adds example of multi layered network

ml5js · Oct 2, 2019 · ab6ef26 · ab6ef26
1 parent 519cfe0
commit ab6ef26
Show file tree

Hide file tree

Showing 2 changed files with 90 additions and 0 deletions.
diff --git a/p5js/NeuralNetwork/NeuralNetwork_multiple-layers/index.html b/p5js/NeuralNetwork/NeuralNetwork_multiple-layers/index.html
@@ -0,0 +1,18 @@
+<html>
+
+<head>
+  <meta charset="UTF-8">
+  <title>Neural network with custom architecture</title>
+
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/p5.js/0.9.0/p5.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/p5.js/0.9.0/addons/p5.dom.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/p5.js/0.9.0/addons/p5.sound.min.js"></script>
+  <script src="http://localhost:8080/ml5.js" type="text/javascript"></script>
+</head>
+
+<body>
+  <h1>Neural network with custom architecture</h1>
+  <script src="sketch.js"></script>
+</body>
+
+</html>
diff --git a/p5js/NeuralNetwork/NeuralNetwork_multiple-layers/sketch.js b/p5js/NeuralNetwork/NeuralNetwork_multiple-layers/sketch.js
@@ -0,0 +1,72 @@
+let nn;
+const nn_options = {
+    inputs: 1,
+    outputs: 1,
+    layers: [
+        ml5.tf.layers.dense({
+            units: 16,
+            inputShape: [1],
+            activation: 'sigmoid',
+        }),
+        ml5.tf.layers.dense({
+            units: 16,
+            inputShape: [1],
+            activation: 'relu',
+        }),
+        ml5.tf.layers.dense({
+            units: 1,
+            activation: 'sigmoid',
+        })
+    ],
+    debug: true
+}
+
+function setup() {
+    createCanvas(400, 400);
+    background(240);
+    nn = ml5.neuralNetwork(nn_options);
+    console.log(nn);
+    createTrainingData();
+
+    nn.normalizeData();
+    const train_options = {
+        epochs: 32
+    }
+    nn.train(train_options, finishedTraining);
+}
+
+function finishedTraining(){
+
+    nn.predict([10], function(err, result){
+        if(err){
+            console.log(err);
+            return
+        }
+        console.log(result)
+    })
+
+    nn.predict([390], function(err, result){
+        if(err){
+            console.log(err);
+            return
+        }
+        console.log(result)
+    })
+}
+
+function createTrainingData(){
+    for(let i = 0; i < 400; i++){
+        if(i%2){
+            const x = floor(random(0, width/2));
+            nn.addData([x], [0])
+        }else {
+            const x = floor(random(width/2, width));
+            nn.addData([x], [1])
+        }
+    }
+
+}
+
+// function draw(){
+
+// }