JNotebooks/tutorial10_step_by_step_MNIST_digits_classification_cnn.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step-by-step MNIST Digits Classification - Convolutional Neural Networks\n",
    "\n",
    "1. Load dataset, explore it (display images, mean, min, max values, etc.) and split it into train, validation and test sets\n",
    "2. Data scaling\n",
    "3. One hot encoding\n",
    "4. Define your model, cost function, optimizer, learning rate\n",
    "5. Define your callbacks (save your model, patience, etc.)\n",
    "6. Train your model\n",
    "    \n",
    "    6.1 If you are satisfied with the train and validation performance go to the next step\n",
    "    \n",
    "    6.2 If you are not satisfied with the train and validation performance go back to step 5\n",
    "7. Test your model on the test and extract relevant metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pylab as plt\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "\n",
    "physical_devices = tf.config.experimental.list_physical_devices('GPU')\n",
    "tf.config.experimental.set_memory_growth(physical_devices[0], True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load dataset, explore it  and split it into train, validation and test sets\n",
    "\n",
    "- [Load MNIST Keras](https://keras.io/api/datasets/mnist/#load_data-function)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Development set\n",
      "Images:  (60000, 28, 28)\n",
      "Labels shape: (60000,)\n",
      "\n",
      "Number of classes: 10\n",
      "\n",
      "Classes: [0 1 2 3 4 5 6 7 8 9]\n",
      "\n",
      "Test set\n",
      "Images:  (10000, 28, 28)\n",
      "Labels shape:  (10000,)\n"
     ]
    }
   ],
   "source": [
    "# Loading the data using the Keras function\n",
    "(X_dev, Y_dev), (X_test, Y_test) = tf.keras.datasets.mnist.load_data() # The data comes already split \n",
    "                                                                        # in dev and test sets\n",
    "print(\"Development set\")\n",
    "print(\"Images: \",X_dev.shape)\n",
    "print(\"Labels shape:\",Y_dev.shape)\n",
    "print(\"\\nNumber of classes:\",np.unique(Y_dev).size)\n",
    "print(\"\\nClasses:\",np.unique(Y_dev))\n",
    "print(\"\\nTest set\")\n",
    "print(\"Images: \",X_test.shape)\n",
    "print(\"Labels shape: \",Y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "plt.hist(Y_dev, bins = range(11))\n",
    "plt.xlabel(\"Labels\")\n",
    "plt.ylabel(\"Label Frequency\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1728x1296 with 30 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Disaplying some samples from the development set\n",
    "sample_indexes = np.random.choice(np.arange(X_dev.shape[0], dtype = int),size = 30, replace = False)\n",
    "plt.figure(figsize = (24,18))\n",
    "for (ii,jj) in enumerate(sample_indexes):\n",
    "    plt.subplot(5,6,ii+1)\n",
    "    plt.imshow(X_dev[jj], cmap = \"gray\")\n",
    "    plt.title(\"Label: %d\" %Y_dev[jj])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Train set\n",
      "Images:  (45000, 28, 28)\n",
      "Labels shape:  (45000,)\n",
      "\n",
      "Validation set\n",
      "Images:  (15000, 28, 28)\n",
      "Labels shape:  (15000,)\n"
     ]
    }
   ],
   "source": [
    "#The number of classes across samples looks balanced\n",
    "# Let's shuffle the samples and split them\n",
    "indexes = np.arange(X_dev.shape[0], dtype = int)\n",
    "np.random.shuffle(indexes)\n",
    "X_dev = X_dev[indexes]\n",
    "Y_dev = Y_dev[indexes]\n",
    "\n",
    "nsplit = int(0.75*X_dev.shape[0]) # Train/validation split\n",
    "\n",
    "# Train and validation split\n",
    "X_train = X_dev[:nsplit]\n",
    "Y_train = Y_dev[:nsplit]\n",
    "X_val = X_dev[nsplit:]\n",
    "Y_val = Y_dev[nsplit:]\n",
    "\n",
    "print(\"\\nTrain set\")\n",
    "print(\"Images: \",X_train.shape)\n",
    "print(\"Labels shape: \",Y_train.shape)\n",
    "print(\"\\nValidation set\")\n",
    "print(\"Images: \",X_val.shape)\n",
    "print(\"Labels shape: \",Y_val.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 255 33.29481490929705 78.541712904855\n",
      "0 255 33.38924107142857 78.64472801055584\n"
     ]
    }
   ],
   "source": [
    "print(X_train.min(),X_train.max(),X_train.mean(),X_train.std())\n",
    "print(X_val.min(),X_val.max(),X_val.mean(),X_val.std())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Data Scaling "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "norm_type = 0 # 0 -> min-max; 1-> standardization\n",
    "\n",
    "if norm_type == 0:\n",
    "    X_train = X_train/255\n",
    "    X_val = X_val/255\n",
    "    X_test = X_test/255\n",
    "elif norm_type == 1:\n",
    "    train_mean, train_std = X_train.mean(),X_train.std() \n",
    "    X_train = (X_train - train_mean)/train_std\n",
    "    X_val = (X_val - train_mean)/train_std\n",
    "    X_test = (X_test - train_mean)/train_std\n",
    "else:\n",
    "    pass"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. One hot encoding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Labels:\n",
      "[3 5 0 6 3]\n",
      "\n",
      "One hot encoded labels:\n",
      "[[0. 0. 0. 1. 0. 0. 0. 0. 0. 0.]\n",
      " [0. 0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
      " [1. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
      " [0. 0. 0. 0. 0. 0. 1. 0. 0. 0.]\n",
      " [0. 0. 0. 1. 0. 0. 0. 0. 0. 0.]]\n"
     ]
    }
   ],
   "source": [
    "Y_train_oh = tf.keras.utils.to_categorical(Y_train)\n",
    "Y_val_oh = tf.keras.utils.to_categorical(Y_val)\n",
