.ipynb_checkpoints/FindingMino_Train-checkpoint.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import itertools\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from keras.utils.np_utils import to_categorical\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n",
    "from tensorflow.keras.models import Sequential, load_model\n",
    "from keras.datasets import mnist\n",
    "import itertools\n",
    "import random"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# FindingMino Miniproject\n",
    "\n",
    "- **Task**: take as input two MNIST images and through a neural network return the value of the minimum digit\n",
    "\n",
    "- **Restrictions**: no intermediate labels can be used, ie a classifier that reveals what digits the images represent cannot be followed by logic to choose the minimum value\n",
    "\n",
    "- **Approach**: as a variation of Gonzalo's proposed method, place images side-by-side and then make the label of that combined image be the minimum value\n",
    "\n",
    "- **Implementation**: convolutional neural network with three convolutional layers interspersed by two max pooling layers, followed by a deep neural network with one hidden layer and a dropout layer prior to the output layer. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load data and split into training and test\n",
    "(train_images, train_labels), (test_images, test_labels) = mnist.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(28, 28)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x10fb346d0>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "image_A = train_images[0]\n",
    "print(image_A.shape)\n",
    "plt.imshow(image_A, cmap = plt.cm.binary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x11f2e14f0>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD4CAYAAAAq5pAIAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAAAOR0lEQVR4nO3df6jUdb7H8df7titBrmF5klNK7l3OP7GQ2iC3jPXc9C4mkS1BKricS4XST5eMbrh/rJSBSNsSFEvuTdYTm9vSWorF7nbFiIVaG+WUVlzrhqHmjxFBkyLX9n3/ON+Wk53vZ8aZ78x39P18wDAz3/d8z/fdt159Z76f+c7H3F0Azn//UnYDADqDsANBEHYgCMIOBEHYgSC+08mNTZgwwadMmdLJTQKh7N27V0ePHrXRai2F3czmSnpS0gWS/tvdV6deP2XKFFWr1VY2CSChUqnk1pp+G29mF0h6WtKNkq6StMjMrmr27wFor1Y+s8+Q9JG7f+zupyT9XtL8YtoCULRWwn6FpH0jnu/Pln2DmS0xs6qZVWu1WgubA9CKtp+Nd/e17l5x90pPT0+7NwcgRythPyBp8ojnk7JlALpQK2F/W1KfmX3fzMZIWihpczFtASha00Nv7n7azO6V9GcND72tc/f3CusMQKFaGmd391clvVpQLwDaiK/LAkEQdiAIwg4EQdiBIAg7EARhB4Ig7EAQhB0IgrADQRB2IAjCDgRB2IEgCDsQBGEHgiDsQBCEHQiCsANBEHYgCMIOBEHYgSAIOxBER6dsxvlnx44dyfpTTz2VW1u/fn1y3YGBgWT9vvvuS9anT5+erEfDkR0IgrADQRB2IAjCDgRB2IEgCDsQBGEHgmCcHUlDQ0PJ+pw5c5L1EydO5NbMLLnu4OBgsr5p06Zk/dixY8l6NC2F3cz2SvpM0leSTrt7pYimABSviCP7v7v70QL+DoA24jM7EESrYXdJfzGzHWa2ZLQXmNkSM6uaWbVWq7W4OQDNajXs17v7dEk3SrrHzH505gvcfa27V9y90tPT0+LmADSrpbC7+4Hs/oiklyTNKKIpAMVrOuxmdpGZfe/rx5J+LGl3UY0BKFYrZ+MnSnopGyv9jqTn3f1PhXSFjtm+fXuyfuuttybrx48fT9ZTY+njxo1LrjtmzJhk/ejR9CDQm2++mVu75pprWtr2uajpsLv7x5KuLrAXAG3E0BsQBGEHgiDsQBCEHQiCsANBcInreeDzzz/Pre3cuTO57uLFi5P1Tz/9tKmeGtHX15esP/TQQ8n6ggULkvWZM2fm1latWpVcd8WKFcn6uYgjOxAEYQeCIOxAEIQdCIKwA0EQdiAIwg4EwTj7eWDp0qW5teeff76DnZydetM9nzx5MlmfNWtWsv7666/n1nbt2pVc93zEkR0IgrADQRB2IAjCDgRB2IEgCDsQBGEHgmCc/RxQbzx6y5YtuTV3b2nb/f39yfpNN92UrD/44IO5tcsvvzy57rRp05L18ePHJ+vbtm3LrbW6X85FHNmBIAg7EARhB4Ig7EAQhB0IgrADQRB2IAjG2bvA0NBQsj5nzpxk/cSJE7m11JTJkjRv3rxkfcOGDcl66ppxSXrsscdya3feeWdy3Z6enmT96qvTkwin/tlfeeWV5Lr1fm9/+vTpyXo3qntkN7N1ZnbEzHaPWHaJmb1mZh9m9+lvNwAoXSNv438rae4Zyx6WtNXd+yRtzZ4D6GJ1w+7ub0g6dsbi+ZLWZ4/XS7ql2LYAFK3ZE3QT3f1g9viQpIl5LzSzJWZWNbNqrVZrcnMAWtXy2XgfvqIg96oCd1/r7hV3r9Q74QKgfZoN+2Ez65Wk7P5IcS0BaIdmw75Z0kD2eEDSpmLaAdAudcfZzWyDpH5JE8xsv6RfSFot6Q9mdoekTyTd1s4mz3V79uxJ1tesWZOsHz9+PFlPfTzq7e1NrjswMJCsjx07Nlmvdz17vXpZUnPaS9Ljjz+erHfz7/HnqRt2d1+UU5pdcC8A2oivywJBEHYgCMIOBEHYgSAIOxAEl7gW4Msvv0zWUz+nLNW/3HLcuHHJ+uDgYG6tUqkk1/3iiy+S9aj27dtXdguF48gOBEHYgSAIOxAEYQeCIOxAEIQdCIKwA0Ewzl6Aej87XG8cvZ5Nm9I/FzBr1qyW/j5i4MgOBEHYgSAIOxAEYQeCIOxAEIQdCIKwA0Ewzl6ABx54IFkfnjQnX39/f7LOOHpz6u33dq3brTiyA0EQdiAIwg4EQdiBIAg7EARhB4Ig7EAQjLM3aMuWLbm1oaGh5LpmlqzffPPNzbSEOlL7vd6/k6lTpxbcTfnqHtnNbJ2ZHTGz3SOWrTSzA2Y2lN3mtbdNAK1q5G38byXNHWX5r9x9anZ7tdi2ABStbtjd/Q1JxzrQC4A2auUE3b1m9m72Nn983ovMbImZVc2sWqvVWtgcgFY0G/ZfS/qBpKmSDkr6Zd4L3X2tu1fcvdLT09Pk5gC0qqmwu/thd//K3f8h6TeSZhTbFoCiNRV2M+sd8fQnknbnvRZAd6g7zm5mGyT1S5pgZvsl/UJSv5lNleSS9kpa2r4Wu0NqHvNTp04l173sssuS9QULFjTV0/mu3rz3K1eubPpvz549O1lfvXp103+7W9UNu7svGmXxs23oBUAb8XVZIAjCDgRB2IEgCDsQBGEHguAS1w648MILk/Xe3t5k/XxVb2ht1apVyfqaNWuS9cmTJ+fWli9fnlx37Nixyfq5iCM7EARhB4Ig7EAQhB0IgrADQRB2IAjCDgTBOHsHRP6p6NTPbNcbJ3/hhReS9fnz5yfrGzduTNaj4cgOBEHYgSAIOxAEYQeCIOxAEIQdCIKwA0Ewzt4gd2+qJkkvv/xysv7kk08201JXeOKJJ5L1Rx99NLd2/Pjx5LqLFy9O1gcHB5N1fBNHdiAIwg4EQdiBIAg7EARhB4Ig7EAQhB0IgnH2BplZUzVJOnToULJ+//33J+u33357sn7ppZfm1t56663kus8991yy/s477yTr+/btS9avvPLK3NrcuXOT6959993JOs5O3SO7mU02s21m9r6ZvWdmy7Lll5jZa2b2YXY/vv3tAmhWI2/jT0ta7u5XSfo3SfeY2VWSHpa01d37JG3NngPoUnXD7u4H3X1n9vgzSR9IukLSfEnrs5etl3RLm3oEUICzOkFnZlMkTZP0N0kT3f1gVjokaWLOOkvMrGpm1Vqt1kqvAFrQcNjNbKykP0r6mbufGFnz4StBRr0axN3XunvF3Ss9PT0tNQugeQ2F3cy+q+Gg/87dv/7JzsNm1pvVeyUdaU+LAIpQd+jNhseVnpX0gbuPvJ5xs6QBSauz+01t6fA8cPr06WT96aefTtZffPHFZP3iiy/Ore3Zsye5bquuu+66ZP2GG27IrT3yyCNFt4OERsbZZ0r6qaRdZjaULVuh4ZD/wczukPSJpNva0iGAQtQNu7v/VVLet0ZmF9sOgHbh67JAEIQdCIKwA0EQdiAIwg4EwSWuDbr22mtzazNmzEiuu3379pa2Xe8S2cOHDzf9tydMmJCsL1y4MFk/l38GOxqO7EAQhB0IgrADQRB2IAjCDgRB2IEgCDsQBOPsDZo0aVJubePGjbk1SXrmmWeS9dS0xq1atmxZsn7XXXcl6319fUW2gxJxZAeCIOxAEIQdCIKwA0EQdiAIwg4EQdiBIGx4MpfOqFQqXq1WO7Y9IJpKpaJqtTrqr0FzZAeCIOxAEIQdCIKwA0EQdiAIwg4EQdiBIOqG3cwmm9k2M3vfzN4zs2XZ8pVmdsDMhrLbvPa3C6BZjfx4xWlJy919p5l9T9IOM3stq/3K3R9vX3sAitLI/OwHJR3MHn9mZh9IuqLdjQEo1ll9ZjezKZKmSfpbtuheM3vXzNaZ2ficdZaYWdXMqrVarbVuATSt4bCb2VhJf5T0M3c/IenXkn4gaaqGj/y/HG09d1/r7hV3r/T09LTeMYCmNBR2M/uuhoP+O3ffKEnuftjdv3L3f0j6jaT07IYAStXI2XiT9KykD9z9iRHLe0e87CeSdhffHoCiNHI2fqakn0raZWZD2bIVkhaZ2VRJLmmvpKVt6A9AQRo5G/9XSaNdH/tq8e0AaBe+QQcEQdiBIAg7EARhB4Ig7EAQhB0IgrADQRB2IAjCDgRB2IEgCDsQBGEHgiDsQBCEHQiio1M2m1lN0icjFk2QdLRjDZydbu2tW/uS6K1ZRfZ2pbuP+vtvHQ37tzZuVnX3SmkNJHRrb93al0RvzepUb7yNB4Ig7EAQZYd9bcnbT+nW3rq1L4nemtWR3kr9zA6gc8o+sgPoEMIOBFFK2M1srpn9r5l9ZGYPl9FDHjPba2a7smmoqyX3ss7MjpjZ7hHLLjGz18zsw+x+1Dn2SuqtK6bxTkwzXuq+K3v6845/ZjezCyTtkfQfkvZLelvSInd/v6ON5DCzvZIq7l76FzDM7EeSTkoadPcfZsvWSDrm7quz/1GOd/f/6pLeVko6WfY03tlsRb0jpxmXdIuk/1SJ+y7R123qwH4r48g+Q9JH7v6xu5+S9HtJ80voo+u5+xuSjp2xeL6k9dnj9Rr+j6XjcnrrCu5+0N13Zo8/k/T1NOOl7rtEXx1RRtivkLRvxPP96q753l3SX8xsh5ktKbuZUUx094PZ40OSJpbZzCjqTuPdSWdMM941+66Z6c9bxQm6b7ve3adLulHSPdnb1a7kw5/BumnstKFpvDtllGnG/6nMfdfs9OetKiPsByRNHvF8UrasK7j7gez+iKSX1H1TUR/+egbd7P5Iyf38UzdN4z3aNOPqgn1X5vTnZYT9bUl9ZvZ9MxsjaaGkzSX08S1mdlF24kRmdpGkH6v7pqLeLGkgezwgaVOJvXxDt0zjnTfNuEred6VPf+7uHb9JmqfhM/L/J+nnZfSQ09e/Snonu71Xdm+SNmj4bd3fNXxu4w5Jl0raKulDSf8j6ZIu6u05SbskvavhYPWW1Nv1Gn6L/q6koew2r+x9l+irI/uNr8sCQXCCDgiCsANBEHYgCMIOBEHYgSAIOxAEYQeC+H8dj1XrNRdSpAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "image_B = train_images[1]\n",
