Examples of GaussNewtonOptimizer

org.apache.commons.math.optimization.general.GaussNewtonOptimizer
Gauss-Newton least-squares solver.
This class solve a least-square problem by solving the normal equations of the linearized problem at each iteration. Either LU decomposition or QR decomposition can be used to solve the normal equations. LU decomposition is faster but QR decomposition is more robust for difficult problems.
@version $Revision: 1073158 $ $Date: 2011-02-21 22:46:52 +0100 (lun. 21 févr. 2011) $ @since 2.0
org.apache.commons.math3.optim.nonlinear.vector.jacobian.GaussNewtonOptimizer
Gauss-Newton least-squares solver.
Constraints are not supported: the call to {@link #optimize(OptimizationData[]) optimize} will throw{@link MathUnsupportedOperationException} if bounds are passed to it.
This class solve a least-square problem by solving the normal equations of the linearized problem at each iteration. Either LU decomposition or QR decomposition can be used to solve the normal equations. LU decomposition is faster but QR decomposition is more robust for difficult problems.
@since 2.0 @deprecated All classes and interfaces in this package are deprecated.The optimizers that were provided here were moved to the {@link org.apache.commons.math3.fitting.leastsquares} package(cf. MATH-1008).
org.apache.commons.math3.optimization.general.GaussNewtonOptimizer
Gauss-Newton least-squares solver.
This class solve a least-square problem by solving the normal equations of the linearized problem at each iteration. Either LU decomposition or QR decomposition can be used to solve the normal equations. LU decomposition is faster but QR decomposition is more robust for difficult problems.
@deprecated As of 3.1 (to be removed in 4.0). @since 2.0

Examples of org.apache.commons.math3.optim.nonlinear.vector.jacobian.GaussNewtonOptimizer

    @Test
    public void testTrivial() {
        LinearProblem problem
            = new LinearProblem(new double[][] { { 2 } }, new double[] { 3 });
        JacobianMultivariateVectorOptimizer underlyingOptimizer
            = new GaussNewtonOptimizer(true, new SimpleVectorValueChecker(1e-6, 1e-6));
        JDKRandomGenerator g = new JDKRandomGenerator();
        g.setSeed(16069223052l);
        RandomVectorGenerator generator
            = new UncorrelatedRandomVectorGenerator(1, new GaussianRandomGenerator(g));
        MultiStartMultivariateVectorOptimizer optimizer

View Full Code Here

Examples of org.apache.commons.math3.optim.nonlinear.vector.jacobian.GaussNewtonOptimizer


    @Test
    public void testIssue914() {
        LinearProblem problem = new LinearProblem(new double[][] { { 2 } }, new double[] { 3 });
        JacobianMultivariateVectorOptimizer underlyingOptimizer =
                new GaussNewtonOptimizer(true, new SimpleVectorValueChecker(1e-6, 1e-6)) {
            @Override
            public PointVectorValuePair optimize(OptimizationData... optData) {
                // filter out simple bounds, as they are not supported
                // by the underlying optimizer, and we don't really care for this test
                OptimizationData[] filtered = optData.clone();

View Full Code Here

Examples of org.apache.commons.math3.optim.nonlinear.vector.jacobian.GaussNewtonOptimizer

     * of the runs succeed.
     */
    @Test(expected=TestException.class)
    public void testNoOptimum() {
        JacobianMultivariateVectorOptimizer underlyingOptimizer
            = new GaussNewtonOptimizer(true, new SimpleVectorValueChecker(1e-6, 1e-6));
        JDKRandomGenerator g = new JDKRandomGenerator();
        g.setSeed(12373523445l);
        RandomVectorGenerator generator
            = new UncorrelatedRandomVectorGenerator(1, new GaussianRandomGenerator(g));
        MultiStartMultivariateVectorOptimizer optimizer