    "Y_test_oh = tf.keras.utils.to_categorical(Y_test)\n",
    "\n",
    "print(\"Labels:\")\n",
    "print(Y_train[:5])\n",
    "print()\n",
    "print(\"One hot encoded labels:\")\n",
    "print(Y_train_oh[:5])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Define your model, cost function, optimizer, learning rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def my_model(ishape = (28,28,1),k = 10, lr = 1e-4):\n",
    "    model_input = tf.keras.layers.Input(shape = ishape)\n",
    "    l1 = tf.keras.layers.Conv2D(48, (3,3), padding='same', activation='relu')(model_input)\n",
    "    l2 = tf.keras.layers.Conv2D(48, (3,3), padding='same', activation='relu')(l1)\n",
    "    l2_drop = tf.keras.layers.Dropout(0.25)(l2)\n",
    "    l3 = tf.keras.layers.MaxPool2D((2,2))(l2_drop)\n",
    "    l4 = tf.keras.layers.Conv2D(96, (3,3), padding='same', activation='relu')(l3)\n",
    "    l5 = tf.keras.layers.Conv2D(96, (3,3), padding='same', activation='relu')(l4)\n",
    "    l5_drop = tf.keras.layers.Dropout(0.25)(l5)\n",
    "    flat = tf.keras.layers.Flatten()(l5_drop)\n",
    "    out = tf.keras.layers.Dense(k,activation = 'softmax')(flat)\n",
    "    model = tf.keras.models.Model(inputs = model_input, outputs = out)\n",
    "    model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=lr), loss='categorical_crossentropy', metrics = [\"accuracy\"])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         [(None, 28, 28, 1)]       0         \n",
      "_________________________________________________________________\n",
      "conv2d (Conv2D)              (None, 28, 28, 48)        480       \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 28, 28, 48)        20784     \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 28, 28, 48)        0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 14, 14, 48)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 14, 14, 96)        41568     \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 14, 14, 96)        83040     \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 14, 14, 96)        0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 18816)             0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 10)                188170    \n",
      "=================================================================\n",
      "Total params: 334,042\n",
      "Trainable params: 334,042\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "model = my_model()\n",
    "print(model.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5. Define your callbacks (save your model, patience, etc.)\n",
    "\n",
    "- [Keras callbacks](https://keras.io/api/callbacks/)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_name = \"best_model_mnist_cnn.h5\"\n",
    "early_stop = tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience = 20)\n",
    "\n",
    "monitor = tf.keras.callbacks.ModelCheckpoint(model_name, monitor='val_loss',\\\n",
    "                                             verbose=0,save_best_only=True,\\\n",
    "                                             save_weights_only=True,\\\n",
    "                                             mode='min')\n",
    "# Learning rate schedule\n",
    "def scheduler(epoch, lr):\n",
    "    if epoch%10 == 0:\n",
    "        lr = lr/2\n",
    "    return lr\n",
    "\n",
    "lr_schedule = tf.keras.callbacks.LearningRateScheduler(scheduler,verbose = 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 6. Train your model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/1000\n",
      "1407/1407 [==============================] - 9s 6ms/step - loss: 0.4151 - accuracy: 0.8834 - val_loss: 0.1801 - val_accuracy: 0.9536 - lr: 5.0000e-05\n",
      "Epoch 2/1000\n",
      "1407/1407 [==============================] - 10s 7ms/step - loss: 0.1235 - accuracy: 0.9630 - val_loss: 0.1032 - val_accuracy: 0.9710 - lr: 5.0000e-05\n",
      "Epoch 3/1000\n",
      "1407/1407 [==============================] - 10s 7ms/step - loss: 0.0836 - accuracy: 0.9750 - val_loss: 0.0844 - val_accuracy: 0.9758 - lr: 5.0000e-05\n",
      "Epoch 4/1000\n",
      "1407/1407 [==============================] - 11s 8ms/step - loss: 0.0675 - accuracy: 0.9797 - val_loss: 0.0715 - val_accuracy: 0.9796 - lr: 5.0000e-05\n",
      "Epoch 5/1000\n",
      " 212/1407 [===>..........................] - ETA: 7s - loss: 0.0496 - accuracy: 0.9853"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-12-165a1d28adcf>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      1\u001b[0m model.fit(X_train[:,:,:,np.newaxis],Y_train_oh,batch_size = 32, epochs = 1000, \\\n\u001b[0;32m----> 2\u001b[0;31m           verbose = 1, callbacks= [early_stop, monitor, lr_schedule],validation_data=(X_val[:,:,:,np.newaxis],Y_val_oh))\n\u001b[0m",
