darwin.js's Introduction

darwin.js

A JavaScript framework for Evolutionary Algorithms.

This framework is work in progress. The goal is to provide a general framework for the following type of evolutionary algorithms:

Genetic Algorithms
Genetic Programming
Evolution Strategies - Support for both (μ + λ) and (μ, λ) evolution strategies.

Recombination

One Point Crossover
N-Point Crossover
Uniform Crossover

Selection

Fitness Proportional Selection
Ranking Selection
Tournament Selection

Callbacks

Callback	Description
ea-started	Fires once, when the EA starts
generation-started	Fires when a new generation starts
population-generated	Fires when the population generated on every generation
population-evaluated	Fires when the population is evaluated at the beginning of a generation
generation-finished	Fires when a new generation ends
ea-finished	Fires once, when the EA ends
reset	Fires when the EA is reset

Examples

Genetic Algorithm - String Evolver

var charPool = "ABCDEFGHIJKLMNOPQRSTUVWXYZ ",
    mutationRate = 0.10,
    wordToFind = "HELLO WORLD",
    stringEvolver = new Darwin.GA({
        populationSize: 100,
        individualFactory: StringEvolver.createRandomStringGenerator(charPool, wordToFind.length),
        fitnessFunction: StringEvolver.createWordFitnessFunction(wordToFind),
        select: Darwin.Selection.randomTopPercent,
        reproduce: Darwin.Operators.singlePointCrossover,
        mutate: StringEvolver.createRandomCharacterMutation(charPool, mutationRate),
        terminationConditions: [new Darwin.Termination.TargetFitness(wordToFind.length)]
    });
    stringEvolver.run()

Evolution Strategy - String Evolver

var wordToFind = "HELLO WORLD",
    charPool = "ABCDEFGHIJKLMNOPQRSTUVWXYZ ",
    mutationRate = 0.10,
    stringEvolver = new Darwin.ES({
        parentsSize: 10,
        childrenSize: 40,
        plusSelection: true,
        individualFactory: StringEvolver.createRandomStringGenerator(charPool, wordToFind.length),
        fitnessFunction: StringEvolver.createWordFitnessFunction(wordToFind),
        mutate: StringEvolver.createRandomCharacterMutation(charPool, mutationRate),
        terminationConditions: [new Darwin.Termination.TargetFitness(wordToFind.length)]
    }),
    stringEvolver.run()