Finalist, Rotterdam / 02-02-2011
Introducing GenPro: Genetic Programming made simple
Genetic programming (GP) is the art of generating a software program which fulfills to a given set of test cases. One can use GP for creating formulas that predict or do an estimation of values in certain circumstances: eg predict turnover in a certain month, estimate house value based on some of its properties, estimate the weight of an egg based on its size etc.
The aim of the open source GenPro project is to add Genetic Programming (GP) to the developer’s general toolbox. This opposed to where GP mostly is exercised: at academic levels with long learning curves, complex programming and cumbersome extension work. The innovation GenPro brings is that it uses code-reflection for it’s inner working. This allows for absolutely easy extending: through POJO's. And the solutions it generates are in plain Java-code.
The workshop will start with a short introduction to GP followed by code examples and demos which show how to define inputs and outputs, create a testset and a fitness-function for a given problem in less than 60 lines of code. Feeding it with your own extension (e.g. function) takes an extra 5 lines of plain java code. After this introduction you have the basic knowledge to start using GenPro .
In the hands-on part of the workshop you can work on some exercises in which you will apply what you learned and start exploring the possibilities. And see the code evolve in front of your eyes in just mere minutes!
The GenPro project is searching for developers who wish to contribute to the project.