      "\u001b[0;32m~/Documents/software/virtualenv/venv01/lib/python3.6/site-packages/tensorflow/python/keras/engine/training.py\u001b[0m in \u001b[0;36m_method_wrapper\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m     64\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0m_method_wrapper\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     65\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_in_multi_worker_mode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m  \u001b[0;31m# pylint: disable=protected-access\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 66\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mmethod\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     67\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     68\u001b[0m     \u001b[0;31m# Running inside `run_distribute_coordinator` already.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Documents/software/virtualenv/venv01/lib/python3.6/site-packages/tensorflow/python/keras/engine/training.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[1;32m    853\u001b[0m                 \u001b[0mcontext\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0masync_wait\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    854\u001b[0m               \u001b[0mlogs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtmp_logs\u001b[0m  \u001b[0;31m# No error, now safe to assign to logs.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 855\u001b[0;31m               \u001b[0mcallbacks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_train_batch_end\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlogs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    856\u001b[0m         \u001b[0mepoch_logs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlogs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    857\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Documents/software/virtualenv/venv01/lib/python3.6/site-packages/tensorflow/python/keras/callbacks.py\u001b[0m in \u001b[0;36mon_train_batch_end\u001b[0;34m(self, batch, logs)\u001b[0m\n\u001b[1;32m    388\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_should_call_train_batch_hooks\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    389\u001b[0m       \u001b[0mlogs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_process_logs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlogs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 390\u001b[0;31m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_call_batch_hook\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mModeKeys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTRAIN\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'end'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlogs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mlogs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    391\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    392\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0mon_test_batch_begin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlogs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Documents/software/virtualenv/venv01/lib/python3.6/site-packages/tensorflow/python/keras/callbacks.py\u001b[0m in \u001b[0;36m_call_batch_hook\u001b[0;34m(self, mode, hook, batch, logs)\u001b[0m\n\u001b[1;32m    296\u001b[0m     \u001b[0;32mfor\u001b[0m \u001b[0mcallback\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcallbacks\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    297\u001b[0m       \u001b[0mbatch_hook\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcallback\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhook_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 298\u001b[0;31m       \u001b[0mbatch_hook\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlogs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    299\u001b[0m     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_delta_ts\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mhook_name\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mt_before_callbacks\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    300\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Documents/software/virtualenv/venv01/lib/python3.6/site-packages/tensorflow/python/keras/callbacks.py\u001b[0m in \u001b[0;36mon_train_batch_end\u001b[0;34m(self, batch, logs)\u001b[0m\n\u001b[1;32m    882\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    883\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0mon_train_batch_end\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlogs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 884\u001b[0;31m     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_batch_update_progbar\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlogs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    885\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    886\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0mon_test_batch_end\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlogs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Documents/software/virtualenv/venv01/lib/python3.6/site-packages/tensorflow/python/keras/callbacks.py\u001b[0m in \u001b[0;36m_batch_update_progbar\u001b[0;34m(self, logs)\u001b[0m\n\u001b[1;32m    926\u001b[0m     \u001b[0madd_seen\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnum_steps\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0muse_steps\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mnum_steps\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    927\u001b[0m     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mseen\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0madd_seen\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 928\u001b[0;31m     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprogbar\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupdate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mseen\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlogs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mitems\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinalize\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    929\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    