<?php
require_once "../core/class_neuralnetwork.php";
// Create a new neural network with 3 input neurons,
// 4 hidden neurons, and 1 output neuron
$n = new NeuralNetwork(3, 4, 1);
$n->setVerbose(false);
// Add test-data to the network. In this case,
// we want the network to learn the 'XOR'-function
$n->addTestData(array(-1, -1, 1), array(-1));
$n->addTestData(array(-1, 1, 1), array(1));
$n->addTestData(array(1, -1, 1), array(1));
$n->addTestData(array(1, 1, 1), array(-1));
// we try training the network for at most $max times
$max = 3;
// train the network in max 1000 epochs, with a max squared error of 0.01
while (!($success = $n->train(1000, 0.01)) && $max-- > 0) {
echo "Nothing found...<hr />";
}
// print a message if the network was succesfully trained
if ($success) {
$epochs = $n->getEpoch();
echo "Success in {$epochs} training rounds!<hr />";
}
// in any case, we print the output of the neural network
for ($i = 0; $i < count($n->trainInputs); $i++) {
$output = $n->calculate($n->trainInputs[$i]);
print "<br />Testset {$i}; ";
print "expected output = (" . implode(", ", $n->trainOutput[$i]) . ") ";
print "output from neural network = (" . implode(", ", $output) . ")\n";
}
function test(NeuralNetwork $neuralNetwork)
{
$trainingDatas = prepareTrainingDatas();
foreach (NUMBERS as $number) {
?>
<div class="displays"><?php
?>
<div class="display"><?php
foreach ($number as $v) {
?>
<div class="display-pixel" style="opacity: <?php
echo ($v + 1) / 2;
?>
"></div><?php
}
?>
</div><?php
?>
<div class="display"><?php
foreach ($neuralNetwork->calculate($number) as $v) {
?>
<div class="display-pixel" style="opacity: <?php
echo ($v + 1) / 2;
?>
"></div><?php
}
?>
</div><?php
?>
</div><?php
}
?>
<div class="graphs">
<div class="graphNetwork" data-data="<?php
echo htmlspecialchars(json_encode($neuralNetwork->toArray()), ENT_QUOTES, 'UTF-8');
?>
"></div>
</div>
<script>draw()</script>
<?php
}
function test(NeuralNetwork $neuralNetwork)
{
?>
<div style="position: relative; width: 1000px; height: 100px; background: #eee; border-radius: 5px; margin: 20px 0px 20px;"><?php
for ($i = 0; $i < 6.5; $i += 0.1) {
$result = $neuralNetwork->calculate([$i])[0];
$left = $i * 100;
$height = $result + 1;
$height *= 50;
?>
<div style="position: absolute; left: <?php
echo $left;
?>
px; width: 10px; bottom: 0px; height: <?php
echo $height;
?>
px; background: green; opacity: 0.5;"></div><?php
}
for ($i = 0; $i < 6.5; $i += 0.1) {
$result = sin($i);
$left = $i * 100;
$height = $result + 1;
$height *= 50;
?>
<div style="position: absolute; left: <?php
echo $left;
?>
px; width: 10px; bottom: 0px; height: <?php
echo $height;
?>
px; background: red; opacity:0.5;"></div><?php
}
?>
</div><?php
}
public function __construct($opts)
{
parent::__construct($opts);
}
function test(NeuralNetwork $neuralNetwork, $withError = true)
{
$accuracy = [];
$errors = [];
foreach ($this->allTestData as $testData) {
$result = $neuralNetwork->calculate($testData->getData());
$resultBool = 0;
if ($result[0] < 0 && $result[1] > 0 && $testData->getExpectedResult() == [-1, 1]) {
$resultBool = 1;
}
if ($result[0] > 0 && $result[1] < 0 && $testData->getExpectedResult() == [1, -1]) {
$resultBool = 1;
}
$accuracy[] = $resultBool;
$errors[] = $neuralNetwork->getError($result, $testData->getExpectedResult());
}
echo '<div>Skuteczność <b>' . avg($accuracy) . '</b></div>';
echo '<div>Błąd <b>' . avg($errors) . '</b></div>';
// echo '<pre>'.print_r($neuralNetwork->toArray(), true).'</pre>';
?>
<div class="graphs">
<?php
if ($withError) {
?>
<div class="graphError" data-data="<?php
echo htmlspecialchars(json_encode($neuralNetwork->getHistoricErrors()));
?>
"></div>
<?php
}
?>
<div class="graphNetwork" data-data="<?php
echo htmlspecialchars(json_encode($neuralNetwork->toArray()), ENT_QUOTES, 'UTF-8');
?>
"></div>
</div>
<script>draw()</script>
<?php
flush();
ob_flush();
}
function test(NeuralNetwork $neuralNetwork)
{
$correctResult = [];
for ($x = MIN; $x <= MAX; $x += STEP) {
for ($y = MIN; $y <= MAX; $y += STEP) {
$result = func($x, $y);
if (is_infinite($result)) {
continue;
}
$correctResult[] = [$x, $y, func($x, $y)];
}
}
$networkResult = [];
for ($x = MIN; $x <= MAX; $x += STEP) {
for ($y = MIN; $y <= MAX; $y += STEP) {
$networkResult[] = [$x, $y, $neuralNetwork->calculate([$x, $y])[0]];
}
}
// echo '<pre>'.print_r($neuralNetwork->toArray(), true).'</pre>';
?>
<div class="graphs">
<div class="graph3d" data-data="<?php
echo json_encode($correctResult);
?>
"></div>
<div class="graph3d" data-data="<?php
echo json_encode($networkResult);
?>
"></div>
<div class="graphNetwork" data-data="<?php
echo htmlspecialchars(json_encode($neuralNetwork->toArray()), ENT_QUOTES, 'UTF-8');
?>
"></div>
</div>
<script>draw()</script>
<?php
flush();
ob_flush();
}
<style type='text/css'>
body { padding:0px; margin:0px; background: #e8e8e8; font-family: "Open Sans", Helvetica, Arial, sans; }
.container { margin: 30px auto; padding:20px; max-width: 900px; background: #f8f8f8; border:2px solid #a8a8a8; border-radius: 10px; -moz-border-radius:10px; -webkit-border-radius:10px; -o-border-radius:10px;}
.container > h1:first-child { margin-top:0px; }
.result { padding:10px; background: #f0f0f0; margin: 0 auto; max-width:650px; }
h1, h2 { padding:0px; margin:40px 0px 0px; }
p { margin: 0px 0px 20px; }
</style>
</head>
<body>
<div class='container'>
<?php
require_once "class_neuralnetwork.php";
// Create a new neural network with 3 input neurons,
// 4 hidden neurons, and 1 output neuron
$n = new NeuralNetwork(3, 4, 1);
$n->setVerbose(false);
// Add test-data to the network. In this case,
// we want the network to learn the 'XOR'-function
$n->addTestData(array(-1, -1, 1), array(-1));
$n->addTestData(array(-1, 1, 1), array(1));
$n->addTestData(array(1, -1, 1), array(1));
$n->addTestData(array(1, 1, 1), array(-1));
// we try training the network for at most $max times
$max = 3;
$i = 0;
echo "<h1>Learning the XOR function</h1>";
// train the network in max 1000 epochs, with a max squared error of 0.01
while (!($success = $n->train(1000, 0.01)) && ++$i < $max) {
echo "Round {$i}: No success...<br />";
}
