Pārlūkot izejas kodu

Added comments to GA algo

pull/18/head
jrtechs pirms 5 gadiem
vecāks
revīzija
b92ba144f3
1 mainītis faili ar 121 papildinājumiem un 29 dzēšanām
  1. +121
    -29
      geneticAlgorithm/geneticAlgo.html

+ 121
- 29
geneticAlgorithm/geneticAlgo.html Parādīt failu

@ -12,6 +12,12 @@
<script>
class Gene
{
/**
* Constructs a new Gene to store in a chromosome.
* @param min minimum value that this gene can store
* @param max value this gene can possibly be
* @param value normalized value
*/
constructor(min, max, value)
{
this.min = min;
@ -19,6 +25,10 @@
this.value = value;
}
/**
* De-normalizes the value of the gene
* @returns {*}
*/
getRealValue()
{
return (this.max - this.min) * this.value + this.min;
@ -48,6 +58,11 @@
class Chromosome
{
/**
* Constructs a chromosome by making a copy of
* a list of genes.
* @param geneArray
*/
constructor(geneArray)
{
this.genes = [];
@ -62,11 +77,20 @@
return this.genes;
}
/**
* Mutates a random gene.
*/
mutate()
{
this.genes[Math.round(Math.random() * (this.genes.length-1))].setValue(Math.random());
}
/**
* Creates a totally new chromosome with same
* genetic structure as this chromosome but different
* values.
* @returns {Chromosome}
*/
createRandomChromosome()
{
let geneAr = [];
@ -79,6 +103,13 @@
}
/**
* Mates two chromosomes using the blending method
* and returns a list of 2 offspring.
* @param father
* @param mother
* @returns {Chromosome[]}
*/
const breed = function(father, mother)
{
let son = new Chromosome(father.getGenes());
@ -89,10 +120,31 @@
let blendCoef = Math.random();
blendGene(son.getGenes()[i], daughter.getGenes()[i], blendCoef);
}
return [son, daughter];
};
/**
* Blends two genes together based on a random blend
* coefficient.
**/
const blendGene = function(gene1, gene2, blendCoef)
{
let value1 = (blendCoef * gene1.getValue()) +
(gene2.getValue() * (1- blendCoef));
let value2 = ((1-blendCoef) * gene1.getValue()) +
(gene2.getValue() * blendCoef);
gene1.setValue(value1);
gene2.setValue(value2);
};
/**
* Helper function to sort an array
*
* @param prop name of JSON property to sort by
* @returns {function(*, *): number}
*/
function predicateBy(prop)
{
return function(a,b)
@ -110,7 +162,17 @@
}
}
/**
* Function which computes the fitness of everyone in the
* population and returns the most fit survivors. Method
* known as elitism.
*
* @param population
* @param keepNumber
* @param fitnessFunction
* @returns {{average: number,
* survivors: Array, bestFit: Chromosome }}
*/
const naturalSelection = function(population, keepNumber, fitnessFunction)
{
let fitnessArray = [];
@ -135,22 +197,17 @@
return {average: total/population.length, survivors: survivors, bestFit: bestFitness, bestChrom: bestChromosome};
};
const blendGene = function(gene1, gene2, blendCoef)
{
let value1 = (blendCoef * gene1.getValue()) +
(gene2.getValue() * (1- blendCoef));
let value2 = ((1-blendCoef) * gene1.getValue()) +
(gene2.getValue() * blendCoef);
gene1.setValue(value1);
gene2.setValue(value2);
};
/**
* Randomly everyone in the population
*
* @param population
* @param desiredPopulationSize
*/
const matePopulation = function(population, desiredPopulationSize)
{
let pairsNeeded = (desiredPopulationSize - population.length)/2;
const originalLength = population.length;
for(let i = 0; i < pairsNeeded; i++)
while(population.length < desiredPopulationSize)
{
let index1 = Math.round(Math.random() * (originalLength-1));
let index2 = Math.round(Math.random() * (originalLength-1));
@ -163,7 +220,9 @@
}
};
/**
* Randomly mutates the population
**/
const mutatePopulation = function(population, mutatePercentage)
{
if(population.length >= 2)
@ -182,6 +241,11 @@
}
};
/**
