[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Tue, 10 Apr 2007 11:38:21 +0200
the online documentation gives the impression that
gsl_linalg_LU_invert() will invert a matrix previously
postprocessed with gsl_linalg_LU_decomp().
I don't quite see how that can be possible, as
gsl_linalg_LU_decomp() will destroy the original matrix
(transform it into an upper triagonal), and doesn't
save enough information to "reconstruct" it.
(only the permutation of the matrix rows is saved, but not
the lower triagonal part (U))
Same goes to the function gsl_linalg_LU_solve().
Is this a bug, or am I overlooking something? In any case,
I can get neither of these functions to work correctly.
(A small example C file which should output a
unit matrix (but doesn't) is attached)
Description: Text Data
- [Help-gsl] gsl_linalg_LU_invert,
Matthias Kramm <=