class TDecompLU: public TDecompBase

 Default constructor

 Constructor for (nrows x nrows) matrix

 Constructor for ([row_lwb..row_upb] x [row_lwb..row_upb]) matrix

 Constructor for matrix a

 Copy constructor

 Matrix A is decomposed in components U and L so that P * A = U * L
 If the decomposition succeeds, bit kDecomposed is set , otherwise kSingular

 Reconstruct the original matrix using the decomposition parts

 Set matrix to be decomposed

 Solve Ax=b assuming the LU form of A is stored in fLU, but assume b has *not*
 been transformed.  Solution returned in b.

 Solve Ax=b assuming the LU form of A is stored in fLU, but assume b has *not*
 been transformed.  Solution returned in b.

 Solve A^T x=b assuming the LU form of A^T is stored in fLU, but assume b has *not*
 been transformed.  Solution returned in b.

 Solve A^T x=b assuming the LU form of A^T is stored in fLU, but assume b has *not*
 been transformed.  Solution returned in b.

 Calculate determinant det = d1*TMath::Power(2.,d2)

 For a matrix A(m,m), its inverse A_inv is defined as A * A_inv = A_inv * A = unit
 (m x m) Ainv is returned .

 For a matrix A(m,n), its inverse A_inv is defined as A * A_inv = A_inv * A = unit
 (n x m) Ainv is returned .

Print internals of this object

assignement operator

 Crout/Doolittle algorithm of LU decomposing a square matrix, with implicit partial
 pivoting.  The decomposition is stored in fLU: U is explicit in the upper triag
 and L is in multiplier form in the subdiagionals .
 Row permutations are mapped out in fIndex. fSign, used for calculating the
 determinant, is +/- 1 for even/odd row permutations. .

 LU decomposition using Gaussain Elimination with partial pivoting (See Golub &
 Van Loan, Matrix Computations, Algorithm 3.4.1) of a square matrix .
 The decomposition is stored in fLU: U is explicit in the upper triag and L is in
 multiplier form in the subdiagionals . Row permutations are mapped out in fIndex.
 fSign, used for calculating the determinant, is +/- 1 for even/odd row permutations.
 Since this algorithm uses partial pivoting without scaling like in Crout/Doolitle.
 it is somewhat faster but less precise .

 Calculate matrix inversion through in place forward/backward substitution