* Introduces x random chromosomes to the population.
* @param population
* @param immigrationSize
*/
const newBlood = function(population, immigrationSize)
{
for(let i = 0; i < immigrationSize; i++)
@ -192,15 +256,28 @@
};
let costx = Math.random() * 10;
let costy = Math.random() * 10;
/** Defines the cost as the "distance" to a 2-d point.
* @param chromosome
* @returns {number}
*/
const basicCostFunction = function(chromosome)
{
console.log(chromosome);
console.log((chromosome.getGenes()[0].getRealValue()));
return Math.abs(chromosome.getGenes()[0].getRealValue() - 6) +
Math.abs(chromosome.getGenes()[1].getRealValue() - 2);
return Math.abs(chromosome.getGenes()[0].getRealValue() - costx) +
Math.abs(chromosome.getGenes()[1].getRealValue() - costy);
};
/**
* Creates a totally random population based on a desired size
* and a prototypical chromosome.
*
* @param geneticChromosome
* @param populationSize
* @returns {Array}
*/
const createRandomPopulation = function(geneticChromosome, populationSize)
{
let population = [];
@ -212,18 +289,34 @@
};
/**
* Runs the genetic algorithm by going through the processes of
* natural selection, mutation, mating, and immigrations. This
* process will continue until an adequately performing chromosome
* is found or a generation threshold is passed.
*
* @param geneticChromosome Prototypical chromosome: used so algo knows
* what the dna of the population looks like.
* @param costFunction Function which defines how bad a Chromosome is
* @param populationSize Desired population size for population
* @param maxGenerations Cut off level for number of generations to run
* @param desiredCost Sufficient cost to terminate program at
* @param mutationRate Number between [0,1] representing proportion of genes
* to mutate each generation
* @param keepNumber Number of Organisms which survive each generation
* @param newBloodNumber Number of random immigrants to introduce into
* the population each generation.
* @returns {*}
*/
const runGeneticOptimization = function(geneticChromosome, costFunction,
populationSize, maxGenerations,
desiredCost, mutationRate, keepNumber,
newBloodNumber)
{
let population = createRandomPopulation(geneticChromosome, populationSize);
let generation = 0;
let bestCost = Number.MAX_VALUE;
let bestChromosome = geneticChromosome;
do
{
matePopulation(population, populationSize);
@ -240,21 +333,20 @@
generation++;
console.log("Generation " + generation + " Best Cost: " + bestCost);
console.log(generationResult);
}while(generation < maxGenerations && bestCost > desiredCost);
return bestChromosome;
};
/**
* Ugly globals used to keep track of population state for the graph.
*/
let genericChromosomeG, costFunctionG,
populationSizeG, maxGenerationsG,
desiredCostG, mutationRateG, keepNumberG,
newBloodNumberG, populationG, generationG,
bestCostG = Number.MAX_VALUE, bestChromosomeG = genericChromosomeG;
const runGeneticOptimizationforGraph = function()
const runGeneticOptimizationForGraph = function()
{
let generationResult = naturalSelection(populationG, keepNumberG, costFunctionG);
@ -416,7 +508,7 @@
{
if(autoRunning === true)
{
runGeneticOptimizationforGraph();
runGeneticOptimizationForGraph();
setTimeout(runAutoOptimizer, 1000);
}
}
@ -449,7 +541,7 @@
<div id="line_chart"></div>
<input class='btn btn-primary' id="runOptimizer" onclick='runGeneticOptimizationforGraph()' type="button" value="Next Generation">
<input class='btn btn-primary' id="runOptimizer" onclick='runGeneticOptimizationForGraph()' type="button" value="Next Generation">
<input class='btn btn-primary' id="runAutoOptimizer" onclick='startStopAutoRun()' type="button" value="Auto Run">
<div class="card">

Notiek ielāde…
Atcelt
Saglabāt