|
From: | John W. Eaton |
Subject: | Re: Singular matrix warning IDs |
Date: | Thu, 18 Dec 2014 11:42:21 -0500 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Icedove/31.2.0 |
On 12/17/2014 11:37 PM, Jordi GutiƩrrez Hermoso wrote:
Don't we already do the approximate equivalent of pinv(A)*b in all cases for A\b? I don't think this needs further explanation. Also, the "minimum norm" solution warning seems to pop up in weird, inconsistent places right now. As far as I can tell, only if you use single precision? I would just drop it, like you said.
I did that here: http://hg.savannah.gnu.org/hgweb/octave/rev/cfe56baccb0c I also added warning IDs for all calls to warning in liboctave, here: http://hg.savannah.gnu.org/hgweb/octave/rev/95c533ed464b jwe
[Prev in Thread] | Current Thread | [Next in Thread] |