What is the best function to obtain a least squares minimum solution from a linear problem like Ax = b in Octave, with A very large but sparse?
x = A\b gives the error:
SparseQR: sparse matrix QR factorization filled" that I don't understand.
Asked
Active
Viewed 466 times
0
IKavanagh
- 6,089
- 11
- 42
- 47
andreaconsole
- 430
- 3
- 17
-
"A" matrix is a matrix of square matrices, n*3 matrices: for each row, the first is a diagonal one, the second is simply sparse and the third is an eye matrix – andreaconsole May 28 '13 at 17:50
-
to be really honest, I delete some rows before computation (they are data from real world, so not everything is perfect...), thus the matrix structure is not exactly the one I described – andreaconsole May 29 '13 at 09:10
-
1not an expert in `octave` but I suppose that you do not have enough memory for the QR factorization of `A` (used for the Dulmage-Mendelsohn decomposition, see the octave [docs](http://www.gnu.org/software/octave/doc/interpreter/Sparse-Linear-Algebra.html). May be this question is better suited for http://scicomp.stackexchange.com – Stefano M May 30 '13 at 07:40
-
What version of Octave are you running? Is it installed with libsuitesparse? I do not seem to have an issue with randomly generated sparse matrices, but I do not have your data either. – Engineero Jun 09 '13 at 18:02