Re: [Bug-gsl] Not correct norm in Jacobi method for eigen-problem

[Brian Gough]

At Wed, 09 Mar 2011 11:57:02 -0500,
Nikolay Simakov wrote:
> I was messing around eigenvalue problems and I found that the norm for 
> Jacobi  method (eigenvalues calculations) is done for whole matrix, but 
> should be done only for off-diagonal elements. That's very crucial since 
> this norm used to move out from Jacobi iteration cycle, and that's the 
> whole idea to make off-diagonal elements zero.
> 
> I knew that Jacobi is not fast enough but if it is a simple method and 
> since it is  already in gsl why not to fix it.

Thanks for the bug report, I've just committed a fix for it.

-- 
Brian Gough