    "plt.imshow(image_B, cmap = plt.cm.binary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x11f34d6a0>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "images_A_and_B = np.hstack((image_A, image_B))\n",
    "plt.imshow(images_A_and_B, cmap = plt.cm.binary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split training set further into training and validation sets\n",
    "sample_size = 40000\n",
    "X_train, X_val, y_train, y_val = train_images[:sample_size], train_images[sample_size:sample_size*2],  train_labels[:sample_size], train_labels[sample_size:sample_size*2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Randomly sampling a subset of the data, to make the combinations more tractable. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Randomly draw from 40k training images\n",
    "\n",
    "num_images_chosen = 400000\n",
    "num_images_total = 40000\n",
    "\n",
    "combo_train_X = []\n",
    "combo_train_y = []\n",
    "\n",
    "for i in range(num_images_chosen):\n",
    "    \n",
    "    # Randomly draw indices\n",
    "    rand_A = random.randint(0, num_images_total-1)\n",
    "    rand_B = random.randint(0, num_images_total-1)\n",
    "    \n",
    "    # Concatenate images\n",
    "    image_A = X_train[rand_A]\n",
    "    image_B = X_train[rand_B]\n",
    "    concatenated_images = np.hstack((image_A, image_B))\n",
    "    \n",
    "    # Find the minimum label\n",
    "    intermediate_label_A = y_train[rand_A]\n",
    "    intermediate_label_B = y_train[rand_B]\n",
    "    minimum_label = min(intermediate_label_A, intermediate_label_B)\n",
    "    \n",
    "    # Append to lists\n",
    "    combo_train_X.append(concatenated_images)\n",
    "    combo_train_y.append(minimum_label)\n",
    "\n",
    "# Convert to numpy arrays so that they can be processed by the network\n",
    "combo_train_X = np.array(combo_train_X)\n",
    "combo_train_y = np.array(combo_train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Randomly draw 20k validation images\n",
    "\n",
    "num_images_chosen = 10000\n",
    "num_images_total = 20000\n",
    "\n",
    "combo_val_X = []\n",
    "combo_val_y = []\n",
    "\n",
    "for i in range(num_images_chosen):\n",
    "    \n",
    "    # Randomly draw indices\n",
    "    rand_A = random.randint(0, num_images_total-1)\n",
    "    rand_B = random.randint(0, num_images_total-1)\n",
    "    \n",
    "    # Concatenate images\n",
    "    image_A = X_val[rand_A]\n",
    "    image_B = X_val[rand_B]\n",
    "    concatenated_images = np.hstack((image_A, image_B))\n",
    "    \n",
    "    # Find the minimum label\n",
    "    intermediate_label_A = y_val[rand_A]\n",
    "    intermediate_label_B = y_val[rand_B]\n",
    "    minimum_label = min(intermediate_label_A, intermediate_label_B)\n",
    "    \n",
    "    # Append to lists\n",
    "    combo_val_X.append(concatenated_images)\n",
    "    combo_val_y.append(minimum_label)\n",
    "    \n",
    "combo_val_X = np.array(combo_val_X)\n",
    "combo_val_y = np.array(combo_val_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(400000, 28, 56)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combo_train_X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reshape the data so they can be fed into the network, one-hot-encode the labels\n",
    "\n",
    "train_images = combo_train_X.reshape((combo_train_X.shape[0], 28, 56, 1))\n",
    "train_images = train_images/ 255\n",
    "\n",
    "val_images = combo_val_X.reshape((combo_val_X.shape[0], 28, 56, 1))\n",
