Examples of de.lmu.ifi.dbs.elki.math.linearalgebra.fitting.LevenbergMarquardtMethod

Package de.lmu.ifi.dbs.elki.math.linearalgebra.fitting

Examples of de.lmu.ifi.dbs.elki.math.linearalgebra.fitting.LevenbergMarquardtMethod

de.lmu.ifi.dbs.elki.math.linearalgebra.fitting.LevenbergMarquardtMethod
Function parameter fitting using Levenberg-Marquardt method. The Levenberg-Marquardt Algorithm (LMA) is a combination of the Gauss-Newton Algorithm (GNA) and the method of steepest descent. As such it usually gives more stable results and better convergence. Implemented loosely based on the book:
Numerical Recipes In C: The Art Of Scientific Computing
ISBN 0-521-43108-5
Press, W.H. and Teukolsky, S.A. and Vetterling, W.T. and Flannery, B.P.
Cambridge University Press, Cambridge, Mass, 1992 Due to their license, we cannot use their code, but we have to implement the mathematics ourselves. We hope the loss in precision isn't too big. TODO: Replace implementation by one based on
M.I.A. Lourakis levmar:
Levenberg-Marquardt nonlinear least squares algorithms in C/C++ Which supposedly offers increased robustness. @author Erich Schubert @apiviz.has FittingFunction @apiviz.uses FittingFunctionResult oneway - - «create»

    for(int i = 0; i < testx.length; i++) {
      s[i] = 1.0;
    }
    double[] params = { mean, stddevq, 1 };
    boolean[] dofit = { true, true, false };
    LevenbergMarquardtMethod fit = new LevenbergMarquardtMethod(new GaussianFittingFunction(), params, dofit, testx, testy, s);
    for(int i = 0; i < 50; i++) {
      fit.iterate();
    }
    double[] ps = fit.getParams();
    // compare results.
    double[] should = { 0.152986763079, 1.00115077, 1 };
    assertEquals("Mean doesn't match.", should[0], ps[0], 0.0001);
    assertEquals("Stddev doesn't match.", should[1], ps[1], 0.0001);
    assertEquals("Scaling doesn't match.", should[2], ps[2], 0.0001);

View Full Code Here

    for(int i = 0; i < testx.length; i++) {
      s[i] = 1.0;
    }
    double[] params = { mean, stddevq, 1 };
    boolean[] dofit = { true, true, false };
    LevenbergMarquardtMethod fit = new LevenbergMarquardtMethod(new GaussianFittingFunction(), params, dofit, testx, testy, s);
    for(int i = 0; i < 50; i++) {
      fit.iterate();
    }
    double[] ps = fit.getParams();
    // compare results.
    double[] should = { 0.1557811515, 1.006463733, 1 };
    assertEquals("Mean doesn't match.", should[0], ps[0], 0.0001);
    assertEquals("Stddev doesn't match.", should[1], ps[1], 0.0001);
    assertEquals("Scaling doesn't match.", should[2], ps[2], 0.0001);

View Full Code Here


  private double[] run(double[] data, double[] params) {
    FittingFunction func = new GaussianFittingFunction();
    boolean[] dofit = { true, true, true };
    KernelDensityEstimator de = new KernelDensityEstimator(data, GaussianKernelDensityFunction.KERNEL);
    LevenbergMarquardtMethod fit = new LevenbergMarquardtMethod(func, params, dofit, data, de.getDensity(), de.getVariance());
    // for(int i = 0; i < 100; i++) {
    // fit.iterate();
    // double[] ps = fit.getParams();
    // System.out.println("Mean: " + ps[0] + " Stddev: " + ps[1] + " Amp: " +
    // ps[2]+" Chi: "+fit.getChiSq());
    // }
    fit.run();
    // double[] ps = fit.getParams();
    // System.out.println("Result: "+ps[0]+" "+ps[1]+" "+ps[2]+" Chi: "+fit.getChiSq()+" Iter: "+fit.maxruns);
    return fit.getParams();
  }

View Full Code Here

    for(int i = 0; i < testx.length; i++) {
      s[i] = 1.0;
    }
    double[] params = { mean, stddevq, 1 };
    boolean[] dofit = { true, true, false };
    LevenbergMarquardtMethod fit = new LevenbergMarquardtMethod(new GaussianFittingFunction(), params, dofit, testx, testy, s);
    for(int i = 0; i < 50; i++) {
      fit.iterate();
    }
    double[] ps = fit.getParams();
    // compare results.
    double[] should = { 0.152986763079, 1.00115077, 1 };
    assertEquals("Mean doesn't match.", should[0], ps[0], 0.0001);
    assertEquals("Stddev doesn't match.", should[1], ps[1], 0.0001);
    assertEquals("Scaling doesn't match.", should[2], ps[2], 0.0001);

View Full Code Here

    for(int i = 0; i < testx.length; i++) {
      s[i] = 1.0;
    }
    double[] params = { mean, stddevq, 1 };
    boolean[] dofit = { true, true, false };
    LevenbergMarquardtMethod fit = new LevenbergMarquardtMethod(new GaussianFittingFunction(), params, dofit, testx, testy, s);
    for(int i = 0; i < 50; i++) {
      fit.iterate();
    }
    double[] ps = fit.getParams();
    // compare results.
    double[] should = { 0.1557811515, 1.006463733, 1 };
    assertEquals("Mean doesn't match.", should[0], ps[0], 0.0001);
    assertEquals("Stddev doesn't match.", should[1], ps[1], 0.0001);
    assertEquals("Scaling doesn't match.", should[2], ps[2], 0.0001);

View Full Code Here


  private double[] run(double[] data, double[] params) {
    FittingFunction func = new GaussianFittingFunction();
    boolean[] dofit = { true, true, true };
    KernelDensityEstimator de = new KernelDensityEstimator(data, GaussianKernelDensityFunction.KERNEL);
    LevenbergMarquardtMethod fit = new LevenbergMarquardtMethod(func, params, dofit, data, de.getDensity(), de.getVariance());
    // for(int i = 0; i < 100; i++) {
    // fit.iterate();
    // double[] ps = fit.getParams();
    // System.out.println("Mean: " + ps[0] + " Stddev: " + ps[1] + " Amp: " +
    // ps[2]+" Chi: "+fit.getChiSq());
    // }
    fit.run();
    // double[] ps = fit.getParams();
    // System.out.println("Result: "+ps[0]+" "+ps[1]+" "+ps[2]+" Chi: "+fit.getChiSq()+" Iter: "+fit.maxruns);
    return fit.getParams();
  }

View Full Code Here

TOP

Related Classes of de.lmu.ifi.dbs.elki.math.linearalgebra.fitting.LevenbergMarquardtMethod

de.lmu.ifi.dbs.elki.math.linearalgebra.LinearEquationSystem

de.lmu.ifi.dbs.elki.math.TestKernelDensityFitting

de.lmu.ifi.dbs.elki.math.TestLevenbergMarquardtGaussianFitting

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.