Esempio n. 1
0
    private static void test01()

    //****************************************************************************80
    //
    //  Purpose:
    //
    //    TEST01 interpolates in 1D.
    //
    //  Licensing:
    //
    //    This code is distributed under the GNU LGPL license.
    //
    //  Modified:
    //
    //    30 September 2012
    //
    //  Author:
    //
    //    John Burkardt
    //
    {
        int i;
        int j;

        Console.WriteLine("");
        Console.WriteLine("TEST01:");
        Console.WriteLine("  Interpolate in 1D, using orders.");
        Console.WriteLine("  LAGRANGE_INTERP_ND_GRID sets the interpolant.");
        Console.WriteLine("  LAGRANGE_INTERP_ND_VALUE evaluates it.");

        int m = 1;

        int[] n_1d = new int[m];
        for (i = 0; i < m; i++)
        {
            n_1d[i] = 5;
        }

        double[] a = new double[m];
        double[] b = new double[m];
        for (i = 0; i < m; i++)
        {
            a[i] = 0.0;
            b[i] = 1.0;
        }

        int nd = LagrangenD.lagrange_interp_nd_size(m, n_1d);

        double[] xd = LagrangenD.lagrange_interp_nd_grid(m, n_1d, a, b, nd);
        double[] zd = f_sinr(m, nd, xd);
        //
        //  Evaluate.
        //
        Console.WriteLine("");
        Console.WriteLine("         Zinterp          Zexact      Error");
        Console.WriteLine("");

        int ni   = 5;
        int seed = 123456789;

        double[] xi = UniformRNG.r8mat_uniform_01_new(m, ni, ref seed);
        double[] ze = f_sinr(m, ni, xi);
        double[] zi = LagrangenD.lagrange_interp_nd_value(m, n_1d, a, b, nd, zd, ni, xi);

        for (j = 0; j < ni; j++)
        {
            Console.WriteLine("  " + zi[j].ToString(CultureInfo.InvariantCulture).PadLeft(14)
                              + "  " + ze[j].ToString(CultureInfo.InvariantCulture).PadLeft(14)
                              + "  " + Math.Abs(zi[j] - ze[j]).ToString(CultureInfo.InvariantCulture).PadLeft(10) + "");
        }
    }
Esempio n. 2
0
    private static void test08()

    //****************************************************************************80
    //
    //  Purpose:
    //
    //    TEST08 repeats test 4 using levels.
    //
    //  Licensing:
    //
    //    This code is distributed under the GNU LGPL license.
    //
    //  Modified:
    //
    //    30 September 2012
    //
    //  Author:
    //
    //    John Burkardt
    //
    {
        int i;
        int l;

        Console.WriteLine("");
        Console.WriteLine("TEST08:");
        Console.WriteLine("  Interpolate in 3D, using levels.");
        Console.WriteLine("  Use a sequence of increasing levels.");

        const int m = 3;

        int[] ind = new int[m];

        double[] a = new double[m];
        double[] b = new double[m];
        for (i = 0; i < m; i++)
        {
            a[i] = 0.0;
            b[i] = 1.0;
        }

        int ni   = 20;
        int seed = 123456789;

        double[] xi = UniformRNG.r8mat_uniform_01_new(m, ni, ref seed);
        double[] ze = f_sinr(m, ni, xi);

        Console.WriteLine("");
        Console.WriteLine("  Level     Order   Average Error");
        Console.WriteLine("");

        for (l = 0; l <= 5; l++)
        {
            for (i = 0; i < m; i++)
            {
                ind[i] = l;
            }

            int      nd = LagrangenD.lagrange_interp_nd_size2(m, ind);
            double[] xd = LagrangenD.lagrange_interp_nd_grid2(m, ind, a, b, nd);
            double[] zd = f_sinr(m, nd, xd);
            //
            //  Evaluate.
            //
            double[] zi = LagrangenD.lagrange_interp_nd_value2(m, ind, a, b, nd, zd, ni, xi);

            double e = typeMethods.r8vec_norm_affine(ni, zi, ze) / ni;

            Console.WriteLine("  " + l.ToString(CultureInfo.InvariantCulture).PadLeft(5)
                              + "  " + nd.ToString(CultureInfo.InvariantCulture).PadLeft(5)
                              + "  " + e.ToString(CultureInfo.InvariantCulture).PadLeft(10) + "");
        }
    }
Esempio n. 3
0
    private static void test09()

    //****************************************************************************80
    //
    //  Purpose:
    //
    //    TEST09 interpolates in 3D, using anisotropic resolution.
    //
    //  Licensing:
    //
    //    This code is distributed under the GNU LGPL license.
    //
    //  Modified:
    //
    //    30 September 2012
    //
    //  Author:
    //
    //    John Burkardt
    //
    {
        int i;
        int l;

