2
$\begingroup$

How can I convert matrices L and U output by dgssvx() of SuperLU to triples format (to matrix market format)? Also how can I convert input matrix A in triples format (in matrix market format) to the format required by SuperLU? What is the easiest way to perform these format conversions?

$\endgroup$
2
  • $\begingroup$ The answer to this question can be found by reading the SuperLU user guide. $\endgroup$
    – pv.
    Commented May 24, 2013 at 20:12
  • $\begingroup$ Dear @py, is there a solution using pysparse or scipy? $\endgroup$
    – Kadir
    Commented May 25, 2013 at 3:29

1 Answer 1

1
$\begingroup$

I would assume that the interfaces had changed and developed within the last 5 years. As of right now, the result can be achieved using the standard functionality of scipy (as requested in the comments).

  • Use scipy.io.mmread to load the matrix from MatrixMarket file into sparse or dense scipy matrix. I would assume that your matrix is originally sparse, therefore you are good to go.
  • Now, scipy.sparse.linalg.splu can be used (or iLU if you opt for incomplete factorization).

Conversely, you can use

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.