start exampel without loading data

ml5js · Nov 4, 2020 · 207c5c2 · 207c5c2
1 parent 6ebea59
commit 207c5c2
Showing 1 changed file with 10 additions and 52 deletions.
diff --git a/examples/p5js/NeuralNetwork/NeuralNetwork_ImageClassifier_Video/sketch.js b/examples/p5js/NeuralNetwork/NeuralNetwork_ImageClassifier_Video/sketch.js
@@ -5,7 +5,7 @@
 
 /* ===
 ml5 Example
-Image classification using MobileNet and p5.js
+Image classification using Convolutional Neural Network
 This example uses a callback pattern to create the classifier
 === */
 let nn;
@@ -26,73 +26,34 @@ function setup() {
  video = createCapture(VIDEO);
  video.size(64, 64);
 
- trainBtn = createButton("train");
+ trainBtn = createButton('train');
  trainBtn.mousePressed(train);
- addDataBtn = createButton("addData");
+ addDataBtn = createButton('addData');
  addDataBtn.mousePressed(addData);
- resultLabel = createDiv("I see:");
-
- labelInput = createInput();
+ resultLabel = createDiv('');
+ labelInput = createInput('label');
 
  const options = {
- task: "imageClassification",
+ task: 'imageClassification',
  debug: true,
  inputs: [IMAGE_WIDTH, IMAGE_HEIGHT, IMAGE_CHANNELS],
- layers: [
- {
- type: "conv2d",
- kernelSize: 5,
- filters: 8,
- strides: 1,
- activation: "relu",
- kernelInitializer: "varianceScaling",
- },
- {
- type: "maxPooling2d",
- poolSize: [2, 2],
- strides: [2, 2],
- },
- {
- type: "conv2d",
- kernelSize: 5,
- filters: 8,
- strides: 1,
- activation: "relu",
- kernelInitializer: "varianceScaling",
- },
- {
- type: "maxPooling2d",
- poolSize: [2, 2],
- strides: [2, 2],
- },
- {
- type: "flatten",
- },
- {
- type: "dense",
- kernelInitializer: "varianceScaling",
- activation: "softmax",
- },
- ],
  };
 
  // construct the neural network
  nn = ml5.neuralNetwork(options);
- nn.loadData("daytime_nightime.json", train);
+ // nn.loadData('daytime_nightime.json', train);
 }
 
 function draw() {
  image(video, 0, 0, width, height);
 }
 
 function addData() {
- console.log("adding data", labelInput.value());
+ console.log('adding data', labelInput.value());
  nn.addData({ image: video }, { label: labelInput.value() });
 }
 
 function train() {
- // nn.normalizeData();
-
  const TRAINING_OPTIONS = {
  batchSize: 16,
  epochs: 4,
@@ -103,7 +64,7 @@ function train() {
 }
 
 function finishedTraining() {
- console.log("finished training");
+ console.log('finished training');
  nn.classify([video], gotResults);
 }
 
@@ -112,9 +73,6 @@ function gotResults(err, result) {
  console.log(err);
  return;
  }
-
- // image(video, 0,0, width, height)
- resultLabel.elt.textContent = `I see ${result[0].label}`;
-
+ resultLabel.html(`${result[0].label}`);
  nn.classify([video], gotResults);
 }