edu.byu.nlp.stats.DirichletMLEOptimizableTest.java Source code

Introduction

Here is the source code for edu.byu.nlp.stats.DirichletMLEOptimizableTest.java

Source

/**
 * Copyright 2012 Brigham Young University
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *    http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */
package edu.byu.nlp.stats;

import static org.fest.assertions.Assertions.assertThat;

import org.apache.commons.math3.linear.Array2DRowRealMatrix;
import org.apache.commons.math3.linear.ArrayRealVector;
import org.apache.commons.math3.linear.LUDecomposition;
import org.apache.commons.math3.linear.RealMatrix;
import org.apache.commons.math3.linear.RealVector;
import org.apache.commons.math3.special.Gamma;
import org.fest.assertions.Delta;
import org.junit.Test;

import edu.byu.nlp.math.optimize.ValueAndObject;
import edu.byu.nlp.util.DoubleArrays;

/**
 * @author rah67
 *
 */
public class DirichletMLEOptimizableTest {

    private static final Delta delta = Delta.delta(1e-10);

    /**
     * Test method for {@link edu.byu.nlp.stats.DirichletMLEOptimizable#computeNext(double[])}.
     */
    @Test
    public void testComputeNextInPlace() {
        final double[][] data = DirichletTestUtils.sampleDataset();

        DirichletMLEOptimizable o = DirichletMLEOptimizable.newOptimizable(data, true);
        double[] alpha = new double[] { 2.0, 3.0, 4.0 };
        // Guard against bugs in the test
        assertThat(alpha.length).isEqualTo(data[0].length);

        double[] alphaCopy = alpha.clone();
        ValueAndObject<double[]> vao = o.computeNext(alphaCopy);

        // Ensure that it was performed in place
        assertThat(vao.getObject()).isSameAs(alphaCopy);

        // Ensure that the update was computed correctly
        RealVector expected = DirichletMLEOptimizableTest.newtonRaphsonUpdate(data, alpha);
        assertThat(alphaCopy).isEqualTo(expected.toArray(), delta);

        // TODO : check value
    }

    /**
     * Test method for {@link edu.byu.nlp.stats.DirichletMLEOptimizable#computeNext(double[])}.
     */
    @Test
    public void testComputeNextNotInPlace() {
        final double[][] data = DirichletTestUtils.sampleDataset();

        DirichletMLEOptimizable o = DirichletMLEOptimizable.newOptimizable(data, false);
        double[] alpha = new double[] { 2.0, 3.0, 4.0 };
        // Guard against bugs in the test
        assertThat(alpha.length).isEqualTo(data[0].length);

        ValueAndObject<double[]> vao = o.computeNext(alpha);

        // Ensure that it was not performed in place
        assertThat(vao.getObject()).isNotSameAs(alpha);
    }

    /**
     * Computes a Newton-Raphson update in-place to alpha.
     */
    private static RealVector newtonRaphsonUpdate(final double[][] data, double[] alpha) {
        // We'll compute the gold-standard value the "long" way (taking the inverse of the Hessian)
        RealMatrix hessian = new Array2DRowRealMatrix(alpha.length, alpha.length);
        for (int r = 0; r < hessian.getRowDimension(); r++) {
            for (int c = 0; c < hessian.getColumnDimension(); c++) {
                hessian.addToEntry(r, c, data.length * Gamma.trigamma(DoubleArrays.sum(alpha)));
                if (r == c) {
                    hessian.addToEntry(r, c, -data.length * Gamma.trigamma(alpha[r]));
                }
            }
        }
        RealVector derivative = new ArrayRealVector(alpha.length);
        for (int k = 0; k < alpha.length; k++) {
            derivative.setEntry(k,
                    data.length * (Gamma.digamma(DoubleArrays.sum(alpha)) - Gamma.digamma(alpha[k])));
            for (double[] theta : data) {
                derivative.addToEntry(k, theta[k]);
            }
        }

        RealMatrix hessianInverse = new LUDecomposition(hessian).getSolver().getInverse();
        RealVector negDiff = hessianInverse.preMultiply(derivative);
        negDiff.mapMultiplyToSelf(-1.0);

        RealVector expected = new ArrayRealVector(alpha, true);
        return expected.add(negDiff);
    }

}