    "val_images = val_images/255\n",
    "\n",
    "train_labels = to_categorical(combo_train_y)\n",
    "val_labels = to_categorical(combo_val_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d (Conv2D)              (None, 26, 54, 32)        320       \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 13, 27, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 11, 25, 64)        18496     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 5, 12, 64)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 3, 10, 64)         36928     \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 1920)              0         \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 1920)              0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 64)                122944    \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 10)                650       \n",
      "=================================================================\n",
      "Total params: 179,338\n",
      "Trainable params: 179,338\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# Model architecture\n",
    "\n",
    "# Convolutional NN\n",
    "model = tf.keras.Sequential()\n",
    "model.add(layers.Conv2D(32, (3,3), activation = 'relu', input_shape = (28,56,1)))\n",
    "model.add(layers.MaxPooling2D(2,2))\n",
    "model.add(layers.Conv2D(64, (3,3), activation = 'relu'))\n",
    "model.add(layers.MaxPooling2D(2,2))\n",
    "model.add(layers.Conv2D(64, (3,3), activation = 'relu'))\n",
    "# Adding a NN Classifier\n",
    "model.add(layers.Flatten())\n",
    "model.add(layers.Dense(64, activation = 'relu'))\n",
    "model.add(layers.Dense(10, activation = 'softmax'))\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compile model\n",
    "opt = tf.keras.optimizers.RMSprop(learning_rate=0.0002)\n",
    "model.compile(optimizer = opt,\n",
    "             loss = 'categorical_crossentropy',\n",
    "             metrics = ['mae', 'accuracy'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/8\n",
      "2734/6250 [============>.................] - ETA: 3:43 - loss: 0.5052 - mae: 0.0491 - accuracy: 0.8365"
     ]
    },
    {
     "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-13-203dbf38761e>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;31m# Fit model\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m model.fit(train_images, \n\u001b[0m\u001b[1;32m      3\u001b[0m           \u001b[0mtrain_labels\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m           \u001b[0mepochs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m8\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m           batch_size = 64)\n",
      "\u001b[0;32m/usr/local/lib/python3.9/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   1181\u001b[0m                 _r=1):\n\u001b[1;32m   1182\u001b[0m               \u001b[0mcallbacks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_train_batch_begin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1183\u001b[0;31m               \u001b[0mtmp_logs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterator\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   1184\u001b[0m               \u001b[0;32mif\u001b[0m \u001b[0mdata_handler\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshould_sync\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1185\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[0;32m/usr/local/lib/python3.9/site-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    