 |
|
List Archives >
Maple User Group List Archive >
Archive by date >
This Month By Date >
This Month By Topic
[MUG] Re: Serious bug in LinearSolve
| [MUG] Re: Serious bug in LinearSolve |
|
Author: Maple User Group
Posted: Tue, 29 Oct 2002 13:33:51 -0500
|
>> From: Maple User Group "maple_gr"
| >> From: Luis Goddyn "goddyn"
| LinearSolve appears to be broken in both Maple 7 and Maple 8,
| and across several platforms.
| The problem appears occationally when the matrix has a real entry
| and the solution is not unique.
-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-
Date: Thu, 24 Oct 2002 16:44:26 -0700 (PDT)
From: Robert Israel "israel"
To: "maple-list"
Subject: Serious bug in LinearSolve
It seems to work with
> LinearSolve(%,method='solve',outputoptions=[datatype=anything]);
(in this example, anyway).
Robert Israel "israel"
Department of Mathematics http://www.math.ubc.ca/~israel
University of British Columbia
Vancouver, BC, Canada V6T 1Z2
-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-
From: "Dr Francis J. Wright" "F.J.Wright"
To: "goddyn"
Subject: Serious bug in LinearSolve
Date: Fri, 25 Oct 2002 15:52:12 +0100
I think this is a problem of numerical instability. As soon as you
introduce floating-point numbers you imply numerical approximation, so the
two identical rows become only approximately identical, the problem becomes
approximately singular and numerical solution procedures run into trouble.
If you have additional information, in this case that two rows really are
identical, you can use that information to make the problem solvable by
removing the redundancy. It seems to work in this case anyway:
evalf(<<1,0>|<1,0>|<0,1>>),evalf(<1,0>); LinearSolve(%);
[1. 1. 0.] [1.]
[ ], [ ]
[0. 0. 1.] [0.]
[1.000000000 - 1. _t0[1]]
[ ]
[ _t0[1] ]
[ ]
[ 0. ]
Francis
---
Dr Francis J. Wright
School of Mathematical Sciences, Queen Mary
University of London, Mile End Road, London E1 4NS, UK
Tel: 020 7882 5453 (direct); Fax: 020 8981 9587 (dept.)
"F.J.Wright" http://centaur.maths.qmul.ac.uk/
-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-
Date: Sun, 27 Oct 2002 16:10:00 -0500 (EST)
From: Denis Sevee "dsevee"
To: "maple-list" "maple-list"
Subject: Serious bug in LinearSolve
Yes, it's annoying. If you haven't noticed yet, the same problem turns up
with GaussianElimination and LUDecomposition, i.e. Maple doesn't move
the rows of zeroes to the bottom.
Denis Sevee
|
[View Complete Thread]
Previous by date: [MUG] Re: Maple and PseudoRandomGenerator, Maple User Group
Next by date: [MUG] Re: Maple 8 under Debian GNU/Linux, Maple User Group
Previous thread: [MUG] solving inequality, M A Suzen
Next thread: [MUG] Maple 8 under Debian GNU/Linux, Curtis Vinson
|
|
|
(NAG)
