digit_recognition_ffnn_cnn.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "digit-recognition-ffnn-cnn.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/kyrcha/deep-learning-pipelines/blob/master/digit_recognition_ffnn_cnn.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "F0TLc64hYXit",
        "colab_type": "text"
      },
      "source": [
        "# Digit classification\n",
        "\n",
        "We will use the MNIST data of the colab notebooks to build a simple feed-forward neural network to classify digits from images."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Xu0YinOhYNGC",
        "colab_type": "code",
        "outputId": "d093d20a-3c55-4976-94f5-58890027613d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "import tensorflow as tf\n",
        "import math\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "\n",
        "print(tf.__version__)\n",
        "\n",
        "# This will go away in the future.\n",
        "# If this gives an error, you might be running TensorFlow 2 or above\n",
        "# If so, then just comment out this line and run this cell again\n",
        "tf.enable_eager_execution()  \n",
        "\n",
        "import logging\n",
        "logger = tf.get_logger()\n",
        "logger.setLevel(logging.ERROR)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "1.14.0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lVjB684tYujy",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "training_data = pd.read_csv(\"sample_data/mnist_train_small.csv\", header=None)\n",
        "testing_data = pd.read_csv(\"sample_data/mnist_test.csv\", header=None)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "r_W_nAE8ZkRE",
        "colab_type": "code",
        "outputId": "0c92f186-221b-4ad6-acfa-686d3799b02a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 253
        }
      },
      "source": [
        "training_data.head()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "      <th>5</th>\n",
              "      <th>6</th>\n",
              "      <th>7</th>\n",
              "      <th>8</th>\n",
              "      <th>9</th>\n",
              "      <th>10</th>\n",
              "      <th>11</th>\n",
              "      <th>12</th>\n",
              "      <th>13</th>\n",
              "      <th>14</th>\n",
              "      <th>15</th>\n",
              "      <th>16</th>\n",
              "      <th>17</th>\n",
              "      <th>18</th>\n",
              "      <th>19</th>\n",
              "      <th>20</th>\n",
              "      <th>21</th>\n",
              "      <th>22</th>\n",
              "      <th>23</th>\n",
              "      <th>24</th>\n",
              "      <th>25</th>\n",
              "      <th>26</th>\n",
              "      <th>27</th>\n",
              "      <th>28</th>\n",
              "      <th>29</th>\n",
              "      <th>30</th>\n",
              "      <th>31</th>\n",
              "      <th>32</th>\n",
              "      <th>33</th>\n",
              "      <th>34</th>\n",
              "      <th>35</th>\n",
              "      <th>36</th>\n",
              "      <th>37</th>\n",
              "      <th>38</th>\n",
              "      <th>39</th>\n",
              "      <th>...</th>\n",
              "      <th>745</th>\n",
              "      <th>746</th>\n",
              "      <th>747</th>\n",
              "      <th>748</th>\n",
              "      <th>749</th>\n",
              "      <th>750</th>\n",
              "      <th>751</th>\n",
              "      <th>752</th>\n",
              "      <th>753</th>\n",
              "      <th>754</th>\n",
              "      <th>755</th>\n",
              "      <th>756</th>\n",
              "      <th>757</th>\n",
              "      <th>758</th>\n",
              "      <th>759</th>\n",
              "      <th>760</th>\n",
              "      <th>761</th>\n",
              "      <th>762</th>\n",
              "      <th>763</th>\n",
              "      <th>764</th>\n",
              "      <th>765</th>\n",
              "      <th>766</th>\n",
              "      <th>767</th>\n",
              "      <th>768</th>\n",
              "      <th>769</th>\n",
              "      <th>770</th>\n",
              "      <th>771</th>\n",
              "      <th>772</th>\n",
              "      <th>773</th>\n",
              "      <th>774</th>\n",
              "      <th>775</th>\n",
              "      <th>776</th>\n",
              "      <th>777</th>\n",
