lab-07-4-mnist_introduction-tf2.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "60000 60000\n",
      "tf.Tensor(\n",
      "[[0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " ...\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]], shape=(60000, 784), dtype=float32) tf.Tensor(\n",
      "[[0. 0. 0. ... 0. 0. 0.]\n",
      " [1. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " ...\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 1. 0.]], shape=(60000, 10), dtype=float32)\n",
      "10000 10000\n",
      "tf.Tensor(\n",
      "[[0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " ...\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]], shape=(10000, 784), dtype=float32) tf.Tensor(\n",
      "[[0. 0. 0. ... 1. 0. 0.]\n",
      " [0. 0. 1. ... 0. 0. 0.]\n",
      " [0. 1. 0. ... 0. 0. 0.]\n",
      " ...\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]], shape=(10000, 10), dtype=float32)\n",
      "Epoch: 0001, Cost: 2.557298183\n",
      "Epoch: 0002, Cost: 1.021223426\n",
      "Epoch: 0003, Cost: 0.808580279\n",
      "Epoch: 0004, Cost: 0.706655800\n",
      "Epoch: 0005, Cost: 0.643622339\n",
      "Epoch: 0006, Cost: 0.599544644\n",
      "Epoch: 0007, Cost: 0.566324413\n",
      "Epoch: 0008, Cost: 0.540010393\n",
      "Epoch: 0009, Cost: 0.518421769\n",
      "Epoch: 0010, Cost: 0.500244260\n",
      "Epoch: 0011, Cost: 0.484631658\n",
      "Epoch: 0012, Cost: 0.471008748\n",
      "Epoch: 0013, Cost: 0.458965391\n",
      "Epoch: 0014, Cost: 0.448202401\n",
      "Epoch: 0015, Cost: 0.438495845\n",
      "Learning finished\n",
      "Accuracy:  0.8946\n",
      "Label:  [2]\n",
      "Prediction:  [2]\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"
    },
    {
     "data": {
      "text/plain": [
       "'\\nEpoch: 0001, Cost: 2.826302672\\nEpoch: 0002, Cost: 1.061668952\\nEpoch: 0003, Cost: 0.838061315\\nEpoch: 0004, Cost: 0.733232745\\nEpoch: 0005, Cost: 0.669279885\\nEpoch: 0006, Cost: 0.624611836\\nEpoch: 0007, Cost: 0.591160344\\nEpoch: 0008, Cost: 0.563868987\\nEpoch: 0009, Cost: 0.541745171\\nEpoch: 0010, Cost: 0.522673578\\nEpoch: 0011, Cost: 0.506782325\\nEpoch: 0012, Cost: 0.492447643\\nEpoch: 0013, Cost: 0.479955837\\nEpoch: 0014, Cost: 0.468893674\\nEpoch: 0015, Cost: 0.458703488\\nLearning finished\\nAccuracy:  0.8951\\n'"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Lab 7 Learning rate and Evaluation\n",
    "from functools import partial as bind\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import random\n",
    "import tensorflow as tf\n",
    "assert(tf.__version__.find('2') == 0)\n",
