from a Chinese user of JGAP

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

from a Chinese user of JGAP

张慧涛
Hello!

I am a new user of JGAP from China. I have spent a whole week to study your programs. Thank you for sharing your work. It's wonderful. However, I found a bug in one of your examples named "MathProblem.java", In this example, the fitness was defined as an error between the selected value and the target value, which means the greater the error is, the greater the fitness is. And in evolve(), the class "FitnessProportionateSelection" which was used to select individual defined that the individual with greater fitness had larger possibility to be selected. As the result of two definations above, the closer the value to the target, the less possible it to be selected. I think it was wrong. I am looking forward to your response.

I appreciate your help.

Regards

                                                                                                Huitao Zhang




-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
jgap-announce mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/jgap-announce