dorr91 / genetic Goto Github PK

genetic is a library for creating, applying, and tuning genetic algorithms. It consists of a package, genetics, containing a few classes, along with an example in genetics.example

Code structure: package genetics class Species: abstract class; subclasses represent a species and instances of the subclass are members of the species contains the fitness function as an instance method score() contains the instance method mate(Individual other) for sexual reproduction class Population: represents a group of members of a species keeps track of each generation determines which individuals should mate (and calls their mate function) class Tuner: Tests different combinations of parameters in order to improve genetic algorithm performance (Should use a genetic algorithm? I like it...)

genetics.example class StringSpecies represents a string class StringRunner: creates a population of strings, specifies a target string, and evolves the population until the target string is generated.

Running the example: java genetics.example.StringRunner

Usage: create a subclass of Species with the signature Subclass extends Species override the methods score() and mate() create a new Population, passing in the initial generation and the number from each generation that should survive call generate(), check the best() individual encountered so far rinse and repeat