              "      <th>778</th>\n",
              "      <th>779</th>\n",
              "      <th>780</th>\n",
              "      <th>781</th>\n",
              "      <th>782</th>\n",
              "      <th>783</th>\n",
              "      <th>784</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>7</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>9</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>15</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 785 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "   0    1    2    3    4    5    6    ...  778  779  780  781  782  783  784\n",
              "0    6    0    0    0    0    0    0  ...    0    0    0    0    0    0    0\n",
              "1    5    0    0    0    0    0    0  ...    0    0    0    0    0    0    0\n",
              "2    7    0    0    0    0    0    0  ...    0    0    0    0    0    0    0\n",
              "3    9    0    0    0    0    0    0  ...    0    0    0    0    0    0    0\n",
              "4    5    0    0    0    0    0    0  ...    0    0    0    0    0    0    0\n",
              "\n",
              "[5 rows x 785 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tF6Cj6KrZt9f",
        "colab_type": "code",
        "outputId": "e9149073-22de-419d-98b7-9aaac764eb2c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 253
        }
      },
      "source": [
        "testing_data.head()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "      <th>5</th>\n",
              "      <th>6</th>\n",
              "      <th>7</th>\n",
              "      <th>8</th>\n",
              "      <th>9</th>\n",
              "      <th>10</th>\n",
              "      <th>11</th>\n",
              "      <th>12</th>\n",
              "      <th>13</th>\n",
              "      <th>14</th>\n",
              "      <th>15</th>\n",
              "      <th>16</th>\n",
              "      <th>17</th>\n",
              "      <th>18</th>\n",
              "      <th>19</th>\n",
              "      <th>20</th>\n",
              "      <th>21</th>\n",
              "      <th>22</th>\n",
              "      <th>23</th>\n",
              "      <th>24</th>\n",
              "      <th>25</th>\n",
              "      <th>26</th>\n",
              "      <th>27</th>\n",
              "      <th>28</th>\n",
              "      <th>29</th>\n",
              "      <th>30</th>\n",
              "      <th>31</th>\n",
              "      <th>32</th>\n",
              "      <th>33</th>\n",
              "      <th>34</th>\n",
              "      <th>35</th>\n",
              "      <th>36</th>\n",
              "      <th>37</th>\n",
              "      <th>38</th>\n",
              "      <th>39</th>\n",
              "      <th>...</th>\n",
              "      <th>745</th>\n",
              "      <th>746</th>\n",
              "      <th>747</th>\n",
              "      <th>748</th>\n",
              "      <th>749</th>\n",
              "      <th>750</th>\n",
              "      <th>751</th>\n",
              "      <th>752</th>\n",
              "      <th>753</th>\n",
              "      <th>754</th>\n",
              "      <th>755</th>\n",
              "      <th>756</th>\n",
              "      <th>757</th>\n",
              "      <th>758</th>\n",
              "      <th>759</th>\n",
              "      <th>760</th>\n",
              "      <th>761</th>\n",
              "      <th>762</th>\n",
              "      <th>763</th>\n",
              "      <th>764</th>\n",
              "      <th>765</th>\n",
              "      <th>766</th>\n",
              "      <th>767</th>\n",
              "      <th>768</th>\n",
              "      <th>769</th>\n",
              "      <th>770</th>\n",
              "      <th>771</th>\n",
              "      <th>772</th>\n",
              "      <th>773</th>\n",
              "      <th>774</th>\n",
              "      <th>775</th>\n",
              "      <th>776</th>\n",
              "      <th>777</th>\n",
              "      <th>778</th>\n",
              "      <th>779</th>\n",
              "      <th>780</th>\n",
              "      <th>781</th>\n",
              "      <th>782</th>\n",
              "      <th>783</th>\n",
              "      <th>784</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>7</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 785 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "   0    1    2    3    4    5    6    ...  778  779  780  781  782  783  784\n",
