private static void sgmga_vcn_coef_test(int dim_num, double[] importance,
double[] level_weight, int level_max_min, int level_max_max)
//****************************************************************************80
//
// Purpose:
//
// SGMGA_VCN_COEF_TEST tests SGMGA_VCN_COEF.
//
// Licensing:
//
// This code is distributed under the GNU LGPL license.
//
// Modified:
//
// 18 May 2010
//
// Author:
//
// John Burkardt
//
{
int dim;
int level_max;
SGMGAniso.SGMGAData data = new();
string cout = "";
int[] level_1d = new int[dim_num];
int[] level_1d_max = new int[dim_num];
int[] level_1d_min = new int[dim_num];
Console.WriteLine("");
Console.WriteLine("SGMGA_VCN_COEF_TEST");
Console.WriteLine(" For anisotropic problems, a \"combinatorial coefficent\"");
Console.WriteLine(" must be computed for each component product grid.");
Console.WriteLine(" SGMGA_VCN_COEF_NAIVE does this in a simple, inefficient way.");
Console.WriteLine(" SGMGA_VCN_COEF tries to be more efficient.");
Console.WriteLine(" Here, we simply compare COEF1 and COEF2, the same");
Console.WriteLine(" coefficient computed by the naive and efficient ways.");
Console.WriteLine("");
Console.WriteLine(" IMPORTANCE:");
cout = "";
for (dim = 0; dim < dim_num; dim++)
{
cout += " " + importance[dim].ToString(CultureInfo.InvariantCulture).PadLeft(14);
}
Console.WriteLine(cout);
Console.WriteLine(" LEVEL_WEIGHT:");
cout = "";
for (dim = 0; dim < dim_num; dim++)
{
cout += " " + level_weight[dim].ToString(CultureInfo.InvariantCulture).PadLeft(14);
}
Console.WriteLine(cout);
for (level_max = level_max_min; level_max <= level_max_max; level_max++)
{
int i = 0;
double coef1_sum = 0.0;
double coef2_sum = 0.0;
//
// Initialization.
//
double level_weight_min_pos = typeMethods.r8vec_min_pos(dim_num, level_weight);
double q_min = level_max * level_weight_min_pos
- typeMethods.r8vec_sum(dim_num, level_weight);
double q_max = level_max * level_weight_min_pos;
for (dim = 0; dim < dim_num; dim++)
{
level_1d_min[dim] = 0;
}
for (dim = 0; dim < dim_num; dim++)
{
switch (level_weight[dim])
{
case > 0.0:
{
level_1d_max[dim] = (int)Math.Floor(q_max / level_weight[dim]) + 1;
if (q_max <= (level_1d_max[dim] - 1) * level_weight[dim])
{
level_1d_max[dim] -= 1;
}
break;
}
default:
level_1d_max[dim] = 0;
break;
}
}
bool more_grids = false;
Console.WriteLine("");
Console.WriteLine(" I Q Coef1 Coef2 X");
cout = " MIN" + " " + q_min.ToString(CultureInfo.InvariantCulture).PadLeft(14)
+ " ";
for (dim = 0; dim < dim_num; dim++)
{
cout += " " + level_1d_min[dim].ToString().PadLeft(2);
}
Console.WriteLine(cout);
//
// Seek all vectors LEVEL_1D which satisfy the constraint:
//
// LEVEL_MAX * LEVEL_WEIGHT_MIN_POS - sum ( LEVEL_WEIGHT )
// < sum ( 0 <= I < DIM_NUM ) LEVEL_WEIGHT[I] * LEVEL_1D[I]
// <= LEVEL_MAX * LEVEL_WEIGHT_MIN_POS.
//
for (;;)
{
SGMGAniso.sgmga_vcn_ordered_naive(ref data, dim_num, level_weight, level_1d_max,
level_1d, q_min, q_max, ref more_grids);
if (!more_grids)
{
break;
}
//
// Compute the combinatorial coefficient.
//
double coef1 = SGMGAniso.sgmga_vcn_coef_naive(dim_num, level_weight, level_1d_max,
level_1d, q_min, q_max);
double coef2 = SGMGAniso.sgmga_vcn_coef(dim_num, level_weight, level_1d, q_max);
i += 1;
double q = 0.0;
for (dim = 0; dim < dim_num; dim++)
{
q += level_weight[dim] * level_1d[dim];
}
coef1_sum += coef1;
coef2_sum += coef2;
cout = " " + i.ToString().PadLeft(4)
+ " " + q.ToString(CultureInfo.InvariantCulture).PadLeft(14)
+ " " + coef1.ToString(CultureInfo.InvariantCulture).PadLeft(10)
+ " " + coef2.ToString(CultureInfo.InvariantCulture).PadLeft(10);
for (dim = 0; dim < dim_num; dim++)
{
cout += " " + level_1d[dim].ToString().PadLeft(2);
}
Console.WriteLine(cout);
}
cout = " MAX" + " " + q_max.ToString(CultureInfo.InvariantCulture).PadLeft(14)
+ " ";
for (dim = 0; dim < dim_num; dim++)
{
cout += " " + level_1d_max[dim].ToString().PadLeft(2);
}
Console.WriteLine(cout);
Console.WriteLine(" SUM "
+ " " + coef1_sum.ToString(CultureInfo.InvariantCulture).PadLeft(10)
+ " " + coef2_sum.ToString(CultureInfo.InvariantCulture).PadLeft(10) + "");
}
}