    "tf.random.set_seed(777)  # for reproducibility\n",
    "\n",
    "(x_train, y_train), (x_test, y_test) = tf.keras.datasets.mnist.load_data()\n",
    "x_train, x_test = x_train / 255.0, x_test / 255.0\n",
    "\n",
    "x_train = tf.reshape(x_train, [-1, 28*28])\n",
    "x_train = tf.cast(x_train, tf.float32)\n",
    "y_train = tf.one_hot(y_train, 10)\n",
    "y_train = tf.cast(y_train, tf.float32)\n",
    "x_train_len = len(x_train)\n",
    "y_train_len = len(y_train)\n",
    "print(x_train_len, y_train_len)\n",
    "print(x_train, y_train)\n",
    "\n",
    "x_test = tf.reshape(x_test, [-1, 28*28])\n",
    "x_test = tf.cast(x_test, tf.float32)\n",
    "y_test = tf.one_hot(y_test, 10)\n",
    "x_test_len = len(x_test)\n",
    "y_test_len = len(y_test)\n",
    "print(x_test_len, y_test_len)\n",
    "print(x_test, y_test)\n",
    "\n",
    "nb_classes = 10\n",
    "\n",
    "W = tf.Variable(tf.random.normal([784, nb_classes]))\n",
    "b = tf.Variable(tf.random.normal([nb_classes]))\n",
    "\n",
    "# Hypothesis (using softmax)\n",
    "hypothesis = lambda X: tf.nn.softmax(tf.matmul(X, W) + b)\n",
    "\n",
    "@tf.function\n",
    "def cost(X, Y):\n",
    "    return tf.reduce_mean(-tf.reduce_sum(Y * tf.math.log(hypothesis(X)), axis=1))\n",
    "\n",
    "optimizer = tf.keras.optimizers.SGD(learning_rate=0.1)\n",
    "\n",
    "# Test model\n",
    "is_correct = lambda X, Y: tf.equal(tf.argmax(hypothesis(X), 1), tf.argmax(Y, 1))\n",
    "# Calculate accuracy\n",
    "accuracy = lambda X, Y: tf.reduce_mean(tf.cast(is_correct(X, Y), tf.float32))\n",
    "\n",
    "# parameters\n",
    "num_epochs = 15\n",
    "batch_size = 100\n",
    "num_iterations = int(x_train_len / batch_size)\n",
    "\n",
    "for epoch in range(num_epochs):\n",
    "    avg_cost = 0\n",
    "\n",
    "    for i in range(num_iterations):\n",
    "        pos = i * batch_size\n",
    "        batch_xs = x_train[pos: pos + batch_size, :]\n",
    "        batch_ys = y_train[pos: pos + batch_size, :]\n",
    "#         print(batch_xs)\n",
    "#         print(batch_ys)\n",
    "        \n",
    "        # Minimize\n",
    "        optimizer.minimize(bind(cost, batch_xs, batch_ys), var_list=[W, b])\n",
    "        cost_val = cost(batch_xs, batch_ys)\n",
    "        avg_cost += cost_val / num_iterations\n",
    "\n",
    "    tf.print(\"Epoch: {:04d}, Cost: {:.9f}\".format(epoch + 1, avg_cost))\n",
    "\n",
    "tf.print(\"Learning finished\")\n",
    "\n",
    "# Test the model using test sets\n",
    "tf.print(\n",
    "    \"Accuracy: \",\n",
    "    accuracy(x_test, y_test)\n",
    ")\n",
    "\n",
