AFAIK, you'll have to write your own Statistics object to do this.  It's not too hard—an explanation of how to do so is found in Tutorial 3.

Personally, the first thing I often do when creating an ECJ application is to write a Statistics object that records the genomes and fitnesses of the entire population for every generation.  That creates a ton of data, but it makes it easy to ask whatever questions I want after a run completes.


On Sun, May 21, 2017 at 11:54 AM, Norliza Abdullah <[log in to unmask]> wrote:
Hi all,

 I am implementing Regression Problem using GP. Is it possible for me to obtain the second best individual?

Thank you.

Best Wishes,



Ph.D student in Computer Science, George Mason University
CFO and Web Director, Journal of Mason Graduate Research