Lil-gp the 1-th

The original version of lil-gp is the work of Dr. Bill Punch and Douglas Zongker from the Michigan State Unverisity. What I did was, just as many other did before me, download lil-gp from the ftp server and take a good look at their soucre code ( I looked twice :-). Instead of reviewing their excellent software here, I chose to copy the list of features of lil-gp as they can be found on the website of the authors.

Overview of lilgp

• Generic 'C' program that runs on both workstations and PC's (though PC's will require windows, or the djgpp gcc compiler (available here) or better OS's to support the needed memory).
• Support for multiple population experiments, using arbitrary and user settable topologies for exchange, for a single processor(i.e., you can do multiple population gp experiments on your PC).
• lil-gp manipulates trees of function pointers which are allocated in single, large memory blocks for speed and to avoid swapping.

This software package provides you with a framework for genetic programming experiments that is both powerfull and flexible to use.

Since it is a framework the user only needs to provide those functions that are really interesting for his genetic programming research/project. This makes it obviously very easy to set up an genetic programming experiment.

The operation of the genetic programming kernel can be control by means of a configuration file. Most of the parameters have some well documented default value which makes it possible to run experiment without the need of creating a very elaborate configuration file.  Interesting as well is the fact that the configuration file mimics the C syntax. This is very convenient since the user does not need to learn yet another syntax to configure his experiments.

The most prominent feature of this software package (from the parallellisation point of view that is) is the support of multiple population. So what was done is basicly change the line "for a single processor( i.e.,you can do multiple population gp experiments on your PC)" to "for a single processor or multiple processors..."