JavaCPP Presets for CMINPACK

Introduction

This directory contains the JavaCPP Presets module for:

• CMINPACK 1.3.6 http://devernay.free.fr/hacks/cminpack/

Please refer to the parent README.md file for more detailed information about the JavaCPP Presets.

Documentation

Java API documentation is available here:

• http://bytedeco.org/javacpp-presets/cminpack/apidocs/

Sample Usage

Here is a simple example of CMINPACK ported to Java from this C source file:

• https://github.com/devernay/cminpack/blob/master/examples/tlmdifc.c

We can use Maven 3 to download and install automatically all the class files as well as the native binaries. To run this sample code, after creating the pom.xml and Tlmdif1c.java source files below, simply execute on the command line:

 $ mvn compile exec:java

The pom.xml build file

<project>
    <modelVersion>4.0.0</modelVersion>
    <groupId>org.bytedeco.cminpack</groupId>
    <artifactId>tlmdif1c</artifactId>
    <version>1.5.3</version>
    <properties>
        <exec.mainClass>Tlmdif1c</exec.mainClass>
    </properties>
    <dependencies>
        <dependency>
            <groupId>org.bytedeco</groupId>
            <artifactId>cminpack-platform</artifactId>
            <version>1.3.6-1.5.3</version>
        </dependency>
    </dependencies>
    <build>
        <sourceDirectory>.</sourceDirectory>
    </build>
</project>

The Tlmdif1c.java source file

/*     driver for lmdif1 example. */

import org.bytedeco.javacpp.*;
import org.bytedeco.javacpp.indexer.*;

import static java.lang.Math.*;
import static org.bytedeco.cminpack.global.cminpack.*;

public class Tlmdif1c {
  public static void main(String[] args)
  {
    Loader.load(org.bytedeco.cminpack.global.cminpack.class);

    int info, lwa, iwa[] = new int[3];
    double tol, fnorm, x[] = new double[3], fvec[] = new double[15], wa[] = new double[75];
    int m = 15;
    int n = 3;
    /* auxiliary data (e.g. measurements) */
    double[] y = {1.4e-1, 1.8e-1, 2.2e-1, 2.5e-1, 2.9e-1, 3.2e-1, 3.5e-1,
                    3.9e-1, 3.7e-1, 5.8e-1, 7.3e-1, 9.6e-1, 1.34, 2.1, 4.39};
    /* the following struct defines the data points */
    DoublePointer data = new DoublePointer(y);

    /* the following starting values provide a rough fit. */

    x[0] = 1.;
    x[1] = 1.;
    x[2] = 1.;

    lwa = 75;

    /* set tol to the square root of the machine precision.  unless high
       precision solutions are required, this is the recommended
       setting. */

    tol = sqrt(dpmpar(1));

    info = lmdif1(new Fcn(), data, m, n, x, fvec, tol, iwa, wa, lwa);

    fnorm = enorm(m, fvec);

    System.out.printf("      final l2 norm of the residuals%15.7g\n\n",(double)fnorm);
    System.out.printf("      exit parameter                %10d\n\n", info);
    System.out.printf("      final approximate solution\n\n %15.7g%15.7g%15.7g\n",
          (double)x[0], (double)x[1], (double)x[2]);
    System.exit(0);
  }

  public static class Fcn extends cminpack_func_mn {
    @Override public int call(Pointer p, int m, int n, DoublePointer x, DoublePointer fvec, int iflag)
    {
      /* function fcn for lmdif1 example */

      int i;
      double tmp1,tmp2,tmp3;
      DoublePointer y = new DoublePointer(p);
      assert m == 15 && n == 3;

      DoubleIndexer xIdx = DoubleIndexer.create(x.capacity(n));
      DoubleIndexer yIdx = DoubleIndexer.create(y.capacity(m));
      DoubleIndexer fvecIdx = DoubleIndexer.create(fvec.capacity(m));
      for (i = 0; i < 15; ++i)
        {
          tmp1 = i + 1;
          tmp2 = 15 - i;
          tmp3 = (i > 7) ? tmp2 : tmp1;
          fvecIdx.put(i, y.get(i) - (x.get(0) + tmp1/(x.get(1)*tmp2 + x.get(2)*tmp3)));
        }
      return 0;
    }
  }
}