrays.toString(networkOutput) );
        }

        //Experiments:
View Full Code Here
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
    layersNeuronsCount.add(0, numberOfInputNeurons);
    layersNeuronsCount.add(numberOfOuputNeurons);
   
    System.out.println("Neuron layer size counts vector = " + layersNeuronsCount);
   
    NeuralNetwork neuralNetwork = new MultiLayerPerceptron(layersNeuronsCount, transferFunctionType);

    neuralNetwork.setLabel(label);
    PluginBase imageRecognitionPlugin = new ImageRecognitionPlugin(samplingResolution, colorMode);
    neuralNetwork.addPlugin(imageRecognitionPlugin);

    assignLabelsToOutputNeurons(neuralNetwork, imageLabels);
                neuralNetwork.setLearningRule(new MomentumBackpropagation());

            return neuralNetwork;
  }
View Full Code Here
TOP

Related Classes of org.neuroph.nnet.MultiLayerPerceptron

  • org.neuroph.contrib.imgrec.ImageRecognitionHelper
  • org.neuroph.contrib.matrixmlp.TestMatrixMLP
  • org.neuroph.core.Layer
  • org.neuroph.nnet.comp.BiasNeuron
  • org.neuroph.nnet.learning.MomentumBackpropagation
  • org.neuroph.samples.intronn.SunSpots
  • org.neuroph.samples.stockmarket.Main
  • org.neuroph.samples.XorMultiLayerPerceptronSample
  • org.neuroph.util.NeuralNetworkFactory
  • org.neuroph.util.NeuronProperties
    • Copyright © 2018 www.massapicom. All rights reserved.
    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.