/// <summary>
///
/// </summary>
/// <param name="wym">dimensions</param>
/// <param name="s1">basevector, used for rotation</param>
/// <param name="a">current orthogonal vectors</param>
public static void Minim(int wym, double[] s1, double[][] a)
{
double c1;
double[][] b = MatrixKniaz.New(wym, wym);
int i, j, k;
for (i = 0; i < wym; i++)
{
for (j = 0; j < wym; j++)
{
c1 = 0;
for (k = i; k < wym; k++)
{
c1 += s1[k] * a[k][j];
}
b[i][j] = c1;
}
}
for (i = 0; i < wym; i++)
{
for (j = 0; j < wym; j++)
{
a[i][j] = b[i][j];
}
}
}
/// <summary>
/// Performs Orthogonalization of the vector
/// </summary>
/// <param name="wym">dimensions</param>
/// <param name="v">current orthogonal vectors, after "minim" was run with basevector</param>
public static void Ortog(int wym, double[][] v)
{
int i, j, k;
double lask = 0.0;
j = 0;
double[] w1 = new double[wym];
MatrixKniaz.Wers(wym, j, v);
for (i = 1; i < wym; i++)
{
for (k = 0; k < wym; k++)
{
w1[k] = 0;
}
for (j = 0; j <= i - 1; j++)
{
lask = MatrixKniaz.scalarProduct(wym, i, j, v);
for (k = 0; k < wym; k++)
{
w1[k] += v[j][k] * lask;
}
}
for (k = 0; k < wym; k++)
{
v[i][k] += -w1[k];
}
MatrixKniaz.Wers(wym, i, v);
}
}
/// <summary>
///
/// </summary>
/// <param name="k"></param>
/// <param name="l"></param>
/// <param name="a"></param>
public static void Wers(int k, int l, double[][] a)
{
int i;
double c;
c = MatrixKniaz.Norma(k, l, a);
for (i = 0; i < k; i++)
{
a[l][i] = a[l][i] / c;
}
}
public static double[][] GramSchmidt(double[] basevector, int nVar, double[][] vMoves)
{
double[][] rotatedMatrix = new double[nVar][];
for (int i = 0; i < nVar; i++)
{
rotatedMatrix[i] = new double[nVar];
rotatedMatrix[i] = vMoves[i];
}
MatrixKniaz.Minim(nVar, basevector, rotatedMatrix);
MatrixKniaz.Ortog(nVar, rotatedMatrix);
return(rotatedMatrix);
}