        Console.WriteLine("");
        Console.WriteLine("TEST09:");
        Console.WriteLine("  Interpolate in 3D, using orders.");
        Console.WriteLine("  Use a sequence of increasing orders.");
        Console.WriteLine("  Use anisotropic resolution.");
        Console.WriteLine("  The interpoland is a polynomial of degrees 3, 5, 2");
        Console.WriteLine("  so our orders need to be at least 4, 6, 3 to match it.");

        const int m = 3;

        int[] n_1d = new int[m];

        double[] a = new double[m];
        double[] b = new double[m];
        for (i = 0; i < m; i++)
        {
            a[i] = 0.0;
            b[i] = 1.0;
        }

        int ni   = 20;
        int seed = 123456789;

        double[] xi = UniformRNG.r8mat_uniform_01_new(m, ni, ref seed);
        double[] ze = f_poly352(m, ni, xi);

        Console.WriteLine("");
        Console.WriteLine("  Level     Orders   Average Error");
        Console.WriteLine("");

        for (l = 0; l <= 10; l++)
        {
            switch (l)
            {
            case 0:
                n_1d[0] = 1;
                n_1d[1] = 1;
                n_1d[2] = 1;
                break;

            case 1:
                n_1d[0] = 2;
                n_1d[1] = 1;
                n_1d[2] = 1;
                break;

            case 2:
                n_1d[0] = 1;
                n_1d[1] = 2;
                n_1d[2] = 1;
                break;

            case 3:
                n_1d[0] = 1;
                n_1d[1] = 1;
                n_1d[2] = 2;
                break;

            case 4:
                n_1d[0] = 4;
                n_1d[1] = 2;
                n_1d[2] = 2;
                break;

            case 5:
                n_1d[0] = 2;
                n_1d[1] = 4;
                n_1d[2] = 2;
                break;

            case 6:
                n_1d[0] = 2;
                n_1d[1] = 2;
                n_1d[2] = 4;
                break;

            case 8:
                n_1d[0] = 6;
                n_1d[1] = 4;
                n_1d[2] = 4;
                break;

            case 9:
                n_1d[0] = 4;
                n_1d[1] = 6;
                n_1d[2] = 4;
                break;

            case 10:
                n_1d[0] = 4;
                n_1d[1] = 4;
                n_1d[2] = 6;
                break;
            }

            int      nd = LagrangenD.lagrange_interp_nd_size(m, n_1d);
            double[] xd = LagrangenD.lagrange_interp_nd_grid(m, n_1d, a, b, nd);
            double[] zd = f_poly352(m, nd, xd);
            //
            //  Evaluate.
            //
            double[] zi = LagrangenD.lagrange_interp_nd_value(m, n_1d, a, b, nd, zd, ni, xi);

            double e = typeMethods.r8vec_norm_affine(ni, zi, ze) / ni;

            Console.WriteLine("  " + l.ToString(CultureInfo.InvariantCulture).PadLeft(5)
                              + "  " + n_1d[0].ToString(CultureInfo.InvariantCulture).PadLeft(5)
                              + "  " + n_1d[1].ToString(CultureInfo.InvariantCulture).PadLeft(5)
                              + "  " + n_1d[2].ToString(CultureInfo.InvariantCulture).PadLeft(5)
                              + "  " + e.ToString(CultureInfo.InvariantCulture).PadLeft(10) + "");
        }
    }
Esempio n. 4
0
    private static void test01(int m, int sparse_max)

    //****************************************************************************80
    //
    //  Purpose:
    //
    //    TEST01: sequence of sparse interpolants to an M-dimensional function.
    //
    //  Discussion:
    //
    //    We have functions that can generate a Lagrange interpolant to data
    //    in M dimensions, with specified order or level in each dimension.
    //
    //    We use the Lagrange function as the inner evaluator for a sparse
    //    grid procedure.
    //
    //    The procedure computes sparse interpolants of levels 0 to SPARSE_MAX
    //    to a given function.
    //
    //  Licensing:
    //
    //    This code is distributed under the GNU LGPL license.
    //
    //  Modified:
    //
    //    06 October 2012
    //
    //  Author:
    //
    //    John Burkardt
    //
    //  Parameters:
    //
    //    Local, int M, the spatial dimension.
    //
    //    Input, int SPARSE_MAX, the maximum sparse grid level to try.
    //
    //  Local Parameters:
    //
    //    Local, double A[M], B[M], the upper and lower variable limits
    //    in each dimension.
    //
    //    Local, double APP_ERROR, the averaged Euclidean norm of the
    //    difference between the sparse interpolant and the exact function at
    //    the interpolation points.
    //
    //    Local, int C[L_MAX+1], the sparse grid coefficient vector.
    //    Results at level L are weighted by C(L).
    //
    //    Local, int IND[M], the 1D indices defining a Lagrange grid.
    //    Each entry is a 1d "level" that specifies the order of a
    //    Clenshaw Curtis 1D grid.
    //
    //    Local, int L, the current Lagrange grid level.
    //
    //    Local, int L_MAX, the current sparse grid level.
    //
    //    Local, int MORE, used to control the enumeration of all the
    //    Lagrange grids at a current grid level.
    //
    //    Local, int ND, the number of points used in a Lagrange grid.
    //
    //    Local, int ND_TOTAL, the total number of points used in all the
    //    Lagrange interpolants for a given sparse interpolant points that occur
    //    in more than one Lagrange grid are counted multiple times.
    //
    //    Local, int NI, the number of interpolant evaluation points.
    //
    //    Local, int SPARSE_MIN, the minimum sparse grid level to try.
    //
    //    Local, double XD[M*ND], the data points for a Lagrange grid.
    //
    //    Local, double XI[M*NI], the interpolant evaluation points.
    //
    //    Local, double ZD[ND], the data values for a Lagrange grid.
    //
    //    Local, double ZE[NI], the exact function values at XI.
    //
    //    Local, double ZI[NI], the sparse interpolant values at XI.
    //
    //    Local, double ZPI[NI], one set of Lagrange interpolant values at XI.
    //
    {
        int h = 0;
        int i;
        int l_max;
        int t = 0;

