Home »
Public Forums »
archive »
Singular Value Decomposition
Singular Value Decomposition [message #9030] |
Fri, 23 May 1997 00:00 |
Paul Morris
Messages: 3 Registered: May 1997
|
Junior Member |
|
|
SINGULAR VALUE DECOMPOSITION
I was wondering if any one had developed alternatives to the build-in
IDL SVDC - singular value deccomposition routine. I was hoping to find
an IDL SVD something like the `Lanczos bidiagonalization' algorithm
where a reduced number of singular values can be calculated, thus saving
a tremendous amount of computation.
At the moment I am applying SVDC to a large (500,500) array but I only
need to keep a small number of output singular values. If any one out
there has looked at this problem in IDL then it would be great to get in
contact. I have looked at the Pete Riley's `searchable list of IDL
routines' but could not find anything other than the built-in SVDC.
much thanks,
Paul Morris,
Department of Physics,
University of Oxford
|
|
|
Current Time: Fri Oct 10 04:58:56 PDT 2025
Total time taken to generate the page: 1.12293 seconds