dolphin278/genetic

Implementation of genetic algorithms for nodejs

genetic's Issues

Traveling Salesmen Problem Solver

May you guide how to use your library to solve the tsp ? Thanks a lot

Cannot read property 'score' of undefined

Someone got this problem before? I cannot run more than 100 generations.

/Users/michaelmalura/Development/bot/node_modules/genetic/lib/genetic/Task.js:210
            level += self.parents[position].score
                                           ^

TypeError: Cannot read property 'score' of undefined
    at /Users/michaelmalura/Development/bot/node_modules/genetic/lib/genetic/Task.js:210:44
    at Object.async.until (/Users/michaelmalura/Development/bot/node_modules/genetic/node_modules/async/lib/async.js:568:13)
    at /Users/michaelmalura/Development/bot/node_modules/genetic/lib/genetic/Task.js:198:13
    at /Users/michaelmalura/Development/bot/node_modules/genetic/node_modules/async/lib/async.js:486:21
    at /Users/michaelmalura/Development/bot/node_modules/genetic/node_modules/async/lib/async.js:185:13
    at iterate (/Users/michaelmalura/Development/bot/node_modules/genetic/node_modules/async/lib/async.js:108:13)
    at /Users/michaelmalura/Development/bot/node_modules/genetic/node_modules/async/lib/async.js:119:25
    at /Users/michaelmalura/Development/bot/node_modules/genetic/node_modules/async/lib/async.js:187:17
    at /Users/michaelmalura/Development/bot/node_modules/genetic/node_modules/async/lib/async.js:491:34
    at /Users/michaelmalura/Development/bot/node_modules/genetic/lib/genetic/Task.js:192:7

function random(start, end) {
    return Math.floor(Math.random() * end) + start;
};

function getRandomSolution(callback) {
    // let threshhold = 0.03;
    // let buyCut = 3;
    // let sellCut = 8;

    callback({
        threshhold: random(0.02, 0.3), // 2% - 30%
        buyCut: random(1, 100), // 1/1 - 1/100
        sellCut: random(1, 100) // 1/1 - 1/100
    })
}

function fitness(solution, callback) {
    const { profit, sells, buys } = simulate(stockData, Object.assign({}, solution));
    callback(profit)
}

function mutate(solution, callback) {
    let s = Object.assign({}, solution);
    if (Math.random() < 0.3) {
        s.threshhold = random(0.02, 0.3);
    }
    if (Math.random() < 0.3) {
        s.buyCut = random(1, 100);
    }
    if (Math.random() < 0.3) {
        s.sellCut = random(1, 100);
    }
    callback(s)
}

function crossover(father, mother, callback) {
    var child = {};

    assert(!!father, `No father`);
    assert(!!mother, `No mother`);

    if (Math.random() >= 0.5) {
        child.threshhold = father.threshhold;
    } else {
        child.threshhold = mother.threshhold;
    }

    if (Math.random() >= 0.5) {
        child.sellCut = father.sellCut;
    } else {
        child.sellCut = mother.sellCut;
    }

    if (Math.random() >= 0.5) {
        child.buyCut = father.buyCut;
    } else {
        child.buyCut = mother.buyCut;
    }
    callback(child);
}

function stopCriteria() {
    return (this.generation === 300)
}

const geneticOptions = {
    getRandomSolution: getRandomSolution,  // previously described to produce random solution
    popSize: 100,  // population size
    stopCriteria: stopCriteria,  // previously described to act as stopping criteria for entire process
    fitness: fitness,  // previously described to measure how good your solution is
    minimize: false,  // whether you want to minimize fitness function. default is `false`, so you can omit it
    mutateProbability: 0.1,  // mutation chance per single child generation
    mutate: mutate,  // previously described to implement mutation
    crossoverProbability: 0.3, // crossover chance per single child generation
    crossover: crossover // previously described to produce child solution by combining two parents
}

const runEvolution = function () {
    const taskInstance = new Task(geneticOptions);
    taskInstance.on('error', function (error) { console.log('ERROR - ', error) })
    taskInstance.run(function (stats) {
        console.log('results', stats);
    });
}