    "# Get one and predict\n",
    "r = random.randint(0, y_test_len - 1)\n",
    "tf.print(\n",
    "    \"Label: \", tf.argmax(y_test[r : r + 1], 1)\n",
    ")\n",
    "tf.print(\n",
    "    \"Prediction: \",\n",
    "    tf.argmax(hypothesis(x_test[r : r + 1]), 1),\n",
    ")\n",
    "\n",
    "plt.imshow(\n",
    "    tf.reshape(x_test[r : r + 1], [28, 28]),\n",
    "    cmap=\"Greys\",\n",
    "    interpolation=\"nearest\",\n",
    ")\n",
    "plt.show()\n",
    "\n",
    "\n",
    "'''\n",
    "Epoch: 0001, Cost: 2.826302672\n",
    "Epoch: 0002, Cost: 1.061668952\n",
    "Epoch: 0003, Cost: 0.838061315\n",
    "Epoch: 0004, Cost: 0.733232745\n",
    "Epoch: 0005, Cost: 0.669279885\n",
    "Epoch: 0006, Cost: 0.624611836\n",
    "Epoch: 0007, Cost: 0.591160344\n",
    "Epoch: 0008, Cost: 0.563868987\n",
    "Epoch: 0009, Cost: 0.541745171\n",
    "Epoch: 0010, Cost: 0.522673578\n",
    "Epoch: 0011, Cost: 0.506782325\n",
    "Epoch: 0012, Cost: 0.492447643\n",
    "Epoch: 0013, Cost: 0.479955837\n",
    "Epoch: 0014, Cost: 0.468893674\n",
    "Epoch: 0015, Cost: 0.458703488\n",
    "Learning finished\n",
    "Accuracy:  0.8951\n",
    "'''\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Label:  [0]\n",
      "Prediction:  [0]\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": [
    "\n",
    "# Get one and predict\n",
    "r = random.randint(0, y_test_len - 1)\n",
    "tf.print(\n",
    "    \"Label: \", tf.argmax(y_test[r : r + 1], 1)\n",
    ")\n",
    "tf.print(\n",
    "    \"Prediction: \",\n",
    "    tf.argmax(hypothesis(x_test[r : r + 1]), 1),\n",
    ")\n",
    "\n",
    "plt.imshow(\n",
    "    tf.reshape(x_test[r : r + 1], [28, 28]),\n",
    "    cmap=\"Greys\",\n",
    "    interpolation=\"nearest\",\n",
    ")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Label:  [6]\n",
      "Prediction:  [6]\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": [
    "\n",
    "# Get one and predict\n",
    "r = random.randint(0, y_test_len - 1)\n",
    "tf.print(\n",
    "    \"Label: \", tf.argmax(y_test[r : r + 1], 1)\n",
    ")\n",
    "tf.print(\n",
    "    \"Prediction: \",\n",
    "    tf.argmax(hypothesis(x_test[r : r + 1]), 1),\n",
    ")\n",
    "\n",
    "plt.imshow(\n",
    "    tf.reshape(x_test[r : r + 1], [28, 28]),\n",
    "    cmap=\"Greys\",\n",
    "    interpolation=\"nearest\",\n",
    ")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Label:  [0]\n",
      "Prediction:  [0]\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": [
    "\n",
    "# Get one and predict\n",
    "r = random.randint(0, y_test_len - 1)\n",
    "tf.print(\n",
