Genetic algorithms and scilab

[COLLETTE Yann]

Hello,

I've made a Genetic Algorithm Toolbox. To test it, you will need to 
install some other packages:

- Some test problems : 
http://www.scilab.org/contrib/index_contrib.php?page=displayContribution&fileID=1082
 There are not loader.sce and builder.sce because I'm still implementing 
some test problems (a set of fortran functions from the La Cumparsita 
data set and the minpack2. The first one is nearly finished, I need to 
write the documentation for the other one).
- The toolbox to manage parameters (like with optimset and optimget): 
http://www.scilab.org/contrib/index_contrib.php?page=displayContribution&fileID=1081
 There is a loader.sce file and a builder.sce ...
  It's easier to manage extra parameters via a list of parameters than 
via a long command line. Using the param toolbox, I can leave on the 
command line only the most important parameters. All the fine tuning is 
done by setting up parameters values via the param toolbox.
- The Genetic Algorithm toolbox: 
http://www.scilab.org/contrib/index_contrib.php?page=displayContribution&fileID=1080
 There is a loader.sce file and a builder.sce ...
 In this toolbox, I've implemented some well known algorithms:
   - the classical genetic algorithm which works on binary string 
(tested in the second part of the demo file GAdemo.sce);
   - the evolutionnary algorithm which works directly on the variable 
(tested in the first part of the demo file GAdemo.sce for a continuous 
variable test problem and in the GAIsing2ddemo.sce for a combinatorial 
test problem);
   - MOGA: the multi-objective genetic algorithm. A classical 
multi-objective algorithm, be not really efficient. Tested in MOGAdemo.sce.
   - NSGA: the niched sharing genetic algorithm. A good improvement wrt 
MOGA. Tested in NSGAdemo.sce.
   - NSGA2: the 2nd version of NSGA. Certainly the best multi-objective 
algorithm so far. No extra parameters and quite efficient. See 
NSGA2demo.sce for a demonstration.

The genetic algorithm is certainly the most flexible algorithm so far 
(not the most efficient). You can solve a large spectrum of problems 
with such a method.

My question is: what do I need to do to see a genetic algorithm being 
included in scilab ?

I am also very interested by some feedback to improve all these packages ...

Your sincerely,

Yann COLLETTE


-- Disclaimer ------------------------------------
Ce message ainsi que les eventuelles pieces jointes constituent une correspondance privee et confidentielle a l'attention exclusive du destinataire designe ci-dessus. Si vous n'etes pas le destinataire du present message ou une personne susceptible de pouvoir le lui delivrer, il vous est signifie que toute divulgation, distribution ou copie de cette transmission est strictement interdite. Si vous avez recu ce message par erreur, nous vous remercions d'en informer l'expediteur par telephone ou de lui retourner le present message, puis d'effacer immediatement ce message de votre systeme.
***
This e-mail and any attachments is a confidential correspondence intended only for use of the individual or entity named above. If you are not the intended recipient or the agent responsible for delivering the message to the intended recipient, you are hereby notified that any disclosure, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify the sender by phone or by replying this message, and then delete this message from your system.