Re: [eigen] Matrix power

[ Thread Index | Date Index | More lists.tuxfamily.org/eigen Archives ]


On Tue, Jun 18, 2013 at 11:45 PM, 何震邦 <jdh8@xxxxxxxxxxxxxx> wrote:
> Sorry for my late reply.

No problem!

>
> If A has a zero eigenvalue, there is a zero on diag of T.  Therefore, there is an 1
> on diag of (I - T), enlarging (I - T).norm(), making Eigen sqrt'ing (I - T).
> However, no matter how many times (I - T) is sqrt'ed, the 1 stands still on its
> diag, leading to an infinite loop.
>

Ok, I think see what is going on.

> I will finish my final exam at 8:50 tomorrow UTC.  I will work on this since
> then.

Cool, let me know if you want me to test any patches out.  I'm usually
on IRC #eigen with handle DLPeterson if you want to chat.

> Thanks for your discovery!

It is I who should be thanking you for working on this.  Thank you.

Luke



Mail converted by MHonArc 2.6.19+ http://listengine.tuxfamily.org/