Code viewer for World: CharRecognition_UsingCNN (...

// Cloned by Dheera on 16 Nov 2021 from World "Character recognition neural network" by "Coding Train" project
// Please leave this clone trail here.


// Ported the CNN implementation in javascript from:
//https://github.com/karpathy/convnetjs
//Ported the EMNIST data from here:
//https://github.com/acl21/Alphabet_Recognition_Gestures/tree/master/data
// with many modifications


// --- defined by MNIST - do not change these ---------------------------------------

const PIXELS        = 28;                       // images in data set are tiny
const PIXELSSQUARED = PIXELS * PIXELS;

// number of training and test exemplars in the data set:
const NOTRAIN = 124800;
const NOTEST  = 20800;



//--- can modify all these --------------------------------------------------

// no of nodes in network
const noinput  = PIXELSSQUARED;
const nohidden = 64;
const nooutput = 10;

const learningrate = 0.1;   // default 0.1

// should we train every timestep or not
let do_training = true;

// how many to train and test per timestep
const TRAINPERSTEP = 15;
const TESTPERSTEP  = 5;

// multiply it by this to magnify for display
const ZOOMFACTOR    = 7;
const ZOOMPIXELS    = ZOOMFACTOR * PIXELS;

// 3 rows of
// large image + 50 gap + small image
// 50 gap between rows

const canvaswidth = ( PIXELS + ZOOMPIXELS ) + 50;
const canvasheight = ( ZOOMPIXELS * 3 ) + 100;


const DOODLE_THICK = 18;    // thickness of doodle lines
const DOODLE_BLUR = 3;      // blur factor applied to doodles


let mnist;
// all data is loaded into this
// mnist.train_images
// mnist.train_labels
// mnist.test_images
// mnist.test_labels

//Dheera says - my global variables mycnn, mycnnTrain, mycnnModel and alphabets defined below:
let mycnn;
let mycnnTrain;
let mycnnModel;
// Dheera says - alphabets array defined for the possible outputs in charater recognition.
let alphabets = ["A", "B", "C", "D", "E", "F", "G", "H", "I", "J", "K", "L", "M", "N", "O", "P", "Q", "R", "S", "T", "U", "V", "W", "X", "Y", "Z"];

let trainrun = 1;
let train_index = 0;
let testrun = 1;
let test_index = 0;
let total_tests = 0;
let total_correct = 0;

// images in LHS:
let doodle, demo;
let doodle_exists = false;
let demo_exists = false;

let mousedrag = false;      // are we in the middle of a mouse drag drawing?


// save inputs to global var to inspect
// type these names in console
var train_inputs, test_inputs, demo_inputs, doodle_inputs;


// Matrix.randomize() is changed to point to this. Must be defined by user of Matrix.

function randomWeight()
{
    return ( AB.randomFloatAtoB ( -0.5, 0.5 ) );
            // Coding Train default is -1 to 1
}


// make run header bigger
AB.headerCSS ( { "max-height": "95vh" } );


//--- start of AB.msgs structure: ---------------------------------------------------------
// We output a serious of AB.msgs to put data at various places in the run header
var thehtml;

  // 1 Doodle header
  //Dheera says - Renamed the button to clear and redraw the doodle.
  thehtml = "<hr> <h1> 1. Doodle </h1> Top row: Doodle (left) and shrunk (right). <br> " +
        " Draw your doodle in top LHS. <button onclick='wipeDoodle();' class='normbutton' >Clear and Re-draw</button> <br> " ;
   AB.msg ( thehtml, 1 );

  // 2 Doodle variable data (guess)

  // 3 Training header
  // Dheera says - Added buttons to pause and re-start training.
  thehtml = "<hr> <h1> 2. Training </h1> Middle row: Training image magnified (left) and original (right). <br>  " +
        " <button onclick='do_training = false;' class='normbutton' >Pause training</button> " +
        " <button onclick='do_training = true;' class='normbutton' >Resume training</button> <br> ";
  AB.msg ( thehtml, 3 );

  // 4 variable training data

  // 5 Testing header
  thehtml = "<h3> Hidden tests </h3> " ;
  AB.msg ( thehtml, 5 );

  // 6 variable testing data

  // 7 Demo header
  thehtml = "<hr> <h1> 3. Demo </h1> Bottom row: Test image magnified (left) and  original (right). <br>" +
        " The network is <i>not</i> trained on any of these images. <br> " +
        " <button onclick='makeDemo();' class='normbutton' >Demo test image</button> <br> ";
   AB.msg ( thehtml, 7 );

  // 8 Demo variable data (random demo ID)
  // 9 Demo variable data (changing guess)

const bluespan = "<span style='font-weight:bold; font-size:large; color:darkblue'> "  ;


