//

// This software is released under the MIT License.

// https://opensource.org/licenses/MIT

​

/* ===

ml5 Example

Image classification using MobileNet and p5.js

This example uses a callback pattern to create the classifier

=== */

​

// Initialize the Image Classifier method with MobileNet. A callback needs to be passed.

let classifier;

let chosen;

​

// A variable to hold the image we want to classify

let img;

let allImg = []

let divLabel;

let divConfidence;

​

function preload() {

  classifier = ml5.imageClassifier('MobileNet');

  // allImg.push(loadImage('images/rabbit.jpeg'));

  // allImg.push(loadImage('images/cow.jpeg'));

  // allImg.push(loadImage('images/z.jpeg'));

  // allImg.push(loadImage('images/tcn.jpeg'));

  // allImg.push(loadImage('images/ttle.jpeg'));

  // allImg.push(loadImage('images/dont_know.jpeg'));

​

  allImg.push(loadImage('puffins/p0.jpeg'));

  allImg.push(loadImage('puffins/p1.jpeg'));

  allImg.push(loadImage('puffins/p2.jpeg'));

  allImg.push(loadImage('puffins/p3.jpeg'));

  allImg.push(loadImage('puffins/p4.jpeg'));

  allImg.push(loadImage('puffins/p5.jpeg'));

  allImg.push(loadImage('puffins/p0.jpeg'));

  allImg.push(loadImage('puffins/p6.jpeg'));

  allImg.push(loadImage('puffins/p7.jpeg'));

  allImg.push(loadImage('puffins/p8.jpeg'));

  allImg.push(loadImage('puffins/p9.jpeg'));

​

  chosen = Math.floor(Math.random() * allImg.length);

}

​

function setup() {

  createCanvas(600, 400);

  classifier.classify(allImg[chosen], gotResult);

  image(allImg[chosen], 0, 0, width - 50, height - 50);

  divLabel = createDiv('Label: processing...');

  divConfidence = createDiv('Confidence: processing...');

}

​

// A function to run when we get any errors and the results

function gotResult(error, results) {

  // Display error in the console

  if (error) {

    console.log('aaa ', error);

  }

  // The results are in an array ordered by confidence.

  console.log(results)

  divLabel.html('Label: ' + results[0].label);

  divConfidence.html('Confidence: ' + results[0].confidence);

}

​

function mousePressed() {

  var randomNumber = floor(random(0, allImg.length));

  image(allImg[randomNumber], 0, 0, width - 50, height - 50);

  classifier.classify(allImg[randomNumber], gotResult);

}