Mercurial > ift6266

<html>
<head><title>Demo</title></head>
<body>

<h2>Deep Self-Taught Learning for Handwritten Character Recognition</h2>

<p>
This demo allows you to compare two neural network detectors of handwritten characters:
<ul>
<li>A <i>shallow</i> multi-layer perceptron (MLP) with one hidden layer of 500 units;</li>
<li>A <i>deep</i> network trained with stacked denoising autoencoders (SDA) of 1000 hidden units each.</li>
</ul>
Both classifiers are trained on P07, a distribution that combines multiple datasets and stochastic transformations.
</p>

<p>
The demo provides real-time detection of characters as you draw (left panel) and apply transforms (middle panel) similar to the ones in P07. The three best results are shown (right panel) along with the confidence of the detector.
</p>

<p><b>The following applet requires <a href="http://www.adobe.com/go/getflashplayer">Adobe Flash Player 10</a>.</b></p>

<object	width="1200" height="770" codebase="http://fpdownload.macromedia.com/get/flashplayer/current/swflash.cab">
	<param name="movie" value="Test1.swf" />
	<param name="quality" value="high" />
	<param name="bgcolor" value="#869ca7" />
	<param name="allowScriptAccess" value="sameDomain" />
	<embed src="Test1.swf" quality="high" bgcolor="#869ca7"
		width="1200" height="770" name="Test1" align="middle"
		play="true"
		loop="false"
		quality="high"
		allowScriptAccess="sameDomain"
		type="application/x-shockwave-flash"
		pluginspage="http://www.adobe.com/go/getflashplayer">
	</embed>
</object>

<p><font size="-1">
Free hosting provided by <a href="http://www.000webhost.com">000webhost.com</a>
</font></p>

</body>
</html>

<!-- www.000webhost.com Analytics Code -->
<script type="text/javascript" src="http://analytics.hosting24.com/count.php"></script>
<noscript><a href="http://www.hosting24.com/"><img src="http://analytics.hosting24.com/count.php" alt="web hosting" /></a></noscript>
<!-- End Of Analytics Code -->
author	fsavard
date	Thu, 30 Sep 2010 17:54:56 -0400
parents	269c39f55134
children