|
Size: 1448
Comment: converted to 1.6 markup
|
← Revision 3 as of 2017-03-22 00:36:25 ⇥
Size: 0
Comment: is an spkg
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 1: | Line 1: |
| = LinBox = == About == From the website: "LinBox is a C++ template library for exact, high-performance linear algebra computation with dense, sparse, and structured matrices over the integers and over finite fields." (http://www.linalg.org) == URL == http://www.linalg.org == Documentation == Note: Credibility of this information: Clement Pernet from the LinBox team. However, any possible mistakes are due to the respective authors (e.g. [[MartinAlbrecht|Martin Albrecht]]) === Design Description === There are two types of dense matrices in LinBox. The first type is in {{{blas-matrix.h}}}. You will find a DenseMatrix and a BlasMatrix class in there. Use the BlasMatrix class. Those classes are templated with a field as template parameter. You can plug in e.g. {{{Modular<double>}}}, {{{Integer}}}, and even {{{GivaroGfq}}}. The other type are black box matrices. These are templated with an Element as template parameter as it is supposed not to rely on ony field implementation. == Tips and Tricks == * If you are looking for a solution, search the solution subdirectory. You don't have to worry about matrix domains there. If you cannot find what you are looking for look in the algorithms subdirectory. * If you are in the algorithms subdirectory use BlasMatrixDomain for everything dense. (Btw. never ever use the MatrixDomain) * Always use DenseMatrix for dense operations. == Random Benchmarks == |
