September 2008


Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Reply To:
ECJ Evolutionary Computation Toolkit <[log in to unmask]>
Fri, 19 Sep 2008 15:32:59 -0400
text/plain (45 lines)
I'll vote for option #2 since I doubt I'd meet the requirements for  
#3.  :)

I will continue to clean up the code and develop some documentation.   
I hope to have something ready for you to review in a few days.


On Sep 19, 2008, at 11:51 AM, Sean Luke wrote:

> Dave, this is cool to hear.
> So there are three levels we can do this at:
> 1. At the very least I can put a link to your stuff on the ECJ web  
> page.
> 2. We can host the code in the 'contrib' directory.  You'd be in  
> charge of it.
> 3. We can import the code into the main ECJ distribution.  To do  
> this, the code has to match ECJ coding norms and design standards  
> (and be good!)
> What do you think?
> Sean
> David wrote:
>> Sean,
>> I am putting the finishing touches on a rudimentary implementation  
>> of Cartesian Genetic Programming (CGP) for ECJ.  CGP is a genetic  
>> programming technique invented by Julian Miller - more information  
>> here:
>> I am interested in submitting my implementation to the ECJ code  
>> base.  If this is agreeable to you, I was thinking of placing my  
>> implementation under the "contrib" directory.  My hope is that  
>> others can improve and build on this work.
>> I've run a handful of test problems with this CGP implementation:  
>> Even-n-parity, regression, breast cancer classifier, and the  
>> classic Iris species classifier.  I would include these as sample  
>> implementations in the contribution.
>> Let me know what you think!
>> Regards,
>> -Dave