On Thursday, January 9, 2014 2:33:54 PM UTC-8, David Roe wrote:
>
> Hm.  I agree with you.  What about having two methods, eigenvalues_exact 
> and eigenvalues_approx?  Working in the splitting field seems like a great 
> choice when the degree is small, and a horrible choice when the degree is 
> large.  Currently eigenvalues has "extend" as a keyword parameter, which 
> can be True or False.  Perhaps allow it to also be something a string 
> "splitting_field" or a parent for the eigenvalues to live in?
> David
>

We are just talking about shorthands for

M.characteristic_polynomial().roots(<FIELD>)

here. By analogy

M.eigenvalues(<FIELD>)

would seem reasonable. In fact, for exact fields there doesn't seem to be a 
good rationale to use "eigenvalues" at all: just look at the characteristic 
polynomial if you have a serious application that goes beyond an 
introductory linear algebra class.

Of course, for matrices over inexact fields, it would make a lot of sense 
to have a separate "eigenvalues" routine that tries to take care of 
numerical stability.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-nt" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send an email to [email protected].
Visit this group at http://groups.google.com/group/sage-nt.
For more options, visit https://groups.google.com/d/optout.

Reply via email to