Exercise_Neural_Networks_4.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "mount_file_id": "1AqMgvg_LtBdu8Ltb2L_9wUx1V8QZvvTq",
      "authorship_tag": "ABX9TyPP0lfufiU77hR7uk9QX4jf",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/yasarigno/NLP_DeepLearning_Course/blob/main/Exercise_Neural_Networks_4.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 1. Import the data from Google Drive"
      ],
      "metadata": {
        "id": "ry9n2SFF6cvi"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "6joLSRXZhk1l"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "%matplotlib inline\n",
        "\n",
        "from tensorflow.keras.layers import Dense\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "from tensorflow.keras.models import Sequential"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import random\n",
        "random.seed(2)\n",
        "\n",
        "import tensorflow as tf\n",
        "tf.random.set_seed(2)\n",
        "\n",
        "import numpy as np\n",
        "np.random.seed(2)"
      ],
      "metadata": {
        "id": "y0Ew4FmSiaQC"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Local = False"
      ],
      "metadata": {
        "id": "Fk0ATv7njPLn"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "if Local:\n",
        "    traffic_data = pd.read_csv('datasets/traffic_data.csv', index_col=0)\n",
        "else:\n",
        "    traffic_data = pd.read_csv(\"/content/drive/MyDrive/Shared Datasets/Traffic/traffic_data.csv\", index_col=0)"
      ],
      "metadata": {
        "id": "GctBDj15idIu"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "The last 3 columns specify the pizza type:\n",
        "\n",
        "Vegan\n",
        "Vegetarian\n",
        "Meat"
      ],
      "metadata": {
        "id": "7gjMWcewikmt"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "traffic_data.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "HUpIR1_sifjB",
        "outputId": "a59095ce-9397-48f6-f2a3-be6889712328"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "      day  minute  hour  second     type\n",
              "0  Monday       0     8      14  traffic\n",
              "1  Monday       0     8      28  traffic\n",
              "2  Monday       0     8      34  traffic\n",
              "3  Monday       0     8      45  traffic\n",
              "4  Monday       0     8      53  traffic"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-92fd587d-850e-4adc-bda5-7f3d5e5defab\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>day</th>\n",
              "      <th>minute</th>\n",
              "      <th>hour</th>\n",
              "      <th>second</th>\n",
              "      <th>type</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Monday</td>\n",
              "      <td>0</td>\n",
              "      <td>8</td>\n",
              "      <td>14</td>\n",
              "      <td>traffic</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Monday</td>\n",
              "      <td>0</td>\n",
              "      <td>8</td>\n",
              "      <td>28</td>\n",
              "      <td>traffic</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Monday</td>\n",
              "      <td>0</td>\n",
              "      <td>8</td>\n",
              "      <td>34</td>\n",
              "      <td>traffic</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Monday</td>\n",
              "      <td>0</td>\n",
              "      <td>8</td>\n",
              "      <td>45</td>\n",
              "      <td>traffic</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Monday</td>\n",
              "      <td>0</td>\n",
              "      <td>8</td>\n",
              "      <td>53</td>\n",
              "      <td>traffic</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-92fd587d-850e-4adc-bda5-7f3d5e5defab')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-92fd587d-850e-4adc-bda5-7f3d5e5defab button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-92fd587d-850e-4adc-bda5-7f3d5e5defab');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 2. EDA"
      ],
      "metadata": {
        "id": "St0cGrBD6ZU-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "traffic_data.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LMLG8Xl2-E7i",
        "outputId": "8ab30418-261d-42e5-b8d2-3a9a589b04e6"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(85638, 5)"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "traffic_data['type'].value_counts()"
      ],
      "metadata": {
        "id": "bPCSE5L-inXy",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3a8c7efc-8281-4918-f1c6-3b946f35e453"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "no_traffic    50769\n",
              "traffic       34869\n",
              "Name: type, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Notice that the dataset is imbalanced."
