Intel® oneAPI Math Kernel Library
Ask questions and share information with other developers who use Intel® Math Kernel Library.
7015 Discussions

How to compute the max eigenvalue of one real matrix?

wang__ningning
Beginner
1,442 Views

How to compute the max eigenvalue of one real matrix? I only need the max eigenvalue, and best a function with fewer arguments in the calling sequence.

Labels (2)
0 Kudos
4 Replies
Ruqiu_C_Intel
Moderator
1,411 Views

Hello wang__ningning,


There are lots of functions to compute eigenvalue, for example ?sygvx can select eigenvalues by specifying either or range of indices for the desired eigenvalues. For more details please refer to the MKL developer reference: https://software.intel.com/content/www/us/en/develop/documentation/mkl-developer-reference-fortran/top/lapack-routines/lapack-least-squares-and-eigenvalue-problem-routines/lapack-least-squares-and-eigenvalue-problem-driver-routines/generalized-symmetric-definite-eigenvalue-problems-lapack-driver-routines/sygvx.html


Regards,

Ruqiu




0 Kudos
wang__ningning
Beginner
1,404 Views

Yeah, I know there are many functions for Fortran 95 to compute the eigenvalue of real matrix. But in my test, the syevd is the most fast function than other functions, such as the sygvx, which can select eigenvalues.  So I want to konw which function is the most efficient to compute the eigenvalues?

0 Kudos
Ruqiu_C_Intel
Moderator
1,212 Views

 Hello Wang_Ningning,

Have you checked this function below in the developer reference for computing the largest/smallest eigenvalues and corresponding eigenvectors of a standard eigenvalue problem:

mkl_sparse_?_ev (intel.com)

 

and set the first input parameter "which" as:

 
Indicates eigenvalues for which to search:
  • which = 'L'
     indicates the largest eigenvalues.
  • which = 'S'
     indicates the smallest eigenvalues.

0 Kudos
MRajesh_intel
Moderator
1,165 Views

Hi,


Since we didn't hear back from you, we are closing this thread for now. If you require any additional assistance from Intel, please start a new thread. Any further interaction in this thread will be considered community only.


Have a Good day.

Regards

Rajesh


0 Kudos
Reply