galgo

Library to calculate minimal and maximum solutions using Genetic Algorithm.

Current specs:

• Each solution for a population generates 2 children solutions
• Only best 50% of generate children survive to generate next solutions
• No parents surviving

Using

var galgo = require('galgo');
galgo.fitnessFunction = myFitnessFunction;
galgo.options = myOptions;    // optionally set 1 or more options (or replace by its own)
var result = galgo.run();

Options

Options with default values, that can be changed through galgo.options:

options: {
    chromosomeLength: 10,       // length of encoding array of bits
    generationsQty: 5,          // quantity of generations to run (#iterations)
    mutationProbability: 0.02,  // probability of mutation occur on next generation of a solution: [0, 1]
    populationSize: 10000       // size of solutions per generation,
    interval: {                 // interval of accepted solution
        min: -2,
        max: 2
    }
}

Fitness Function

On this initial version, galgo expects fitness function, or fitness function, only with 1 or 2 variables.

It's defined by galgo.fitnessFunction:

var galgo = require('galgo');
galgo.fitnessFunction = function myFitnessFn(x, y) {
    return x * x + 4 * y * y + 4 * y + x;
}

TODO

• #1: Allow n variables in fitness function
• #2: Choose to min or max the fitness function
• #3: Allow Elitism
• #4: Surviving parents with max-age option