|
|
<!DOCTYPE html>
<html>
<head>
<title></title>
</head>
<body>
<style type="text/css"><!--
#example-holder
{
border: 1px solid #000;
padding:5px;
background: #c0c0c0;
width: 650px;
height: 445px;
position: relative;
}
#drawing-area
{
border: 1px solid #000;
background: white;
width:650px;
height:300px;
position: absolute;
display: inline;
}
#example-btn1
{
border: 1px solid #000;
position: absolute;
display: inline;
width: 650px;
top:310px;
}
#example-btn2
{
border: 1px solid #000;
position: absolute;
display: inline;
width: 650px;
top:340px;
}
#example-btn3
{
border: 1px solid #000;
position: absolute;
display: inline;
width: 650px;
top:397px;
}
.test { color: red }
--></style><div id="example-holder">
<div id="universe-area"></div><div id="example-btn1"><input id="btnStart" type="button" value="Start" /><input id="btnStop" type="button" value="Stop" /><input id="btnSingle" type="button" value="Single" /><input id="btnRandomCities" type="button" value="Random Cities" /><input id="btnCircleCities" type="button" value="Circle Cities" /><input id="btnRandomPath" type="button" value="New Population" /></div>
<div id="example-btn2">Cities: <input id="txtCities" type="text" size="2" value="50" />, Stop after <input id="txtStable" type="text" size="2" value="50" /> stable iterations.<br />Population: <input id="txtPopulation" type="text" size="2" value="1000" />, Mutation %: <input id="txtMutationPercent" type="text" size="2" value="0.2" />, % to Mate: <input id="txtMatePercent" type="text" size="2" value="0.2" />, Eligible Pop %: <input id="txtMatingPopulationPercent" type="text" size="2" value="0.5" /><input id="btnSet" type="button" value="Set" /></div>
<div id="example-btn3">
<p id="paraStatus">Ready.</p>
</div>
</div>
<script src="../encog.js"></script>
<script src="../encog-widget.js"></script>
<script type="text/javascript">
<!--//--><![CDATA[// ><!--
"use strict";
if(window.addEventListener) {
window.addEventListener('load', function () {
var MARGIN = 10;
var universe, genetic, backgroundTimer;
var btnStart, btnStop, btnSet, btnSingle, paraStatus, btnRandomCities, btnCircleCities, btnRandomPath;
var txtCities, txtStable,txtPopulation,txtMutationPercent, txtMatePercent, txtMutationPercent;
var cities = [];
var iteration;
var constCities = 50;
var constStable = 50;
var constPopulationSize = 1000;
var lastBest;
var stableFor;
var population = [];
var constMutationPercent = 0.1;
var constMatePercent = 0.24;
var constMatingPopulationPercent = 0.5;
function init () {
// Find the canvas element.
universe = ENCOG.GUI.TSP.create('universe-area', 650, 300);
paraStatus = document.getElementById('paraStatus');
btnStart = document.getElementById('btnStart');
btnStop = document.getElementById('btnStop');
btnSet = document.getElementById('btnSet');
btnSingle = document.getElementById('btnSingle');
btnRandomCities = document.getElementById('btnRandomCities');
btnCircleCities = document.getElementById('btnCircleCities');
btnRandomPath = document.getElementById('btnRandomPath');
btnStart.addEventListener('click', ev_start, false);
btnStop.addEventListener('click', ev_stop, false);
btnSet.addEventListener('click', ev_set, false);
btnSingle.addEventListener('click', ev_single, false);
btnRandomCities.addEventListener('click', ev_randomCities, false);
btnCircleCities.addEventListener('click', ev_circleCities, false);
btnRandomPath.addEventListener('click', ev_startOver, false);
txtPopulation = document.getElementById('txtPopulation');
txtMutationPercent = document.getElementById('txtMutationPercent');
txtMatePercent = document.getElementById('txtMatePercent');
txtCities = document.getElementById('txtCities');
txtStable = document.getElementById('txtStable');
ev_set();
universe.render();
}
/////////////////////////////////////////////////////////////////////////////
// Event functions
/////////////////////////////////////////////////////////////////////////////
function ev_start(ev)
{
backgroundTimer = self.setInterval(ev_animate,100);
btnStart.disabled = true;
btnStop.disabled = false;
btnSingle.disabled = true;
btnRandomCities.disabled = true;
btnCircleCities.disabled = true;
btnRandomPath.disabled = true;
btnSet.disabled = true;
}
function ev_stop(ev)
{
self.clearInterval(backgroundTimer);
btnStart.disabled = false;
btnStop.disabled = true;
btnSingle.disabled = false;
btnRandomCities.disabled = false;
btnCircleCities.disabled = false;
btnRandomPath.disabled = false;
btnSet.disabled = false;
}
function ev_randomCities(ev)
{
universe.reset(constCities);
ev_startOver(ev);
}
function ev_circleCities(ev)
{
universe.resetCircle(constCities);
ev_startOver(ev);
}
function ev_startOver(ev)
{
iteration = 1;
genetic = ENCOG.Genetic.create();
genetic.crossover = function performCrossover(motherArray, fatherArray, child1Array, child2Array)
{
// the chromosome must be cut at two positions, determine them
var cutLength = motherArray.length / 5;
var cutpoint1 = Math.floor(Math.random() * (motherArray.length - cutLength));
var cutpoint2 = cutpoint1 + cutLength;
// keep track of which genes have been taken in each of the two
// offspring, defaults to false.