      ],
      "metadata": {
        "id": "UFeRxSaQ6S-y"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "traffic_data['day'].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zI65c8sWipQ_",
        "outputId": "eacaf95c-c4ea-4572-ba1e-cb0f3a12020f"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Sunday       12278\n",
              "Saturday     12251\n",
              "Thursday     12241\n",
              "Wednesday    12237\n",
              "Tuesday      12217\n",
              "Monday       12212\n",
              "Friday       12202\n",
              "Name: day, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Group the types by day"
      ],
      "metadata": {
        "id": "5NLXBrHj6yBc"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "traffic_data.groupby('day')['type'].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ICwbwqwIirVs",
        "outputId": "72a4368f-ba1d-42b5-bd3b-eaff1e46052f"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "day        type      \n",
              "Friday     traffic        6989\n",
              "           no_traffic     5213\n",
              "Monday     traffic        6972\n",
              "           no_traffic     5240\n",
              "Saturday   no_traffic    12251\n",
              "Sunday     no_traffic    12278\n",
              "Thursday   traffic        6965\n",
              "           no_traffic     5276\n",
              "Tuesday    traffic        6990\n",
              "           no_traffic     5227\n",
              "Wednesday  traffic        6953\n",
              "           no_traffic     5284\n",
              "Name: type, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "traffic_data['hour'].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NRP8-w4diuKj",
        "outputId": "6e4bbe71-eca7-42f5-a36b-ec555a24db47"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "8     6211\n",
              "16    6204\n",
              "20    6195\n",
              "18    6188\n",
              "14    6180\n",
              "10    6156\n",
              "12    6147\n",
              "15    6110\n",
              "17    6076\n",
              "13    6052\n",
              "11    6046\n",
              "21    6046\n",
              "19    6036\n",
              "9     5988\n",
              "22       3\n",
              "Name: hour, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Hours start at 8 and stop at 22. \n",
        "22 has very little data."
      ],
      "metadata": {
        "id": "rWye06mA66x3"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "traffic_data['c_type'] = traffic_data['type'].apply(lambda x: 1 if x == 'traffic' else 0)"
      ],
      "metadata": {
        "id": "8BZSxHeEiu4V"
      },
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Visualisation by scatter plot"
      ],
      "metadata": {
        "id": "BSYPJ0Bw7IRR"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "from matplotlib.pyplot import colorbar, figure"
      ],
      "metadata": {
        "id": "BlFXVqUzi0SX"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "figure(num=None, figsize=(12, 10))\n",
        "plt.scatter(\n",
        "    traffic_data['day'], \n",
        "    traffic_data['hour'], \n",
        "    c=traffic_data['c_type'], \n",
        "    cmap='jet',\n",
        ")\n",
        "cbar = colorbar()\n"
      ],
      "metadata": {
        "id": "UalHJyoKi02f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 596
        },
        "outputId": "f9e00b25-b7b2-432f-c754-6ea92ae1aad7"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x720 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "there is no traffic on Saturdays or Sundays. It seems to be mostly around some key hours, such as 8-10, 12-13, and 16-20."
      ],
      "metadata": {
        "id": "TgwzQVYb7QyQ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "traffic_data = traffic_data.join(pd.get_dummies(traffic_data['day']))"
      ],
      "metadata": {
        "id": "G-LU1Uv3i61W"
      },
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "traffic_data"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "PQAjbDTP7XhJ",
        "outputId": "c08757a1-5404-4cff-accb-c98ac7742f4b"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          day  minute  hour  second        type  c_type  Friday  Monday  \\\n",
              "0      Monday       0     8      14     traffic       1       0       1   \n",
              "1      Monday       0     8      28     traffic       1       0       1   \n",
              "2      Monday       0     8      34     traffic       1       0       1   \n",
              "3      Monday       0     8      45     traffic       1       0       1   \n",
              "4      Monday       0     8      53     traffic       1       0       1   \n",
              "...       ...     ...   ...     ...         ...     ...     ...     ...   \n",
              "98042  Sunday      59    17      51  no_traffic       0       0       0   \n",
              "98043  Sunday       0    18       0  no_traffic       0       0       0   \n",
              "98044  Sunday       0    16      24  no_traffic       0       0       0   \n",
              "98045  Sunday       0    16      35  no_traffic       0       0       0   \n",
              "98047  Sunday       1    16      28  no_traffic       0       0       0   \n",
              "\n",
              "       Saturday  Sunday  Thursday  Tuesday  Wednesday  \n",
              "0             0       0         0        0          0  \n",
              "1             0       0         0        0          0  \n",
              "2             0       0         0        0          0  \n",
              "3             0       0         0        0          0  \n",
              "4             0       0         0        0          0  \n",
              "...         ...     ...       ...      ...        ...  \n",
              "98042         0       1         0        0          0  \n",
              "98043         0       1         0        0          0  \n",
              "98044         0       1         0        0          0  \n",
