libGeneiAL

GeneiAL is an open source extensible genetic algorithm library. It provides a framework to solve optimization problems by maximizing fitness value of the chromosomes in a population over the course of generations using a custom fitness function. The library offers the user high flexibility and extensibility while simultaneously featuring good scalability for the core functionality.

Installation and Documentation

Please, follow the Readme.md to install the library on your system and run the demos. A doxygen-generated documentation will be available soon.

Implemented Features

• support for user-defined chromosome types (e.g., the optimization of sequences of genes)
• generation of new chromosomes using random values within a user-specified interval
• generation of new chromosomes by partially/fully reusing population from a previous optimization
• multi-threading support for the evaluation of the fitness function
• diagnostic tools to analyze the genetic algorithm's behavior (i.e., runtime measurement, analysis of genetic operators, etc.)
• extendible by user-specified analysis tools with customized pre- and post-processing logic

Genetic Operators

• roulette-wheel selection
• N-point crossover, smoothed variant of N-point crossover
• uniform mutation, smoothed variant of uniform mutation
• extendible by customized genetic operators

Stopping Criteria

• basic stopping criteria (fixed-point-criterion,maximal number of generations, etc.)
• extendible by user-defined stopping criteria
• support for a combinations (AND, OR, XOR, Negation) of multiple stopping criteria