let nn;

let training = true;

let classifyX;

let classifyY;

let output = "";

​

​

function setup() {

  createCanvas(400, 400);

​

  let options = {

    // object literal

    task: "regression",

    debug: true,

  };

​

  nn = ml5.neuralNetwork(options);

}

​

function draw() {

  background(255);

  textSize(48);

  text(currentNumbers[0], 50, height/2);

  text(currentNumbers[1], width-50, height/2);

  text(output, width/2, height-100);

}

​

function mousePressed() {

  if (training) {

    let inputs = [currentNumbers[0], currentNumbers[1]];

    let average = (currentNumbers[0] + currentNumbers[1]) /2;

    let outputs = [ average ];

    console.log("Added " + currentNumbers[0] + " AVG " + currentNumbers[1] + " = " + average);

    nn.addData(inputs, outputs);

  } else {

    let inputs = [currentNumbers[0], currentNumbers[1]];

    nn.predict(inputs, donePredicting);

  }

}

​

function keyPressed() {

  if (key >= '0' && key <= '9') {

    currentNumbers.push(parseInt(key));

    if(currentNumbers.length > 2) {

      currentNumbers.shift(); // keep only last two numbers

    }

  } else if(key == "s" || key == "S") {

    nn.saveData();

  } else if(key == "1" || key == "L") {

    nn.loadData("data.json");

  } else if (key == "t") {

    nn.normalizeData();

    let options = {

      epochs: 64,

      batchSize: 12,

    } ;

    nn.train(options, doneTraining);

  }

}

​

function doneTraining() {

  console.log("Done training!");

  training = false;

}

​

function donePredicting(error, results) {

  console.log(results);

  output = results[0].value;

}

​