function learn(NeuronSigmoidal $neuron, array $trainingDatas)
{
/* @var $trainingDatas TrainingData[] */
test($neuron);
$i = 0;
while (true) {
$errors = [];
foreach ($trainingDatas as $trainingData) {
$result = $neuron->calculate($trainingData->getData());
$error = $neuron->getError($result, $trainingData->getExpectedResult());
$errors[] = abs($error);
$neuron->correctWeights($result, $trainingData->getExpectedResult());
}
// echo avg($errors).'<br>';
if (avg($errors) < 0.05) {
break;
}
// if ($i % 10 == 0)
// test($neuron);
$i++;
if ($i == 100) {
break;
}
}
// echo '<pre>'.print_r($neuron->getWeights(), true).'</pre>';
// echo '<pre>'.print_r($neuron->getBetaWeight(), true).'</pre>';
echo avg($errors) . '<br>';
echo $i . '<br>';
}
function hideMovingObjects($imageFiles)
{
$xMax = 0;
$yMax = 0;
$images = array();
foreach ($imageFiles as $fn) {
$images[] = imagecreatefromjpeg($fn);
$info = getimagesize($fn);
$infoX = $info[0];
$infoY = $info[1];
$infoBits = $info['bits'];
$infoChannels = $info['channels'];
$xMax = max($xMax, $infoX);
$yMax = max($yMax, $infoY);
}
$colorGd = imagecreatetruecolor($xMax, $yMax);
$imagesnumber = sizeof($images);
$trimRounds = ceil(log($imagesnumber) * log($imagesnumber) / 2);
$trimStop = $imagesnumber - $trimRounds - $trimRounds;
//for all images and all pixel
//calculate the Trimmed mean of color
for ($x = 0; $x < $xMax; $x++) {
for ($y = 0; $y < $yMax; $y++) {
$colorR = array();
$colorG = array();
$colorB = array();
foreach ($images as $img) {
$colorRgb = imagecolorat($img, $x, $y);
$colorR[] = RGB::getRed($colorRgb);
$colorG[] = RGB::getGreen($colorRgb);
$colorB[] = RGB::getBlue($colorRgb);
}
sort($colorR);
sort($colorG);
sort($colorB);
$colorRed = avg(array_slice($colorR, $trimRounds, $trimStop));
$colorGreen = avg(array_slice($colorG, $trimRounds, $trimStop));
$colorBlue = avg(array_slice($colorB, $trimRounds, $trimStop));
imagesetpixel($colorGd, $x, $y, RGB::colorSet($colorRed, $colorGreen, $colorBlue));
}
}
//done return the merged image!
//imagepng($colorGd, "last.jpg" );
return $colorGd;
}
/* Upload */
$i = 0;
do {
$uploadNow = $datas['upload'][count($datas['upload']) - 1 - $i]->value;
$i++;
} while ($uploadNow == null);
$uploadNow = avg(array_reverse(array_slice($datas['upload'], -12 - $i, 12)));
$uploadLast = avg(array_reverse(array_splice($datas['upload'], 0, 12)));
/* Download */
$i = 0;
do {
$downloadNow = $datas['download'][count($datas['download']) - 1 - $i]->value;
$i++;
} while ($downloadNow == null);
$downloadNow = avg(array_reverse(array_slice($datas['download'], -12 - $i, 12)));
$downloadLast = avg(array_splice($datas['download'], 0, 12));
if ($ping && $uploadLast && $uploadNow && $downloadNow && $downloadLast) {
$insertStat->execute();
}
echo $service . " : {$ping} : {$uploadNow} : {$uploadLast} : {$downloadNow} : {$downloadLast}<br/>";
/* Détection des erreurs */
$errors = array();
/* Ping */
if ($ping > $config['pingDanger']) {
$errors[] = '<span class="text-danger">Ping ' . round($ping, 2) . '</span>';
} elseif ($ping > $config['pingWarn']) {