        Console.WriteLine("");
        Console.WriteLine("TEST01:");
        Console.WriteLine("  Sparse interpolation for a function f(x) of M-dimensional argument.");
        Console.WriteLine("  Use a sequence of sparse grids of levels 0 through SPARSE_MAX.");
        Console.WriteLine("  Invoke a general Lagrange interpolant function to do this.");
        Console.WriteLine("");
        Console.WriteLine("  Compare the exact function and the interpolants at a grid of points.");
        Console.WriteLine("");
        Console.WriteLine("  The \"order\" is the sum of the orders of all the product grids");
        Console.WriteLine("  used to make a particular sparse grid.");
        //
        //  User input.
        //
        Console.WriteLine("");
        Console.WriteLine("  Spatial dimension M = " + m + "");
        Console.WriteLine("  Maximum sparse grid level = " + sparse_max + "");
        //
        //  Define the region.
        //
        double[] a = new double[m];
        double[] b = new double[m];

        for (i = 0; i < m; i++)
        {
            a[i] = 0.0;
            b[i] = 1.0;
        }

        //
        //  Define the interpolation evaluation information.
        //
        const int ni   = 100;
        int       seed = 123456789;

        double[] xi = UniformRNG.r8mat_uniform_abvec_new(m, ni, a, b, ref seed);

        Console.WriteLine("  Number of interpolation points is NI = " + ni + "");

        double[] ze = f_sinr(m, ni, xi);
        //
        //  Compute a sequence of sparse grid interpolants of increasing level.
        //
        Console.WriteLine("");
        Console.WriteLine("   L     Order    ApproxError");
        Console.WriteLine("");

        int[]    ind = new int[m];
        double[] zi  = new double[ni];

        const int sparse_min = 0;

        for (l_max = sparse_min; l_max <= sparse_max; l_max++)
        {
            int[] c = new int[l_max + 1];
            int[] w = new int[l_max + 1];
            Smolyak.smolyak_coefficients(l_max, m, ref c, ref w);

            for (i = 0; i < ni; i++)
            {
                zi[i] = 0.0;
            }

            int nd_total = 0;

            int l_min = Math.Max(l_max + 1 - m, 0);

            int l;
            for (l = l_min; l <= l_max; l++)
            {
                bool more = false;
                while (true)
                {
                    //
                    //  Define the next product grid.
                    //
                    Comp.comp_next(l, m, ref ind, ref more, ref h, ref t);
                    //
                    //  Count the grid, find its points, evaluate the data there.
                    //
                    int      nd = LagrangenD.lagrange_interp_nd_size2(m, ind);
                    double[] xd = LagrangenD.lagrange_interp_nd_grid2(m, ind, a, b, nd);
                    double[] zd = f_sinr(m, nd, xd);
                    //
                    //  Use the grid to evaluate the interpolant.
                    //
                    double[] zpi = LagrangenD.lagrange_interp_nd_value2(m, ind, a, b, nd, zd, ni, xi);
                    //
                    //  Weighted the interpolant values and add to the sparse grid interpolant.
                    //
                    nd_total += nd;
                    for (i = 0; i < ni; i++)
                    {
                        zi[i] += c[l] * zpi[i];
                    }

                    if (!more)
                    {
                        break;
                    }
                }
            }

            //
            //  Compare sparse interpolant and exact function at interpolation points.
            //
            double app_error = typeMethods.r8vec_norm_affine(ni, zi, ze) / ni;

            Console.WriteLine("  " + l.ToString(CultureInfo.InvariantCulture).PadLeft(2)
                              + "  " + nd_total.ToString(CultureInfo.InvariantCulture).PadLeft(8)
                              + "  " + app_error.ToString(CultureInfo.InvariantCulture).PadLeft(8) + "");
        }
    }