|
|
| Re: Lanczos diagonalisation with IDL [message #51812 is a reply to message #51811] |
Thu, 07 December 2006 20:13  |
Richard French
Messages: 173
Registered: December 2000
|
Senior Member |
|
|
You might check out the IMSL routines in IDL Analyst - extra bucks for this
software, I know, but it might be worth it. For example, IMSL_SP_LUSOL
solves a sparse system of linear equations without having to specify the
full sparse matrix. IMSL_SP_LUFAC computes an LU factorization of a sparse
matrix without having to store the full matrix. The documentation for IDL
Analyst is available on the RSI web site someplace.
Dick French
On 12/7/06 8:13 AM, in article
1165497198.963840.266160@l12g2000cwl.googlegroups.com,
"philip.tregenna@psi.ch" <philip.tregenna@psi.ch> wrote:
> I would like to diagonalise a large, sparse, symmetric matrix, and find
> a limited set of eigenvalues. The best way to do this is to use an
> algorithm based upon the Lanczos method. Has a Lanczos diagonalisation
> routine been written in the IDL language? The procedure with a
> functionality close to that which I need is "LA_EIGENQL". However,
> this routine is not ideal for large sparse matrices as it requires as
> input all the elements of the matrix, including those with a value of
> zero.
>
> Regards,
>
> Philip.
>
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