              "0    7    0    0    0    0    0    0  ...    0    0    0    0    0    0    0\n",
              "1    2    0    0    0    0    0    0  ...    0    0    0    0    0    0    0\n",
              "2    1    0    0    0    0    0    0  ...    0    0    0    0    0    0    0\n",
              "3    0    0    0    0    0    0    0  ...    0    0    0    0    0    0    0\n",
              "4    4    0    0    0    0    0    0  ...    0    0    0    0    0    0    0\n",
              "\n",
              "[5 rows x 785 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XA-VKFcCZ1PX",
        "colab_type": "code",
        "outputId": "52ed88db-593e-48fe-e382-b2a6ac82a075",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "source": [
        "# get dims\n",
        "print(training_data.shape)\n",
        "print(testing_data.shape)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(20000, 785)\n",
            "(10000, 785)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Tth1KsvQaEqQ",
        "colab_type": "text"
      },
      "source": [
        "## Pre-processing\n",
        "\n",
        "We will normalize the pixel values between [0,1] by dividing with 255.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KuShv7ozaBAF",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "y_train = training_data[0]\n",
        "y_test = testing_data[0]\n",
        "X_train = training_data.drop(0, axis=1).divide(255)\n",
        "X_test = testing_data.drop(0, axis=1).divide(255)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GwAwyldvaao5",
        "colab_type": "code",
        "outputId": "6b9ce0d2-d61e-481d-b7fc-b732297b288c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 286
        }
      },
      "source": [
        "image = np.asarray(X_train.iloc[0]).reshape((28,28))\n",
        "print('Label:',y_train.iloc[0])\n",
        "\n",
        "# Plot the image - voila a piece of fashion clothing\n",
        "plt.figure()\n",
        "plt.imshow(image, cmap=plt.cm.binary)\n",
        "plt.colorbar()\n",
        "plt.grid(False)\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Label: 6\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAS4AAAD8CAYAAADJwUnTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n7LhwmVl2XLjMLDsuXGaWHRcuM8vO/wNNeA6lY63NeAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "p0EoemyybQEK",
        "colab_type": "code",
        "outputId": "f2ed90a7-fe37-4038-c708-af9122f43730",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 592
        }
      },
      "source": [
        "# Plot the next 25\n",
        "plt.figure(figsize=(10,10))\n",
        "i = 0\n",
        "for j in range(0,25):\n",
        "    image = np.asarray(X_train.iloc[j]).reshape((28,28))\n",
        "    plt.subplot(5,5,i+1)\n",
        "    plt.xticks([])\n",
        "    plt.yticks([])\n",
        "    plt.grid(False)\n",
        "    plt.imshow(image, cmap=plt.cm.binary)\n",
        "    plt.xlabel(y_train.iloc[j])\n",
        "    i += 1\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkAAAAI/CAYAAACf7mYiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
            "text/plain": [
              "<Figure size 720x720 with 25 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kMCP-O1wd3Vs",
        "colab_type": "text"
      },
      "source": [
        "## Model\n",
        "\n",
        "Let's build the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cIVt1KJ2dDso",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model = tf.keras.Sequential([\n",
        "    tf.keras.layers.Dense(128, activation=tf.nn.relu, input_shape=(784,)),\n",
        "    tf.keras.layers.Dense(10,  activation=tf.nn.softmax)\n",
        "])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oNsl9cDDd64g",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model.compile(optimizer='adam', \n",
        "              loss='sparse_categorical_crossentropy',\n",
        "              metrics=['accuracy'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "N2nR-8tBeF7O",
        "colab_type": "text"
      },
      "source": [
        "## Train the model\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nrAsDUleeC1R",
        "colab_type": "code",
        "outputId": "c642abdf-7ed5-4c3f-ab38-47be17b46d4d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        }
      },
      "source": [
        "BATCH_SIZE = 32\n",
        "\n",
