ConvNetJS-minified-library
ConvNetJS minified library, the source is from :https://cs.stanford.edu/people/karpathy/convnetjs/build/convnet-min.js
Link the minified file in the index file in your text-editor to get started instanatly:
Example code to get started:
<html>
<head>
<title>Getting Started</title>
<!--CSS goes here-->
<style>
body{
background-color: red; /**Example**/
}
</style>
<!--Importing the convetjs library-->
<script src="minified.js">
</script>
<!--Javascript Content-->
<script type="text/javascript">
function periodic() {
var d = document.getElementById('eg-divison');
d.innerHTML = 'Random number: ' + Math.random()
}
var net; //global variable declared outside
function start() {
// this gets executed on startup
//...
net = new convnetjs.Net();
// ...
// example of running something every 1 second
setInterval(periodic, 1000);
}
</script>
</head>
<body onload="start()">
<div id="eg-division"></div>
</body>
</html>
Neural Net Classification
Simple two layer network binary classifier, 2 dimensional data points. First Layer of network - input layer ,declare size of input data. ConvNetJS layers - based on Vol, 3 dimensional(sx,sy,depth) volume of numbers. Next three layers - fully connected ('fc'). Last layer - Classifier layer ('softmax') , outputs probability.
In case of not using images , input volume be 1x1x2 Declaring size of input volume (out_sx=1,out_sy=1,out_depth=2)
//layer definitions
var layer_defs=[];
//Declaring size- 1x1x2 (vol class - 3D volume)
layer_defs.push({type:'input', out_sx:1, out_sy:1, out_depth:2});
//Fully Connected Layers ('fc')
layer_defs.push({type:'fc', num_neurons:20, activation:'relu'});
layer_defs.push({type:'fc', num_neurons:20, activation:'relu'});
//Softmax
layer_defs.push({type:'softmax', num_classes:2});
//Creating a net
var net = new convnetjs.Net();
net.makeLayers(layer_defs);
//Creating a volume
var x = new convnetjs.Vol([0.5, -1.3]);
var probability_volume = net.forward(x);
console.log('probability that x is class 0: ' + probability_volume.w[0]);
// prints 0.50101