ec.app.star.README Maven / Gradle / Ivy

Go to download

Show more of this group Show more artifacts with this name
Show all versions of ecj Show documentation

ECJ, A Java-based Evolutionary Computation Research System. ECJ is a research EC system written in Java. It was designed to be highly flexible, with nearly all classes (and all of their settings) dynamically determined at runtime by a user-provided parameter file. All structures in the system are arranged to be easily modifiable. Even so, the system was designed with an eye toward efficiency. ECJ is developed at George Mason University's ECLab Evolutionary Computation Laboratory. The software has nothing to do with its initials' namesake, Evolutionary Computation Journal. ECJ's sister project is MASON, a multi-agent simulation system which dovetails with ECJ nicely.

The newest version!

This directory contains three examples for the ECJ master/slave evaluator.
See ec/eval/README for more information on how to run the evaluator
and the ec/eval/master.params file for more parameter information and
options.

NOTE: these examples are worst-case scenarios for the master/slave
evaluator -- they evaluate rapidly and so the time is CONSUMED by
transfering over the net.  The examples are only here to give you 
the idea of how to get your own jobs up and running.

NOTE ABOUT JOBS: ECJ has recently had its job-size parameter set to
1 by default (it used to be 100).  Likewise the max-jobs-per-slave
parameter is now 1 by default (it used to be 3).  These are safer
default settings but they may slow you down.  Consult the manual
about how and when to increase these values.


Artificial Ant (change the eval.master.host in ant.slave.params to the
	appropriate IP of the master's machine.  By default it's
	127.0.0.1, which only works if all the slaves are on the
	same machine as the host; this configuration is likely 
	only useful for testing purposes)

MASTER:		java ec.Evolve -file ant.master.params
EACH SLAVE:	java ec.eval.Slave -file ant.slave.params



Coevolve1 - an example of competitive coevolution using the
	master-slave architecture. (change the eval.master.host 
	in coevolve1.slave.params to the
        appropriate IP of the master's machine.  By default it's
        127.0.0.1, which only works if all the slaves are on the
        same machine as the host; this configuration is likely 
        only useful for testing purposes). 

MASTER:         java ec.Evolve -file coevolve1.master.params
EACH SLAVE:     java ec.eval.Slave -file coevolve1.slave.params


Coevolve2 - an example of cooperative coevolution using the
	master-slave architecture. (change the eval.master.host 
	in coevolve2.slave.params to the
        appropriate IP of the master's machine.  By default it's
        127.0.0.1, which only works if all the slaves are on the
        same machine as the host; this configuration is likely
        only useful for testing purposes).

MASTER:         java ec.Evolve -file coevolve2.master.params
EACH SLAVE:     java ec.eval.Slave -file coevolve2.slave.params



Meta-level Evolution -- an example of combining the meta-level
evolutionary facility with the master-slave architecture.  You
will need to change eval.master.host in slavemeta.params to the
appropriate IP of the master machine.  By default it's
127.0.0.1, which only works if all the slaves are on the
same machine as the host; this configuration is likely only
useful for testing purposes.

MASTER:		java ec.Evolve -file mastermeta.params
EACH SLAVE:	java ec.eval.Slave -file slavemeta.params