$errors[] = '<span class="text-warning">Ping ' . round($ping, 2) . '</span>';
}
/* Upload */
if (!$uploadNow) {
$uploadPercent = 100;
function learn(NeuralNetwork $neuralNetwork, array $trainingDatas)
{
/* @var $trainingDatas TrainingData[] */
test($neuralNetwork);
$historicErrors = [];
$i = 0;
while (true) {
$errors = [];
/*?>
<table style=""><tr>
<th>dane</th>
<th>wynik</th>
<th>spodziewany</th>
<th>błąd</th>
</tr><?php*/
foreach ($trainingDatas as $trainingData) {
/*?><tr><?php*/
$result = $neuralNetwork->calculate($trainingData->getData());
$error = $neuralNetwork->getError($result, $trainingData->getExpectedResult());
$errors[] = $error;
/*?><td><?php foreach($trainingData->getData() as $v) { echo round($v, 2).' '; } ?></td><?php
?><td><?= round($result[0], 2) ?></td><?php
?><td><?= round($trainingData->getExpectedResult()[0], 2) ?></td><?php
?><td><?= round($error, 2) ?></td><?php
?><td><?php*/
$neuralNetwork->correctWeights($result, $trainingData->getExpectedResult());
/*?></td><?php
?></tr><?php*/
}
// if ($i % 5000 == 0) {
// test($neuralNetwork);
// }
/*?></table><?php*/
// test($neuralNetwork);
// echo '<div style="width: '.(avg($errors) * 1000).'px; height: 1px; background: #666;"></div>';
// echo '<b>'.round(avg($errors), 3).'</b><br>';
if (avg($errors) < 0.01) {
return;
}
$i++;
if ($i > 1) {
$lastHistoricErrors = array_slice($historicErrors, -50);
if (avg($lastHistoricErrors) > avg($historicErrors)) {
$historicErrors = [];
$i = 0;
test($neuralNetwork);
$neuralNetwork->rerandomizeWeights();
//test($neuralNetwork);
}
}
$historicErrors[] = avg($errors);
if ($i == 100000) {
echo 'break';
break;
}
shuffle($trainingDatas);
}
}
}
fclose($file_count);
@(list($today, $yesterday, $total, $date, $days) = split("%", $data));
echo ceil($total / $days);
}
?>
Đang online: <?php
echo online();
?>
<br>
Truy cập hôm nay: <?php
echo today();
?>
<br>
Truy cập hôm qua: <?php
echo yesterday();
?>
<br>
Tổng số truy cập: <?php
total();
?>
<br>
Truy cập trung bình: <?php
avg();
?>
<br>
</td>
</tr>
</table>
$sum += $r;
}
// echo "min=".$min."*";
return $sum -= $min;
}
function sev($n)
{
$sum = 0;
$min = 20;
for ($i = 0; $i < $n; $i++) {
$t = roll(4);
if ($min > $t) {
$min = $t;
}
$sum += $t;
}
return ($sum - $min) / ($n - 1);
}
function avg($n)
{
$sum = 0;
for ($i = 0; $i < $n; $i++) {
$sum += sev(7);
}
echo $sum / $n . "<br />";
}
avg(1000);
?>
</html>
function learn(NeuralNetwork $neuralNetwork, array $trainingDatas)
{
/* @var $trainingDatas TrainingData[] */
test($neuralNetwork);
$historicErrors = [];
$i = 0;
while (true) {
$errors = [];
/*?>
<table style="left: <?= $i * 400 ?>px; top: 0px; position: absolute; display: none;"><tr>
<th>dane</th>
<th>wynik</th>
<th>spodziewany</th>
<th>błąd</th>
<th>poprawa</th>
</tr><?php*/
foreach ($trainingDatas as $trainingData) {
/*?><tr><?php*/
$result = $neuralNetwork->calculate($trainingData->getData());
$error = $neuralNetwork->getError($result, $trainingData->getExpectedResult());
$errors[] = $error;
/*?><td><?= round($trainingData->getData()[0], 2) ?></td><?php
