|
From: | Nir Krakauer |
Subject: | [Octave-bug-tracker] [bug #39903] Error solving linear systems in single precision |
Date: | Fri, 30 Aug 2013 14:42:23 +0000 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/28.0.1500.95 Safari/537.36 |
URL: <http://savannah.gnu.org/bugs/?39903> Summary: Error solving linear systems in single precision Project: GNU Octave Submitted by: nir_krakauer Submitted on: Fri 30 Aug 2013 02:42:22 PM GMT Category: None Severity: 3 - Normal Priority: 5 - Normal Item Group: Incorrect Result Status: None Assigned to: None Originator Name: Nir Krakauer Originator Email: Open/Closed: Open Discussion Lock: Any Release: 3.6.2 Operating System: GNU/Linux _______________________________________________________ Details: I've been getting sporadic errors of this sort. With the attached example data: load demo.mat %contains single-precision arrays a and b c = a \ b; ** On entry to SLASD4 parameter number -1 had an illegal value error: exception encountered in Fortran subroutine sgelsd_ However, either of the following does work without problem: c = pinv(a) * b; c = double(a) \ double(b); _______________________________________________________ File Attachments: ------------------------------------------------------- Date: Fri 30 Aug 2013 02:42:22 PM GMT Name: demo.mat Size: 119kB By: nir_krakauer <http://savannah.gnu.org/bugs/download.php?file_id=28960> _______________________________________________________ Reply to this item at: <http://savannah.gnu.org/bugs/?39903> _______________________________________________ Message sent via/by Savannah http://savannah.gnu.org/
[Prev in Thread] | Current Thread | [Next in Thread] |