930\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0m_finalize_progbar\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlogs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Documents/software/virtualenv/venv01/lib/python3.6/site-packages/tensorflow/python/keras/utils/generic_utils.py\u001b[0m in \u001b[0;36mupdate\u001b[0;34m(self, current, values, finalize)\u001b[0m\n\u001b[1;32m    653\u001b[0m         \u001b[0minfo\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;34m'\\n'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    654\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 655\u001b[0;31m       \u001b[0msys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstdout\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwrite\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minfo\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    656\u001b[0m       \u001b[0msys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstdout\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mflush\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    657\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Documents/software/virtualenv/venv01/lib/python3.6/site-packages/ipykernel/iostream.py\u001b[0m in \u001b[0;36mwrite\u001b[0;34m(self, string)\u001b[0m\n\u001b[1;32m    402\u001b[0m             \u001b[0mis_child\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_is_master_process\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    403\u001b[0m             \u001b[0;31m# only touch the buffer in the IO thread to avoid races\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 404\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpub_thread\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mschedule\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mlambda\u001b[0m \u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_buffer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwrite\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstring\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    405\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mis_child\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    406\u001b[0m                 \u001b[0;31m# mp.Pool cannot be trusted to flush promptly (or ever),\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Documents/software/virtualenv/venv01/lib/python3.6/site-packages/ipykernel/iostream.py\u001b[0m in \u001b[0;36mschedule\u001b[0;34m(self, f)\u001b[0m\n\u001b[1;32m    203\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_events\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    204\u001b[0m             \u001b[0;31m# wake event thread (message content is ignored)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 205\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_event_pipe\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mb''\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    206\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    207\u001b[0m             \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Documents/software/virtualenv/venv01/lib/python3.6/site-packages/ipykernel/iostream.py\u001b[0m in \u001b[0;36m_event_pipe\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m     95\u001b[0m         \u001b[0;34m\"\"\"thread-local event pipe for signaling events that should be processed in the thread\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     96\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 97\u001b[0;31m             \u001b[0mevent_pipe\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_local\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mevent_pipe\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     98\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mAttributeError\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     99\u001b[0m             \u001b[0;31m# new thread, new event pipe\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "model.fit(X_train[:,:,:,np.newaxis],Y_train_oh,batch_size = 32, epochs = 1000, \\\n",
    "          verbose = 1, callbacks= [early_stop, monitor, lr_schedule],validation_data=(X_val[:,:,:,np.newaxis],Y_val_oh))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 7. Test your model on the test and extract relevant metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 1s 3ms/step - loss: 0.0585 - accuracy: 0.9837\n",
      "Categorical cross-entropy: 0.05848461762070656\n",
      "Accuracy: 0.9836999773979187\n"
     ]
    }
   ],
   "source": [
    "model.load_weights(model_name)\n",
    "metrics = model.evaluate(X_test[:,:,:,np.newaxis],Y_test_oh)\n",
    "print(\"Categorical cross-entropy:\", metrics[0])\n",
    "print(\"Accuracy:\", metrics[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "163\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1728x1296 with 30 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Ypred = model.predict(X_test[:,:,:,np.newaxis]).argmax(axis = 1)\n",
    "wrong_indexes = np.where(Ypred != Y_test)[0]\n",
    "print(wrong_indexes.size)\n",
    "\n",
    "# Disaplying some samples from the development set\n",
    "sample_indexes = np.random.choice(np.arange(wrong_indexes.shape[0], dtype = int),size = 30, replace = False)\n",
    "plt.figure(figsize = (24,18))\n",
    "for (ii,jj) in enumerate(sample_indexes):\n",
    "    plt.subplot(5,6,ii+1)\n",
    "    plt.imshow(X_test[wrong_indexes[jj]], cmap = \"gray\")\n",
    "    plt.title(\"Label: %d, predicted: %d\" %(Y_test[wrong_indexes[jj]],Ypred[wrong_indexes[jj]]))\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}