View Full Code Here

Examples of org.apache.commons.math3.optim.nonlinear.vector.jacobian.GaussNewtonOptimizer

        final SimpleVectorValueChecker checker = new SimpleVectorValueChecker(tol, tol);
        final double[] init = new double[] { 0, 0 };
        final int maxEval = 3;


        final double[] lm = doMath798(new LevenbergMarquardtOptimizer(checker), maxEval, init);
        final double[] gn = doMath798(new GaussNewtonOptimizer(checker), maxEval, init);


        for (int i = 0; i <= 1; i++) {
            Assert.assertEquals(lm[i], gn[i], tol);
        }
    }

View Full Code Here

Examples of org.apache.commons.math3.optim.nonlinear.vector.jacobian.GaussNewtonOptimizer

        final SimpleVectorValueChecker checker = new SimpleVectorValueChecker(tol, tol);
        final double[] init = new double[] { 0, 0 };
        final int maxEval = 10000; // Trying hard to fit.


        @SuppressWarnings("unused")
        final double[] gn = doMath798(new GaussNewtonOptimizer(checker), maxEval, init);
    }

View Full Code Here

Examples of org.apache.commons.math3.optim.nonlinear.vector.jacobian.GaussNewtonOptimizer

        final double[] init = new double[] { 0, 0 };
        final int maxEval = 10000; // Trying hard to fit.
        final SimpleVectorValueChecker checker = new SimpleVectorValueChecker(tol, tol, maxEval);


        final double[] lm = doMath798(new LevenbergMarquardtOptimizer(checker), maxEval, init);
        final double[] gn = doMath798(new GaussNewtonOptimizer(checker), maxEval, init);


        for (int i = 0; i <= 1; i++) {
            Assert.assertEquals(lm[i], gn[i], 1e-15);
        }
    }

View Full Code Here

Examples of org.apache.commons.math3.optim.nonlinear.vector.jacobian.GaussNewtonOptimizer

    }


    @Test
    public void testRedundantUnsolvable() {
        // Gauss-Newton should not be able to solve redundant information
        checkUnsolvableProblem(new GaussNewtonOptimizer(true, new SimpleVectorValueChecker(1e-15, 1e-15)), false);
    }

View Full Code Here

Examples of org.apache.commons.math3.optim.nonlinear.vector.jacobian.GaussNewtonOptimizer


        final double safeFactor = 1e2;
        final SimpleVectorValueChecker checker
            = new SimpleVectorValueChecker(safeFactor * Precision.EPSILON,
                                           safeFactor * Precision.SAFE_MIN);
        xFitter = new PolynomialFitter(new GaussNewtonOptimizer(false, checker));
        yFitter = new PolynomialFitter(new GaussNewtonOptimizer(false, checker));
    }

View Full Code Here

Examples of org.apache.commons.math3.optim.nonlinear.vector.jacobian.GaussNewtonOptimizer


    @Test(expected=NullPointerException.class)
    public void testGetOptimaBeforeOptimize() {


        JacobianMultivariateVectorOptimizer underlyingOptimizer
            = new GaussNewtonOptimizer(true, new SimpleVectorValueChecker(1e-6, 1e-6));
        JDKRandomGenerator g = new JDKRandomGenerator();
        g.setSeed(16069223052l);
        RandomVectorGenerator generator
            = new UncorrelatedRandomVectorGenerator(1, new GaussianRandomGenerator(g));
        MultiStartMultivariateVectorOptimizer optimizer

View Full Code Here

Examples of org.apache.commons.math3.optim.nonlinear.vector.jacobian.GaussNewtonOptimizer

    @Test
    public void testTrivial() {
        LinearProblem problem
            = new LinearProblem(new double[][] { { 2 } }, new double[] { 3 });
        JacobianMultivariateVectorOptimizer underlyingOptimizer
            = new GaussNewtonOptimizer(true, new SimpleVectorValueChecker(1e-6, 1e-6));
        JDKRandomGenerator g = new JDKRandomGenerator();
        g.setSeed(16069223052l);
        RandomVectorGenerator generator
            = new UncorrelatedRandomVectorGenerator(1, new GaussianRandomGenerator(g));
        MultiStartMultivariateVectorOptimizer optimizer

View Full Code Here

0 1 2 3 4 5

TOP

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.