//--- end of AB.msgs structure: ---------------------------------------------------------




function setup()
{
  createCanvas ( canvaswidth, canvasheight );

  doodle = createGraphics ( ZOOMPIXELS, ZOOMPIXELS );       // doodle on larger canvas
  doodle.pixelDensity(1);
  //Dheera says - wiped the screen before drawing the doodle so that it drwas on black screen.
  wipeDoodle();


// JS load other JS
// maybe have a loading screen while loading the JS and the data set

      AB.loadingScreen();

 $.getScript ( "/uploads/codingtrain/matrix.js", function()
 {
   //Dheera says - uploaded the plug-n-play library of convnetJS here imported from https://github.com/karpathy/convnetjs
   $.getScript ( "/uploads/dheera0704/convnet3.js", function() 
   {
        //Dheera says - fetching the EMNIST files from yhe links specified in this file.
        $.getScript ( "/uploads/dheera0704/mnistDhe.js", function()
        {
            console.log ("All JS Files loaded");
          //Dheera says - Defining the CNN layers below
          let layer_defs = [];
          //Dheera says - This is the first input layer. The input is a 28x28 Character image in 1 dimention.
          layer_defs.push({
          type : "input",
          out_sx : 28,
          out_sy : 28,
          out_depth : 1
        });
        //Dheera says - 1st convulution layer with 8 filters of filter size 5x5, with stride =1 and padding = 2 and "relu" activation.
        layer_defs.push({
          type : "conv",
          sx : 5,
          filters : 8,
          stride : 1,
          pad : 2,
          activation : "relu"
        });
        //Dheera says - 1st pooling layer of the network of size 2x2
        layer_defs.push({
          type : "pool",
          sx : 2,
          stride : 2
        });
        //Dheera says - 2nd convolution layer with with 16 filters and size 5x5, with stride = 1 and padding = 2 and "relu" activation.
        layer_defs.push({
          type : "conv",
          sx : 5,
          filters : 16,
          stride : 1,
          pad : 2,
          activation : "relu"
        });
        //Dheera says - 2nd pooling layer of size 3x3
        layer_defs.push({
          type : "pool",
          sx : 3,
          stride : 3
        });
        //Dheera says - last layer which is also called the loss layer. 
        //Here softmax activation has been chosen since there are more than 2 labels in output.
        layer_defs.push({
          type : "softmax",
          num_classes : 26
        });
        //Dheera says - creating an empty network called mycnnModel here
        mycnnModel = new convnetjs.Net();
        //Dheera says - Defining the layers of my model according to the layer_defs defined above.
        mycnnModel.makeLayers(layer_defs);
        //Dheera says - Taken the SGD trainer model  here with momentum 0.9 and batch_size of 10. L2_decay has been kept small because the size of training set is not very small.
        mycnnTrain = new convnetjs.SGDTrainer(mycnnModel, {method : "adadelta", momentum: 0.9, batch_size : 10, l2_decay : .001});
            loadData();
        });
   });
 });
}



// load data set from local file (on this server)

function loadData()
{
  loadMNIST (function(data){

    mnist = data;
    //Dheera says - Rotating the training and testing dataset images as they are sizeways.
    let loc = 0;
    for (; loc < NOTRAIN; loc++) {
      
      rotateImage(mnist.train_images[loc]);
    }
    
    for (loc = 0; loc < NOTEST; loc++) {
      rotateImage(mnist.test_images[loc]);
    }
    console.log ("All data loaded into Emnist object.")
    console.log(mnist);
    AB.removeLoading();     // if no loading screen exists, this does nothing
  });
}


function getImage ( img )      // make a P5 image object from a raw data array
{
    let theimage  = createImage (PIXELS, PIXELS);    // make blank image, then populate it
    theimage.loadPixels();

    for (let i = 0; i < PIXELSSQUARED ; i++)
    {
        let bright = img[i];
        let index = i * 4;
        theimage.pixels[index + 0] = bright;
        theimage.pixels[index + 1] = bright;
        theimage.pixels[index + 2] = bright;
        theimage.pixels[index + 3] = 255;
    }
    theimage.updatePixels();
    return theimage;
}


function getInputs ( img )      // convert img array into normalised input array
{
    let inputs = [];
    for (let i = 0; i < PIXELSSQUARED ; i++)
    {
        let bright = img[i];
        inputs[i] = bright / 255;       // normalise to 0 to 1
    }
    return ( inputs );
}

//Dheera says - new function defined to rotate the images
function rotateImage(img) {
  for (let e = 0; e < PIXELS; e++) {
    for (let f = e; f < PIXELS; f++) {
      let a = e * PIXELS + f;
      let b = f * PIXELS + e;
      let c = img[a];
      img[a] = img[b];
      img[b] = c;
    }
  }
} 


function trainit (show)        // train the network with a single exemplar, from global var "train_index", show visual on or off
{
  let img   = mnist.train_images[train_index];
  let label = mnist.train_labels[train_index];