887\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    888\u001b[0m       \u001b[0;32mwith\u001b[0m \u001b[0mOptionalXlaContext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_jit_compile\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--> 889\u001b[0;31m         \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_call\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[0mkwds\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    890\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    891\u001b[0m       \u001b[0mnew_tracing_count\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexperimental_get_tracing_count\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/usr/local/lib/python3.9/site-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36m_call\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    915\u001b[0m       \u001b[0;31m# In this case we have created variables on the first call, so we run the\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    916\u001b[0m       \u001b[0;31m# defunned version which is guaranteed to never create variables.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 917\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_stateless_fn\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[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m  \u001b[0;31m# pylint: disable=not-callable\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    918\u001b[0m     \u001b[0;32melif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_stateful_fn\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    919\u001b[0m       \u001b[0;31m# Release the lock early so that multiple threads can perform the call\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.9/site-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   3021\u001b[0m       (graph_function,\n\u001b[1;32m   3022\u001b[0m        filtered_flat_args) = self._maybe_define_function(args, kwargs)\n\u001b[0;32m-> 3023\u001b[0;31m     return graph_function._call_flat(\n\u001b[0m\u001b[1;32m   3024\u001b[0m         filtered_flat_args, captured_inputs=graph_function.captured_inputs)  # pylint: disable=protected-access\n\u001b[1;32m   3025\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.9/site-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m_call_flat\u001b[0;34m(self, args, captured_inputs, cancellation_manager)\u001b[0m\n\u001b[1;32m   1958\u001b[0m         and executing_eagerly):\n\u001b[1;32m   1959\u001b[0m       \u001b[0;31m# No tape is watching; skip to running the function.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1960\u001b[0;31m       return self._build_call_outputs(self._inference_function.call(\n\u001b[0m\u001b[1;32m   1961\u001b[0m           ctx, args, cancellation_manager=cancellation_manager))\n\u001b[1;32m   1962\u001b[0m     forward_backward = self._select_forward_and_backward_functions(\n",
      "\u001b[0;32m/usr/local/lib/python3.9/site-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36mcall\u001b[0;34m(self, ctx, args, cancellation_manager)\u001b[0m\n\u001b[1;32m    589\u001b[0m       \u001b[0;32mwith\u001b[0m \u001b[0m_InterpolateFunctionError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\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    590\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mcancellation_manager\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 591\u001b[0;31m           outputs = execute.execute(\n\u001b[0m\u001b[1;32m    592\u001b[0m               \u001b[0mstr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msignature\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mname\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    593\u001b[0m               \u001b[0mnum_outputs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_num_outputs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.9/site-packages/tensorflow/python/eager/execute.py\u001b[0m in \u001b[0;36mquick_execute\u001b[0;34m(op_name, num_outputs, inputs, attrs, ctx, name)\u001b[0m\n\u001b[1;32m     57\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[1;32m     58\u001b[0m     \u001b[0mctx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mensure_initialized\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---> 59\u001b[0;31m     tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,\n\u001b[0m\u001b[1;32m     60\u001b[0m                                         inputs, attrs, num_outputs)\n\u001b[1;32m     61\u001b[0m   \u001b[0;32mexcept\u001b[0m \u001b[0mcore\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_NotOkStatusException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\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": [
    "# Fit model\n",
    "model.fit(train_images, \n",
    "          train_labels, \n",
    "          epochs = 8, \n",
    "          batch_size = 64)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a set of predictions\n",
    "y_predict = model.predict(val_images)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Val Accuracy: 0.983\n"
     ]
    }
   ],
   "source": [
    "# Compare predictions to validation labels and return the proportion correctly classified\n",
    "correct = 0\n",
    "\n",
    "for label in range(len(val_labels)):\n",
    "    \n",
    "    if np.argmax(val_labels[label], axis=None, out=None) == np.argmax(y_predict[label], axis=None, out=None):\n",
    "        correct = correct + 1\n",
    "        \n",
    "print(\"Val Accuracy: \" + str(correct/len(val_labels)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Print loss, mae and accuracy\n",
    "eval_model = model.evaluate(val_images, val_labels)\n",
    "eval_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved model to disk\n"
     ]
    }
   ],
   "source": [
    "# serialize model to JSON\n",
    "model_json = model.to_json()\n",
    "with open(\"Trained_Models/Model.json\", \"w\") as json_file:\n",
    "    json_file.write(model_json)\n",
    "# serialize weights to HDF5\n",
    "model.save_weights(\"Trained_Models/Model.h5\")\n",
    "print(\"Saved model to disk\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Verify above by visually inspecting images from the predicted set\n",
    "num_val_images = 10000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicted: 5\n",
      "True: 7\n"
     ]
    },
    {
     "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": [
    "# Run this cell repeatedly to see multiple examples\n",
    "check_index = random.randint(0,num_val_images)\n",
    "plt.imshow(val_images[check_index].reshape(28,56), cmap = plt.cm.binary)\n",
    "print(\"Predicted: \" + str(np.argmax(y_predict[check_index])))\n",
    "print(\"True: \" + str(np.argmax(val_labels[check_index])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "general_ML",
   "language": "python",
   "name": "general_ml"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}