        "model.fit(x = X_train, y = y_train, epochs=5, batch_size=BATCH_SIZE, steps_per_epoch=math.ceil(X_train.shape[0]/BATCH_SIZE))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/5\n",
            "625/625 [==============================] - 47s 75ms/step - loss: 0.1567 - acc: 0.9609\n",
            "Epoch 2/5\n",
            "625/625 [==============================] - 46s 74ms/step - loss: 0.0079 - acc: 0.9998\n",
            "Epoch 3/5\n",
            "625/625 [==============================] - 46s 73ms/step - loss: 0.0018 - acc: 1.0000\n",
            "Epoch 4/5\n",
            "625/625 [==============================] - 46s 74ms/step - loss: 7.6392e-04 - acc: 1.0000\n",
            "Epoch 5/5\n",
            "625/625 [==============================] - 46s 73ms/step - loss: 3.9443e-04 - acc: 1.0000\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7f4c71e80128>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Celua9DYgsGO",
        "colab_type": "text"
      },
      "source": [
        "## Evaluation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Qhm8FAq4eKPP",
        "colab_type": "code",
        "outputId": "cf2fff71-1478-40a1-8d0f-3d19d90f9194",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "source": [
        "test_loss, test_accuracy = model.evaluate(x = X_test, y = y_test, steps=math.ceil(testing_data.shape[0]/32))\n",
        "print('Accuracy on test dataset:', test_accuracy)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "313/313 [==============================] - 11s 36ms/step - loss: 0.2069 - acc: 0.9599\n",
            "Accuracy on test dataset: 0.9599\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Cj0sso8tnFOr",
        "colab_type": "text"
      },
      "source": [
        "# CNNs\n",
        "\n",
        "Now let's do the same with CNNs"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aWBZ0gLOoKdR",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "train_img = X_train.values.reshape(X_train.shape[0], 28,28,1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oeMAa3m0g89J",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "cnn = tf.keras.Sequential([\n",
        "    tf.keras.layers.Conv2D(32, (3,3), padding='same', activation=tf.nn.relu,\n",
        "                           input_shape=(28,28,1)),\n",
        "    tf.keras.layers.MaxPooling2D((2, 2), strides=2),\n",
        "    tf.keras.layers.Conv2D(64, (3,3), padding='same', activation=tf.nn.relu),\n",
        "    tf.keras.layers.MaxPooling2D((2, 2), strides=2),\n",
        "    tf.keras.layers.Flatten(),\n",
        "    tf.keras.layers.Dense(128, activation=tf.nn.relu),\n",
        "    tf.keras.layers.Dense(10,  activation=tf.nn.softmax)\n",
        "])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0axaK6iEsdgl",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "cnn.compile(optimizer='adam', \n",
        "              loss='sparse_categorical_crossentropy',\n",
        "              metrics=['accuracy'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0ldWvQS3nITd",
        "colab_type": "code",
        "outputId": "52625cfa-f73f-4285-9209-fb4d39175ce8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        }
      },
      "source": [
        "cnn.fit(x=train_img, y=y_train, epochs=3, steps_per_epoch=math.ceil(X_train.shape[0]/BATCH_SIZE))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/3\n",
            "625/625 [==============================] - 273s 437ms/step - loss: 2.0012e-05 - acc: 1.0000\n",
            "Epoch 2/3\n",
            "625/625 [==============================] - 273s 437ms/step - loss: 9.5733e-06 - acc: 1.0000\n",
            "Epoch 3/3\n",
            "625/625 [==============================] - 274s 438ms/step - loss: 5.5347e-06 - acc: 1.0000\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7f4c6f58e5f8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_9qYn6ZQsPnV",
        "colab_type": "code",
        "outputId": "b855fe0a-fe95-4bf5-b67d-ffac95580815",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "source": [
        "test_img = X_test.values.reshape(X_test.shape[0], 28,28,1)\n",
        "test_loss, test_accuracy = cnn.evaluate(x = test_img, y = y_test, steps=math.ceil(testing_data.shape[0]/32))\n",
        "print('Accuracy on test dataset:', test_accuracy)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "313/313 [==============================] - 29s 92ms/step - loss: 0.0970 - acc: 0.9849\n",
            "Accuracy on test dataset: 0.9849\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ge2Q9SSbzwjU",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}