  // optional - show visual of the image
  if (show)
  {
    var theimage = getImage ( img );    // get image from data array
    image ( theimage,   0,                ZOOMPIXELS+50,    ZOOMPIXELS,     ZOOMPIXELS  );      // magnified
    image ( theimage,   ZOOMPIXELS+50,    ZOOMPIXELS+50,    PIXELS,         PIXELS      );      // original
  }

  // set up the inputs
  let inputs = getInputs ( img );       // get inputs from data array
  train_inputs = inputs;

  {
    //Dheera says - calling function getmycnnInputs below to create a volume of inputs activations required for the trainer.
    let set = getmycnnInputs(inputs);
    //Dheera says - trainer is passed the volume of exemplars from training images  plus their labels.
    mycnnTrain.train(set, label);
  }

         // can inspect in console

  thehtml = " trainrun: " + trainrun + "<br> no: " + train_index ;
  AB.msg ( thehtml, 4 );

  train_index++;
  if ( train_index == NOTRAIN )
  {
    train_index = 0;
    console.log( "finished trainrun: " + trainrun );
    trainrun++;
  }
}
 //Dheera says - New function defined to create a volume of input activations of size 28x28X1 initialized by zeros.
function getmycnnInputs(t) {
  var x = new convnetjs.Vol(28, 28, 1, 0);

  for (var j = 0; j < PIXELSSQUARED; j++) {
    x.w[j] = t[j];
  }

  return x;
}

function testit()    // test the network with a single exemplar, from global var "test_index"
{
  let id   = mnist.test_images[test_index];
  let label = mnist.test_labels[test_index];

  // set up the inputs
  let inputs = getInputs ( id );
  //Dheera says - getting inputs of test image and finding the value with the maximum weight for prediction.
  let myinput = getmycnnInputs(inputs);
  test_inputs = inputs;        // can inspect in console
  let place = findMax(mycnnModel.forward(myinput).w);
  var img = getImage(id);
  image(img, 0, ZOOMPIXELS + 50, ZOOMPIXELS, ZOOMPIXELS);
  image(img, ZOOMPIXELS + 50, ZOOMPIXELS + 50, PIXELS, PIXELS);

  total_tests++;
  if (place == label)  total_correct++;

  let percent = (total_correct / total_tests) * 100 ;

  thehtml =  " testrun: " + testrun + "<br> no: " + total_tests + " <br> " +
        " correct: " + total_correct + "<br>" +
        "  score: " + bluespan + percent.toFixed(2) + "</span>";
  AB.msg ( thehtml, 6 );

  test_index++;
  if ( test_index == NOTEST )
  {
    console.log( "finished testrun: " + testrun + " score: " + percent.toFixed(2) );
    testrun++;
    test_index = 0;
    total_tests = 0;
    total_correct = 0;
  }
}




//--- find no.1 (and maybe no.2) output nodes ---------------------------------------
// (restriction) assumes array values start at 0 (which is true for output nodes)


function find12 (a)         // return array showing indexes of no.1 and no.2 values in array
{

  let no1 = 0;
  let no2 = 0;
  let no1value = 0;
  let no2value = 0;

  for (let i = 0; i < a.length; i++)
  {
    if (a[i] > no1value)   // new no1
    {
      // old no1 becomes no2
      no2 = no1;
      no2value = no1value;
      // now put in the new no1
      no1 = i;
      no1value = a[i];
    }
    else if (a[i] > no2value)  // new no2
    {
      no2 = i;
      no2value = a[i];
    }
  }

  var b = [ no1, no2 ];
  return b;
}



// just get the maximum - separate function for speed - done many times
// find our guess - the max of the output nodes array

function findMax (a)
{
  let no1 = 0;
  let no1value = 0;

  for (let i = 0; i < a.length; i++)
  {
    if (a[i] > no1value)
    {
      no1 = i;
      no1value = a[i];
    }
  }

  return no1;
}




// --- the draw function -------------------------------------------------------------
// every step:

function draw()
{
  // check if libraries and data loaded yet:
  if ( typeof mnist == 'undefined' ) return;

    background ('black');
    strokeWeight(1);
    stroke('green');
    rect(0,0,ZOOMPIXELS,ZOOMPIXELS); //Dheera says - draws a rectangle with source 0,0 and length and bredth defined by Zoompixels variable.
    textSize(10);
    textAlign(CENTER);
    text("DOODLE HERE",ZOOMPIXELS/2,ZOOMPIXELS/2);


if ( do_training )
{

  // do some training per step
    for (let i = 0; i < TRAINPERSTEP; i++)
    {
      if (i === 0)    trainit(true);    // show only one per step - still flashes by
      else           trainit(false);
    }