              "98045         0       1         0        0          0  \n",
              "98047         0       1         0        0          0  \n",
              "\n",
              "[85638 rows x 13 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-715cdedb-ae26-4e0f-bebf-6252f55f6d3c\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>day</th>\n",
              "      <th>minute</th>\n",
              "      <th>hour</th>\n",
              "      <th>second</th>\n",
              "      <th>type</th>\n",
              "      <th>c_type</th>\n",
              "      <th>Friday</th>\n",
              "      <th>Monday</th>\n",
              "      <th>Saturday</th>\n",
              "      <th>Sunday</th>\n",
              "      <th>Thursday</th>\n",
              "      <th>Tuesday</th>\n",
              "      <th>Wednesday</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Monday</td>\n",
              "      <td>0</td>\n",
              "      <td>8</td>\n",
              "      <td>14</td>\n",
              "      <td>traffic</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\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",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Monday</td>\n",
              "      <td>0</td>\n",
              "      <td>8</td>\n",
              "      <td>28</td>\n",
              "      <td>traffic</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\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",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Monday</td>\n",
              "      <td>0</td>\n",
              "      <td>8</td>\n",
              "      <td>34</td>\n",
              "      <td>traffic</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\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",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Monday</td>\n",
              "      <td>0</td>\n",
              "      <td>8</td>\n",
              "      <td>45</td>\n",
              "      <td>traffic</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\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",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Monday</td>\n",
              "      <td>0</td>\n",
              "      <td>8</td>\n",
              "      <td>53</td>\n",
              "      <td>traffic</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\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",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>98042</th>\n",
              "      <td>Sunday</td>\n",
              "      <td>59</td>\n",
              "      <td>17</td>\n",
              "      <td>51</td>\n",
              "      <td>no_traffic</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>98043</th>\n",
              "      <td>Sunday</td>\n",
              "      <td>0</td>\n",
              "      <td>18</td>\n",
              "      <td>0</td>\n",
              "      <td>no_traffic</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>98044</th>\n",
              "      <td>Sunday</td>\n",
              "      <td>0</td>\n",
              "      <td>16</td>\n",
              "      <td>24</td>\n",
              "      <td>no_traffic</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>98045</th>\n",
              "      <td>Sunday</td>\n",
              "      <td>0</td>\n",
              "      <td>16</td>\n",
              "      <td>35</td>\n",
              "      <td>no_traffic</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>98047</th>\n",
              "      <td>Sunday</td>\n",
              "      <td>1</td>\n",
              "      <td>16</td>\n",
              "      <td>28</td>\n",
              "      <td>no_traffic</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>85638 rows × 13 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-715cdedb-ae26-4e0f-bebf-6252f55f6d3c')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-715cdedb-ae26-4e0f-bebf-6252f55f6d3c button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-715cdedb-ae26-4e0f-bebf-6252f55f6d3c');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Normalization?"
      ],
      "metadata": {
        "id": "IXgrDe6n7iy7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "training_dataset = traffic_data.sample(frac=0.8)\n",
        "testing_dataset = traffic_data[~traffic_data.index.isin(training_dataset.index)]"
      ],
      "metadata": {
        "id": "q88PPT-0i807"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "input_columns = [\n",
        "    'Monday', \n",
        "    'Tuesday', \n",
        "    'Wednesday', \n",
        "    'Thursday', \n",
        "    'Friday', \n",
        "    'Saturday', \n",
        "    'Sunday', \n",
        "    'hour', \n",
        "    'minute', \n",
        "    'second'\n",
        "]"
      ],
      "metadata": {
        "id": "tu2iRxh7jAm_"
      },
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "We have to build a network with more layers, as the function that needs to fit this data is now more complex.\n",
        "\n",
        "---\n",
        "\n",
        "There is no hard and fast rule for the number of layers or neurons, but it is useful to think about the networks like this:\n",
        "\n",
        "- If you add a layer with more neurons than the previous layer or the input, you add more information into the network. More neurons create more signals for the next layer.\n",
        "\n",
        "- If you add a layer with fewer neurons than the previous layer or the input, you compress information. Now, there will be fewer signals for the next layer.\n",
        "\n",
        "- In general, you would use your intuition and experimentation (similarity to other problems) to select the number of neurons.\n",
        "\n",
        "---\n",
        "\n",
        "As a starting point, use three layers with 50 neurons in the input and hidden layer, and just one in the output layer. If you find it works, you can try to reduce that number and see what happens. If it doesn't work, you can always try to increase it.\n",
        "\n",
        "There are two more changes to make: one, to the activation function, and the second, to the dropout rate."
