comp.lang.idl-pvwave archive
Messages from Usenet group comp.lang.idl-pvwave, compiled by Paulo Penteado

Home » Public Forums » archive » Re: TNMIN limits
Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend 
Return to the default flat view Create a new topic Submit Reply
Re: TNMIN limits [message #57119 is a reply to message #57118] Mon, 03 December 2007 05:08 Go to previous messageGo to previous message
Brian Larsen is currently offline  Brian Larsen
Messages: 270
Registered: June 2006
Senior Member
While admittedly I have not used TNMIN before, it is often good
practice to use several different methods of function minimization on
the given problem. There are issues associated with each solver, what
it requires ad what it can handle. For instance there is a class of
solvers that require derivatives and a class that don't. If your
function has regions that are not differential then that method will
have issues... there is a list in idl of built-in minimization
routines (there are 5-6). Everyone has their own favorite but I often
use ameoba() first, it doesn't need derivatives and seems to work well
and fast. The function that you are minimizing has to be written to
enforce limits on variables but thats not too bad to do,

a few web resources:
ameoba (a c code base but lots of good information)
http://solar.physics.montana.edu/kankel/ph567/examples/minim ization/
conjugate-gradient:
http://solar.physics.montana.edu/kankel/ph567/resources/conj ugate_gradient/


Cheers,

Brian

------------------------------------------------------------ --------------
Brian Larsen
Boston University
Center for Space Physics
[Message index]
 
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic: Re: IDL 6.4 seg-faults when drawing (Solaris-8)
Next Topic: TNMIN limits

-=] Back to Top [=-
[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Fri Oct 10 23:47:00 PDT 2025

Total time taken to generate the page: 3.84250 seconds