    "    \"Label: \", tf.argmax(y_test[r : r + 1], 1)\n",
    ")\n",
    "tf.print(\n",
    "    \"Prediction: \",\n",
    "    tf.argmax(hypothesis(x_test[r : r + 1]), 1),\n",
    ")\n",
    "\n",
    "plt.imshow(\n",
    "    tf.reshape(x_test[r : r + 1], [28, 28]),\n",
    "    cmap=\"Greys\",\n",
    "    interpolation=\"nearest\",\n",
    ")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Label:  [1]\n",
      "Prediction:  [1]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD4CAYAAAAq5pAIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAMIUlEQVR4nO3db6gc9b3H8c9Hm4CmeRDNeu7BBNNb8uBKoWlY4gWleCmRmCdJH3htEElBmioGWqhwxT5oHmrpHypcCmkTerz0WgptMIjcmxBEDWh1DacaG3vjDblNmpBszINan7Sabx+csRzj2TmbnZmdvfm+X7DM7vxmZ74M53Nmdn6z+3NECMDV75q2CwAwHoQdSIKwA0kQdiAJwg4k8alxbmzlypWxZs2acW4SSOXkyZO6cOGCF2qrFHbbmyT9SNK1kn4aEY+XLb9mzRr1er0qmwRQotvtDmwb+TTe9rWS/l3S3ZJulbTN9q2jrg9As6p8Zt8g6Z2IOBERf5H0C0lb6ikLQN2qhP1mSafmvT5dzPsY2zts92z3+v1+hc0BqKJK2Be6CPCJe28jYndEdCOi2+l0KmwOQBVVwn5a0up5r1dJOlOtHABNqRL21ySttf0Z20slfUXS/nrKAlC3kbveIuID2zsl/bfmut72RsRbtVUGoFaV+tkj4jlJz9VUC4AGcbsskARhB5Ig7EAShB1IgrADSRB2IAnCDiRB2IEkCDuQBGEHkiDsQBKEHUiCsANJEHYgCcIOJEHYgSQIO5AEYQeSIOxAEoQdSIKwA0kQdiAJwg4kQdiBJAg7kARhB5Ig7EAShB1IgrADSVQaxRWTb3Z2trT9vvvuK21fvnx5afsrr7xyxTWhHZXCbvukpPckfSjpg4jo1lEUgPrVcWT/l4i4UMN6ADSIz+xAElXDHpIO2H7d9o6FFrC9w3bPdq/f71fcHIBRVQ377RGxXtLdkh62/cXLF4iI3RHRjYhup9OpuDkAo6oU9og4U0zPS9onaUMdRQGo38hht73M9vKPnku6S9LRugoDUK8qV+OnJO2z/dF6/jMi/quWqlCbnTt3lra//fbbldb/6quvlrZv2MDJ3qQYOewRcULS52usBUCD6HoDkiDsQBKEHUiCsANJEHYgCb7iepU7ceJEo+s/c+ZMo+tHfTiyA0kQdiAJwg4kQdiBJAg7kARhB5Ig7EAS9LNf5e69997S9ieffLLS+o8cOVLavnXr1krrR304sgNJEHYgCcIOJEHYgSQIO5AEYQeSIOxAEvSzX+XWrl3b6PrXr1/f6PpRH47sQBKEHUiCsANJEHYgCcIOJEHYgSQIO5AE/ezJRUSl99922201VYKmLXpkt73X9nnbR+fNu8H2QdvHi+mKZssEUNUwp/E/k7TpsnmPSjoUEWslHSpeA5hgi4Y9Il6UdPGy2VskzRTPZyTx20PAhBv1At1URJyVpGJ606AFbe+w3bPd6/f7I24OQFWNX42PiN0R0Y2IbqfTaXpzAAYYNeznbE9LUjE9X19JAJowatj3S9pePN8u6Zl6ygHQlEX72W0/LelOSSttn5b0HUmPS/ql7Qck/UHSPU0WiebYrvT+U6dOlbZPT09XWj/qs2jYI2LbgKYv1VwLgAZxuyyQBGEHkiDsQBKEHUiCsANJ8BVXVDI1NdV2CRgSR3YgCcIOJEHYgSQIO5AEYQeSIOxAEoQdSIJ+dlRyyy23tF0ChsSRHUiCsANJEHYgCcIOJEHYgSQIO5AEYQeSIOxAEoQdSIKwA0kQdiAJwg4kQdiBJAg7kARhB5Lg++zJXbp0qbS96pDOmByLHtlt77V93vbRefN22f6j7dnisbnZMgFUNcxp/M8kbVpg/g8jYl3xeK7esgDUbdGwR8SLki6OoRYADapygW6n7TeK0/wVgxayvcN2z3av3+9X2ByAKkYN+48lfVbSOklnJX1/0IIRsTsiuhHR7XQ6I24OQFUjhT0izkXEhxFxSdJPJG2otywAdRsp7Lan5738sqSjg5YFMBkW7We3/bSkOyWttH1a0nck3Wl7naSQdFLS1xusEQ265hruq8pi0bBHxLYFZu9poBYADeLfOpAEYQeSIOxAEoQdSIKwA0nwFVdUcvDgwdL2jRs3jqkSLIYjO5AEYQeSIOxAEoQdSIKwA0kQdiAJwg4kQT/7VW7TpoV+K7Q+77//fqPrR304sgNJEHYgCcIOJEHYgSQIO5AEYQeSIOxAEvSzX+WmpqYaXf