?><td><?= round($result[0], 2) ?></td><?php
?><td><?= round($trainingData->getExpectedResult()[0], 2) ?></td><?php
?><td><?= round($error, 2) ?></td><?php*/
/*?><td><?php*/
$neuralNetwork->correctWeights($result, $trainingData->getExpectedResult());
/*?></td><?php
?></tr><?php*/
if ($i % 100 == 0) {
// test($neuralNetwork);
}
}
/*?></table><?php*/
echo '<div style="width: ' . avg($errors) * 1000 . 'px; height: 1px; background: #666;"></div>';
//echo '<b>'.round(avg($errors), 3).'</b><br>';
if (avg($errors) < 0.05) {
return;
}
$i++;
if ($i > 1) {
$lastHistoricErrors = array_slice($historicErrors, -50);
if (avg($lastHistoricErrors) > avg($historicErrors)) {
$historicErrors = [];
$i = 0;
test($neuralNetwork);
$neuralNetwork->rerandomizeWeights();
test($neuralNetwork);
}
}
$historicErrors[] = avg($errors);
if ($i == 5000) {
break;
}
shuffle($trainingDatas);
}
}
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();
}
$table["classes (ignored)"] = count($classes);
$table["Objects"] = $resourceobjects;
$table["Triples"] = $triples;
$table["Triples_with_Object_ratio"] = pretty($resourceobjects, $triples);
$table["unique_subjects"] = count($uniquesubjects);
$table["subsAreObjects"] = count($subsAreObjects);
$table["subsAreObjectsRatio"] = pretty(count($subsAreObjects), count($uniquesubjects));
$table["avg_indegree_subjects"] = avg($subsAreObjects);
$table["avg_outdegree_subjects"] = avg($uniquesubjects);
$table["unique_predicates"] = count($uniquepredicates);
$table["avg_usage_predicates"] = avg($uniquepredicates);
$table["unique_objects"] = count($uniqueobjects);
$table["objsAreSubjects"] = count($objsAreSubjects);
$table["objsAreSubjectsRatio"] = pretty(count($objsAreSubjects), count($uniqueobjects));
$table["avg_indegree_objects"] = avg($uniqueobjects);
$table["avg_outdegree_objects"] = avg($objsAreSubjects);
p($table);
if ($detailed) {
echo "classes\n";
detailed($classes);
echo "predicates\n";
detailed($uniquepredicates);
}
}
//foreach
function detailed($arr)
{
foreach ($arr as $key => $value) {
echo "{$key}\t{$value}\n";
}
echo "\n\n";
}
}
?>
<form method="post" action="#">
<p><input type="submit" name="save" value="GUARDAR CAMBIOS"></p>
<table>
<tr>
<th>Alumno</th>
<th>1ªev</th>
<th>2ªev</th>
<th>3ªev</th>
<th>Final</th>
<th>x</th>
</tr>
<?php
// var_dump($_SESSION["alumnos"]);
for ($i = 0; $i < count($_SESSION["alumnos"]); $i++) {
$_SESSION["alumnos"][$i]["nom"];
$alumno = $_SESSION["alumnos"][$i];
echo "<tr>";
echo "<td><input type='text'class='nom' name='name-" . $i . "' value='" . $alumno["nom"] . "'/></td>";
echo "<td class='nota'><input type='number' name='ev1-" . $i . "' value='" . $alumno["ev1"] . "'/></td>";
echo "<td class='nota'><input type='number' name='ev2-" . $i . "' value='" . $alumno["ev2"] . "'/></td>";
echo "<td class='nota'><input type='number' name='ev3-" . $i . "' value='" . $alumno["ev3"] . "'/></td>";
echo "<td class='nota'>" . avg($alumno) . "</td>";
echo "<td style='width:1px'><input type='submit' name='del-" . $i . "' class='submit-img img-del' value=''/></td>";
echo "</tr>";
}
?>
</table>
</form>
