alleles

alleles is purely functional highly flexible scala library for genetic algorithms

Manifesto

This library is and always should be:

1. Easy in use

   You don't have to know how genetic algorithms work, if you know what you can use one for. alleles allows you to call GeneticAlgorithm.evolve directly with standard or custom set of operators

   • GeneticAlgorithm.evolve - for simple sequential evolution of generation for certain number of iterations
   • GeneticAlogrithm.par.evolve- for parallel sequential evolution of generation for certain number of iterations
   • GeneticAlgorithm.stream.evolve - for evolving initial population as continuous stream of evolved ones

2. Elastic for user needs

   You are no longer restricted with byte arrays or sequence of doubles. With support of type classes alleles allows you to use genetic algorithm with any data type you want: Books, Shapes, Colors as well as any already existing model from your domain without need of nasty adapters or modifications.

3. Purely functional and highly expandable

   alleles enforces immutability which increases code readability and allows efficient caching. It is also built with modular and extensible architecture, so users can easily extend any genetic algorithm or operator for particular needs.

4. Fast and optimized

   Using scala parallel collections and streaming libraries , like akka-streams it is possible to use many genetic operators in parallel and asynchronous way with no-cost transition between them.