help-octave
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Genetic Algorith Classification


From: George Kousiouris
Subject: Re: Genetic Algorith Classification
Date: Thu, 27 Feb 2014 14:41:02 +0200
User-agent: Mozilla/5.0 (Windows NT 6.2; WOW64; rv:24.0) Gecko/20100101 Thunderbird/24.3.0


hi,

I have used both (Matlab and Octave) ga packages, from experience the options are pretty much the same (so you can get an idea of them from matlab docs). But there are also some differences, mainly in the way the population is created and passed in the internal function in each case. AFAICR, one tool passed the entire population as an array (so you should handle the iteration over the population members inside the internal script that calculates the function) while the other passed the population member by member.

Furthermore, beware that Octave's ga version seemed to lose after some generations (at least in the version i used 3-4 years ago) the ability to limit the created population members values to a fixed normalized interval e.g. 0-1.

And a general question: are you sure you can use GA's for classification? I have not heard of such a use case before, GA's are normally used for optimization purposes. If you want classification, you can look at Artificial Neural Networks, Support Vector Machines etc.

BR,
George

On 2/27/2014 3:37 AM, mpender wrote:
mina wrote
Hello. I need to use GA to classify some data into two categories (in the
way K-means does).
I want to use octave ga package but I can't find any examples or some
tutorial on how to use it. Could anyone suggest any sources or tips
please?
I found a few references on the Internet, but essentially, everything points
back to the Matlab documentation for ga as being the basis for
compatibility:
http://www.mathworks.com/help/gads/ga.html





--
View this message in context: 
http://octave.1599824.n4.nabble.com/Genetic-Algorith-Classification-tp4662364p4662385.html
Sent from the Octave - General mailing list archive at Nabble.com.
_______________________________________________
Help-octave mailing list
address@hidden
https://mailman.cae.wisc.edu/listinfo/help-octave





---
This email is free from viruses and malware because avast! Antivirus protection 
is active.
http://www.avast.com



reply via email to

[Prev in Thread] Current Thread [Next in Thread]