  // do some testing per step
    for (let i = 0; i < TESTPERSTEP; i++)
      testit();
}

  // keep drawing demo and doodle images
  // and keep guessing - we will update our guess as time goes on

  if ( demo_exists )
  {
    drawDemo();
    guessDemo();
  }
  if ( doodle_exists )
  {
    drawDoodle();
    guessDoodle();
  }


// detect doodle drawing
// (restriction) the following assumes doodle starts at 0,0

  if ( mouseIsPressed )         // gets called when we click buttons, as well as if in doodle corner
  {
     // console.log ( mouseX + " " + mouseY + " " + pmouseX + " " + pmouseY );
     var MAX = ZOOMPIXELS + 20;     // can draw up to this pixels in corner
     if ( (mouseX < MAX) && (mouseY < MAX) && (pmouseX < MAX) && (pmouseY < MAX) )
     {
        mousedrag = true;       // start a mouse drag
        doodle_exists = true;
        doodle.stroke('red');
        strokeJoin(ROUND);
        doodle.strokeWeight( DOODLE_THICK );
        doodle.line(mouseX, mouseY, pmouseX, pmouseY);
     }
  }
  else
  {
      // are we exiting a drawing
      if ( mousedrag )
      {
            mousedrag = false;
            // console.log ("Exiting draw. Now blurring.");
            doodle.filter (BLUR, DOODLE_BLUR);    // just blur once
            //   console.log (doodle);
      }
  }
}




//--- demo -------------------------------------------------------------
// demo some test image and predict it
// get it from test set so have not used it in training


function makeDemo()
{
    demo_exists = true;
    var  i = AB.randomIntAtoB ( 0, NOTEST - 1 );

    demo        = mnist.test_images[i];
    var label   = mnist.test_labels[i];

   thehtml =  "Test image no: " + i + "<br>" +
            "Classification: " + alphabets[label - 1] + "<br>" ;
   AB.msg ( thehtml, 8 );

   // type "demo" in console to see raw data
}


function drawDemo()
{
    var theimage = getImage ( demo );
     //  console.log (theimage);

    image ( theimage,   0,                canvasheight - ZOOMPIXELS,    ZOOMPIXELS,     ZOOMPIXELS  );      // magnified
    image ( theimage,   ZOOMPIXELS+50,    canvasheight - ZOOMPIXELS,    PIXELS,         PIXELS      );      // original
}


function guessDemo()
{
   let inputs = getInputs ( demo );

  demo_inputs = inputs;  // can inspect in console
//Dheera says - fetching the imput and finding the one with mx weight for demo prediction.
let myinput = getmycnnInputs(inputs);
let j = findMax(mycnnModel.forward(myinput).w);

   thehtml =   " We classify it as: " + bluespan + alphabets[j-1] + "</span>" ;
   AB.msg ( thehtml, 9 );
}




//--- doodle -------------------------------------------------------------

function drawDoodle()
{
    // doodle is createGraphics not createImage
    let theimage = doodle.get();
    // console.log (theimage);

    image ( theimage,   0,                0,    ZOOMPIXELS,     ZOOMPIXELS  );      // original
    image ( theimage,   ZOOMPIXELS+20,    0,    PIXELS,         PIXELS      );      // shrunk
}


function guessDoodle()
{
   // doodle is createGraphics not createImage
   let img = doodle.get();

  img.resize ( PIXELS, PIXELS );
  img.loadPixels();

  // set up inputs
  let inputs = [];
  for (let i = 0; i < PIXELSSQUARED ; i++)
  {
     inputs[i] = img.pixels[i * 4] / 255;
  }

  doodle_inputs = inputs;     // can inspect in console

//Dheera says- fetching the inputs and displaying the 1st two predictions for the doodle.
let myinput = getmycnnInputs(inputs);
  let a = find12(mycnnModel.forward(myinput).w);
  thehtml =   " Our 1st Guess is: " + bluespan + alphabets[a[0] - 1] +  "</span> <br>" +
            " Our 2nd Guess is: " + bluespan + alphabets[a[1] - 1] + "</span>";
  AB.msg ( thehtml, 2 );
}


function wipeDoodle()
{
    doodle_exists = false;
    doodle.background('black');
}




// --- debugging --------------------------------------------------
// in console
// showInputs(demo_inputs);
// showInputs(doodle_inputs);


function showInputs ( inputs )
// display inputs row by row, corresponding to square of pixels
{
    var str = "";
    for (let i = 0; i < inputs.length; i++)
    {
      if ( i % PIXELS === 0 )    str = str + "\n";                                   // new line for each row of pixels
      var value = inputs[i];
      str = str + " " + value.toFixed(2) ;
    }
    console.log (str);
}