Example usage for org.apache.commons.math3.distribution MixtureMultivariateNormalDistribution MixtureMultivariateNormalDistribution

Introduction

In this page you can find the example usage for org.apache.commons.math3.distribution MixtureMultivariateNormalDistribution MixtureMultivariateNormalDistribution.

Prototype

public MixtureMultivariateNormalDistribution(double[] weights, double[][] means, double[][][] covariances) 

Source Link

Document

Creates a multivariate normal mixture distribution.

Usage

From source file:sly.speakrecognizer.test.math.MultivariateNormalMixtureExpectationMaximizationFitterTest.java

@Test
public void testFitForDifferentSizesData() {
    System.out.println("TEST DLA ROZNYCH ROZMAROW DANYCH");
    double[][][] covariances = { new double[][] { new double[] { 1.74, -0.59 }, new double[] { -0.59, 1.02 }, },
            new double[][] { new double[] { 4.24, 2.58 }, new double[] { 2.58, 3.92 }, }, };
    double[][] means = { new double[] { -1.42, 1.69 }, new double[] { 4.21, 7.98 }, };

    double[] weights = new double[] { 0.296, 0.704 };
    MixtureMultivariateNormalDistribution mmnd = new MixtureMultivariateNormalDistribution(weights, means,
            covariances);//from ww  w.  ja v  a  2s. c o  m

    int[] lengths = new int[] { 10, 50, 250, 1500, 10000, 100000 };
    double[] errors = new double[lengths.length];
    int counter = 0;
    for (int length : lengths) {
        System.out.println("Dla dugoci danych: " + length);
        double[][] data = getTestSamples(length, mmnd);
        MultivariateNormalMixtureExpectationMaximization fitter = new MultivariateNormalMixtureExpectationMaximization(
                data);
        MixtureMultivariateNormalDistribution initialMix = MultivariateNormalMixtureExpectationMaximization
                .estimate(data, 2);
        fitter.fit(initialMix);
        MixtureMultivariateNormalDistribution fittedMix = fitter.getFittedModel();
        printMMND(fittedMix);
        errors[counter++] = printErrorsMMND(mmnd, fittedMix);
    }
    System.out.println("Podsumowanie");
    for (int x = 0; x < lengths.length; x++) {
        System.out.println("Dla length = " + lengths[x] + " bd wynosi = " + errors[x]);
    }

}

---

From source file:sly.speakrecognizer.test.math.MultivariateNormalMixtureExpectationMaximizationFitterTest.java

private double[][] getTestSamples(int length, double[] weights, double[][] means, double[][][] covariances) {
    return getTestSamples(length, new MixtureMultivariateNormalDistribution(weights, means, covariances));
}

---