var taken1 = {};
var taken2 = {};
// handle cut section
for (var i = 0; i < motherArray.length; i++)
{
if (!((i < cutpoint1) || (i > cutpoint2)))
{
child1Array[i] = fatherArray[i];
child2Array[i] = motherArray[i];
taken1[fatherArray[i]] = true;
taken2[motherArray[i]] = true;
}
}
// handle outer sections
for (var i = 0; i < motherArray.length; i++)
{
if ((i < cutpoint1) || (i > cutpoint2))
{
child1Array[i] = getNotTaken(motherArray,taken1);
child2Array[i] = getNotTaken(fatherArray,taken2);
}
}
};
genetic.mutate = function performMutation(data)
{
var iswap1 = Math.floor(Math.random() * data.length);
var iswap2 = Math.floor(Math.random() * data.length);
// can't be equal
if (iswap1 == iswap2)
{
// move to the next, but
// don't go out of bounds
if (iswap1 > 0)
{
iswap1--;
} else {
iswap1++;
}
}
var t = data[iswap1];
data[iswap1] = data[iswap2];
data[iswap2] = t;
}
genetic.scoreSolution = function(path) {
return universe.calculatePathLength(path);
};
genetic.createPopulation(constPopulationSize, function()
{
return universe.generatePath();
});
universe.render();
};
function ev_set(ev)
{
// collect parameters
constCities = parseInt(txtCities.value);
constStable = parseInt(txtStable.value);
constPopulationSize = parseInt(txtPopulation.value);
constMutationPercent = parseFloat(txtMutationPercent.value);
constMatePercent = parseFloat(txtMatePercent.value);
ev_randomCities();
ev_startOver();
}
function ev_single(ev)
{
ev_animate();
}
function getNotTaken(source, taken)
{
for(var i=0;i<constCities;i++)
{
var trial = source[i];
if( taken[trial] != true )
{
taken[trial] = true;
return trial;
}
}
return -1;
}
function ev_animate()
{
genetic.iteration();
universe.currentPath = genetic.getSolution();
universe.render();
iteration++;
var l = Math.floor(universe.calculatePathLength(universe.currentPath));
paraStatus.innerHTML = "Iteration "+iteration+": Path length = " + l;
if (l == lastBest) {
stableFor++;
if (stableFor > constStable) {
paraStatus.innerHTML = "Stable solution found after " + iteration + " iterations: Path length = " + l;
ev_stop();
}
}
else {
lastBest = l;
stableFor = 0;
}
}
/////////////////////////////////////////////////////////////////////////////
// Downsampling functions
/////////////////////////////////////////////////////////////////////////////
/////////////////////////////////////////////////////////////////////////////
// Drawing functions
/////////////////////////////////////////////////////////////////////////////
// cause the init function to be called.
init();
}, false); }
//--><!]]>
</script>
</body>
</html>
|
By viewing downloads associated with this article you agree to the Terms of Service and the article's licence.
If a file you wish to view isn't highlighted, and is a text file (not binary), please
let us know and we'll add colourisation support for it.
Jeff Heaton, Ph.D., is a data scientist, an
adjunct instructor for the Sever Institute at Washington University, and the author of several books about artificial intelligence. Jeff holds a Master of Information Management (MIM) from Washington University and a PhD in computer science from Nova Southeastern University. Over twenty years of experience in all aspects of software development allows Jeff to bridge the gap between complex data science problems and proven software development. Working primarily with the Python, R, Java/C#, and JavaScript programming languages he leverages frameworks such as TensorFlow, Scikit-Learn, Numpy, and Theano to implement deep learning, random forests, gradient boosting machines, support vector machines, T-SNE, and generalized linear models (GLM). Jeff holds numerous certifications and credentials, such as the Johns Hopkins Data Science certification, Fellow of the Life Management Institute (FLMI), ACM Upsilon Pi Epsilon (UPE), a senior membership with IEEE. He has published his research through peer reviewed papers with the Journal of Machine Learning Research and IEEE.
Submit an article or tip