+ePeU/NLx169ZGt4/hcWQHkiDsQBKEHUiCsANJEHYgCcIOJEHYgSToZ7/KLV26tLT9jjvuKG1/6aWXSttffvnl0vZ33313YNuNN95Y+l7Ua9Eju+3Vtp+3fcz2W7a/Ucy/wfZB28eL6YrmywUwqmFO4z+Q9K2I+CdJ/yzpYdu3SnpU0qGIWCvpUPEawIRaNOwRcTYijhTP35N0TNLNkrZImikWm5HEfZHABLuiC3S210j6gqTfSJqKiLPS3D8ESTcNeM8O2z3bvX6/X61aACMbOuy2Py3pV5K+GRF/GvZ9EbE7IroR0e10OqPUCKAGQ4Xd9hLNBf3nEfHrYvY529NF+7Sk882UCKAOi3a92bakPZKORcQP5jXtl7Rd0uPF9JlGKkQlS5YsKW1f7Cuohw8fLm2/ePFiaTtdb5NjmH722yXdL+lN27PFvMc0F/Jf2n5A0h8k3dNMiQDqsGjYI+KwJA9o/lK95QBoCrfLAkkQdiAJwg4kQdiBJAg7kARfcU1u8+bNpe2PPPJIpfU/+OCDA9ueffbZ0vdef/31lbaNj+PIDiRB2IEkCDuQBGEHkiDsQBKEHUiCsANJ0M+e3KpVq0rbN27cWNp+4MCB0vYXXnhhYNvMzMzANkl66KGHSttxZTiyA0kQdiAJwg4kQdiBJAg7kARhB5Ig7EAS9LMnt2zZstL2ffv2lbY/8cQTpe3Hjx8f2HbPPfz6+DhxZAeSIOxAEoQdSIKwA0kQdiAJwg4kQdiBJIYZn321pKck/YOkS5J2R8SPbO+S9DVJ/WLRxyLiuaYKRTuuu+660vZdu3aNpxBUNsxNNR9I+lZEHLG9XNLrtg8WbT+MiO81Vx6AugwzPvtZSWeL5+/ZPibp5qYLA1CvK/rMbnuNpC9I+k0xa6ftN2zvtb1iwHt22O7Z7vX7/YUWATAGQ4fd9qcl/UrSNyPiT5J+LOmzktZp7sj//YXeFxG7I6IbEd1Op1NDyQBGMVTYbS/RXNB/HhG/lqSIOBcRH0bEJUk/kbShuTIBVLVo2G1b0h5JxyLiB/PmT89b7MuSjtZfHoC6DHM1/nZJ90t60/ZsMe8xSdtsr5MUkk5K+nojFQKoxTBX4w9L8gJN9KkD/49wBx2QBGEHkiDsQBKEHUiCsANJEHYgCcIOJEHYgSQIO5AEYQeSIOxAEoQdSIKwA0kQdiAJR8T4Nmb3Jf3fvFkrJV0YWwFXZlJrm9S6JGobVZ213RIRC/7+21jD/omN272I6LZWQIlJrW1S65KobVTjqo3TeCAJwg4k0XbYd7e8/TKTWtuk1iVR26jGUlurn9kBjE/bR3YAY0LYgSRaCbvtTbZ/b/sd24+2UcMgtk/aftP2rO1ey7XstX3e9tF5826wfdD28WK64Bh7LdW2y/Yfi303a3tzS7Wttv287WO237L9jWJ+q/uupK6x7Lexf2a3fa2k/5G0UdJpSa9J2hYRvxtrIQPYPimpGxGt34Bh+4uS/izpqYj4XDHvu5IuRsTjxT/KFRHxbxNS2y5Jf257GO9itKLp+cOMS9oq6atqcd+V1PWvGsN+a+PIvkHSOxFxIiL+IukXkra0UMfEi4gXJV28bPYWSTPF8xnN/bGM3YDaJkJEnI2II8Xz9yR9NMx4q/uupK6xaCPsN0s6Ne/1aU3WeO8h6YDt123vaLuYBUxFxFlp7o9H0k0t13O5RYfxHqfLhhmfmH03yvDnVbUR9oWGkpqk/r/bI2K9pLslPVycrmI4Qw3jPS4LDDM+EUYd/ryqNsJ+WtLqea9XSTrTQh0LiogzxfS8pH2avKGoz300gm4xPd9yPX83ScN4LzTMuCZg37U5/HkbYX9N0lrbn7G9VNJXJO1voY5PsL2suHAi28sk3aXJG4p6v6TtxfPtkp5psZaPmZRhvAcNM66W913rw59HxNgfkjZr7or8/0r6dhs1DKjrHyX9tni81XZtkp7W3GndXzV3RvSApBslHZJ0vJjeMEG1/YekNyW9oblgTbdU2x2a+2j4hqTZ4rG57X1XUtdY9hu3ywJJcAcdkARhB5Ig7EAShB1IgrADSRB2IAnCDiTxNwDin1KhIxyKAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Get one and predict\n",
    "r = random.randint(0, y_test_len - 1)\n",
    "tf.print(\n",
    "    \"Label: \", tf.argmax(y_test[r : r + 1], 1)\n",
    ")\n",
    "tf.print(\n",
    "    \"Prediction: \",\n",
    "    tf.argmax(hypothesis(x_test[r : r + 1]), 1),\n",
    ")\n",
    "\n",
    "plt.imshow(\n",
    "    tf.reshape(x_test[r : r + 1], [28, 28]),\n",
    "    cmap=\"Greys\",\n",
    "    interpolation=\"nearest\",\n",
    ")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}