Package org.apache.commons.math3.ml.neuralnet

Examples of org.apache.commons.math3.ml.neuralnet.Network


     20----21----22----23---24
     */
    @Test
    public void testConcentricNeighbourhood2() {
        final FeatureInitializer[] initArray = { init };
        final Network net = new NeuronSquareMesh2D(5, true,
                                                   5, true,
                                                   SquareNeighbourhood.MOORE,
                                                   initArray).getNetwork();

        Collection<Neuron> neighbours;
        Collection<Neuron> exclude = new HashSet<Neuron>();

        // Level-1 neighbourhood.
        neighbours = net.getNeighbours(net.getNeuron(8));
        for (long nId : new long[] { 2, 3, 4, 7, 9, 12, 13, 14 }) {
            Assert.assertTrue(neighbours.contains(net.getNeuron(nId)));
        }
        // Ensures that no other neurons is in the neihbourhood set.
        Assert.assertEquals(8, neighbours.size());

        // 1. Add the neuron to the "exclude" list.
        exclude.add(net.getNeuron(8));
        // 2. Add all neurons from level-1 neighbourhood.
        exclude.addAll(neighbours);
        // 3. Retrieve level-2 neighbourhood.
        neighbours = net.getNeighbours(neighbours, exclude);
        for (long nId : new long[] { 1, 6, 11, 16, 17, 18, 19, 15, 10, 5, 0, 20, 24, 23, 22, 21 }) {
            Assert.assertTrue(neighbours.contains(net.getNeuron(nId)));
        }
        // Ensures that no other neurons is in the neihbourhood set.
        Assert.assertEquals(16, neighbours.size());
    }
View Full Code Here


        wrapRows = wrapRowDim;
        wrapColumns = wrapColDim;
        neighbourhood = neighbourhoodType;

        final int fLen = featuresList[0][0].length;
        network = new Network(0, fLen);
        identifiers = new long[numberOfRows][numberOfColumns];

        // Add neurons.
        for (int i = 0; i < numberOfRows; i++) {
            for (int j = 0; j < numberOfColumns; j++) {
View Full Code Here

        wrapColumns = wrapColDim;
        neighbourhood = neighbourhoodType;
        identifiers = new long[numberOfRows][numberOfColumns];

        final int fLen = featureInit.length;
        network = new Network(0, fLen);

        // Add neurons.
        for (int i = 0; i < numRows; i++) {
            for (int j = 0; j < numCols; j++) {
                final double[] features = new double[fLen];
View Full Code Here

        // Multi-start loop.
        for (int i = 0; i < starts; i++) {
            // CHECKSTYLE: stop IllegalCatch
            try {
                // Decrease number of allowed evaluations.
                optimData[maxEvalIndex] = new MaxEval(maxEval - totalEvaluations);
                // New start value.
                final double s = (i == 0) ?
                    startValue :
                    min + generator.nextDouble() * (max - min);
                optimData[searchIntervalIndex] = new SearchInterval(min, max, s);
View Full Code Here

     * {@link #DEFAULT_INVERSE_ABSOLUTE_ACCURACY}).
     * @throws NotStrictlyPositiveException if {@code mean <= 0}.
     * @since 2.1
     */
    public ExponentialDistribution(double mean, double inverseCumAccuracy) {
        this(new Well19937c(), mean, inverseCumAccuracy);
    }
View Full Code Here

     * @param upper Upper bound (inclusive) of this distribution.
     * @throws NumberIsTooLargeException if {@code lower >= upper}.
     */
    public UniformIntegerDistribution(int lower, int upper)
        throws NumberIsTooLargeException {
        this(new Well19937c(), lower, upper);
    }
View Full Code Here

     * @throws NumberIsTooLargeException if {@code a >= b} or if {@code c > b}.
     * @throws NumberIsTooSmallException if {@code c < a}.
     */
    public TriangularDistribution(double a, double c, double b)
        throws NumberIsTooLargeException, NumberIsTooSmallException {
        this(new Well19937c(), a, c, b);
    }
View Full Code Here

        for (int i = 0; i < k; ++i) {
            sumImpl[i]     = new Sum();
            sumSqImpl[i]   = new SumOfSquares();
            minImpl[i]     = new Min();
            maxImpl[i]     = new Max();
            sumLogImpl[i= new SumOfLogs();
            geoMeanImpl[i] = new GeometricMean();
            meanImpl[i]    = new Mean();
        }

        covarianceImpl =
View Full Code Here

        geoMeanImpl = new StorelessUnivariateStatistic[k];
        meanImpl    = new StorelessUnivariateStatistic[k];

        for (int i = 0; i < k; ++i) {
            sumImpl[i]     = new Sum();
            sumSqImpl[i]   = new SumOfSquares();
            minImpl[i]     = new Min();
            maxImpl[i]     = new Max();
            sumLogImpl[i= new SumOfLogs();
            geoMeanImpl[i] = new GeometricMean();
            meanImpl[i]    = new Mean();
View Full Code Here

     * @param checker Convergence checker.
     */
    protected BaseOptimizer(ConvergenceChecker<PAIR> checker) {
        this.checker = checker;

        evaluations = new Incrementor(0, new MaxEvalCallback());
        iterations = new Incrementor(0, new MaxIterCallback());
    }
View Full Code Here

TOP

Related Classes of org.apache.commons.math3.ml.neuralnet.Network

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.