      ],
      "metadata": {
        "id": "VUgx2LrX7pe8"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.layers import Dropout\n"
      ],
      "metadata": {
        "id": "kErZUJ6O-msm"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "traffic_model = Sequential([\n",
        "    Dense(32, input_dim=len(input_columns), activation='relu'),\n",
        "    Dropout(0.1),\n",
        "    Dense(32, activation='relu'),\n",
        "    Dropout(0.1),\n",
        "    Dense(1, activation='sigmoid'),\n",
        "])\n",
        "\n",
        "adam = Adam()\n",
        "\n",
        "traffic_model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
        "traffic_model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ni_PO6AQ-nhC",
        "outputId": "5e8f99c8-87f1-414f-95e1-9dc64aa4844c"
      },
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " dense (Dense)               (None, 32)                352       \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 32)                0         \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 32)                1056      \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 32)                0         \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 1)                 33        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 1,441\n",
            "Trainable params: 1,441\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Remember that the activation function determines if that particular unit should activate or become insignificant.\n",
        "\n",
        "The activation function of the input and hidden layers to use a rectified linear unit (ReLU)\n",
        "\n",
        "Two possible problems :\n",
        "\n",
        "1. Gradient vanishing is a phenomenon that happens when the error that backpropagation pushes back throughout the network gets smaller and smaller as it moves from right to left (output layer towards input layer). That means that the weights of neurons in the first layers end up being updated by minimal values. When this happens, the weights take too long to get to the required value, and the network takes a lot longer to learn.\n",
        "\n",
        "2. Having too many neurons and layers make the network prone to overfitting. During weight updating, your model already penalizes weights via a method called L2 regularization. It ensures that weights never get too large.\n",
        "\n",
        "---\n",
        "\n",
        "Remember also that training a neural network consists of two steps: a forward pass and a backward pass.\n",
        "\n",
        "When you add dropout to a layer on the forward pass, a fraction of the neurons don't activate, and the next layer receives a 0 value instead of the neuron's result. On the backward pass, the same neurons that ended up not activating also don't get their weights updated. They are basically frozen in that epoch.\n",
        "\n",
        "Dropout randomly selects some neurons to skip during training to prevent overfitting.\n",
        "\n",
        "In large networks, neurons start being reliant on certain preceding neurons that end up hindering learning. By employing dropout, those preceding neurons are randomly dropped, which breaks the formed dependency.\n",
        "\n",
        "Research shows that the best dropout rate is 50%; however, you should start small as dropout can stop a network from learning. I recommend you use a 10% dropout rate here, but you should try larger numbers of neurons and higher dropouts and see how the network behaves during training:"
      ],
      "metadata": {
        "id": "sZkvmmYh8Tvz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "batch_size = 100"
      ],
      "metadata": {
        "id": "tLcI_6ZH-s-S"
      },
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "history_traffic_model = traffic_model.fit(\n",
        "    training_dataset[input_columns], \n",
        "    training_dataset[['c_type']], \n",
        "    epochs=30,\n",
        "    validation_split=0.1,\n",
        "    batch_size=batch_size\n",
        ")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EFMFHBdb-txi",
        "outputId": "e8f33b43-decd-4dd9-8c90-d5820c303d17"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/30\n",
            "617/617 [==============================] - 3s 4ms/step - loss: 0.9406 - accuracy: 0.5588 - val_loss: 0.6463 - val_accuracy: 0.6636\n",
            "Epoch 2/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.6101 - accuracy: 0.6400 - val_loss: 0.5285 - val_accuracy: 0.7065\n",
            "Epoch 3/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.5473 - accuracy: 0.6748 - val_loss: 0.4969 - val_accuracy: 0.6998\n",
            "Epoch 4/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.5095 - accuracy: 0.6834 - val_loss: 0.4861 - val_accuracy: 0.6961\n",
            "Epoch 5/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.4932 - accuracy: 0.6929 - val_loss: 0.4830 - val_accuracy: 0.6951\n",
            "Epoch 6/30\n",
            "617/617 [==============================] - 4s 7ms/step - loss: 0.4888 - accuracy: 0.6959 - val_loss: 0.4812 - val_accuracy: 0.6981\n",
            "Epoch 7/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.4840 - accuracy: 0.7055 - val_loss: 0.4701 - val_accuracy: 0.7348\n",
            "Epoch 8/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.4748 - accuracy: 0.7211 - val_loss: 0.4549 - val_accuracy: 0.7529\n",
            "Epoch 9/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.4500 - accuracy: 0.7497 - val_loss: 0.4081 - val_accuracy: 0.7897\n",
            "Epoch 10/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.4037 - accuracy: 0.7725 - val_loss: 0.3430 - val_accuracy: 0.8019\n",
            "Epoch 11/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.3537 - accuracy: 0.7944 - val_loss: 0.3052 - val_accuracy: 0.8136\n",
            "Epoch 12/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.3138 - accuracy: 0.8163 - val_loss: 0.2521 - val_accuracy: 0.8552\n",
            "Epoch 13/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.2867 - accuracy: 0.8409 - val_loss: 0.2482 - val_accuracy: 0.8660\n",
            "Epoch 14/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.2742 - accuracy: 0.8464 - val_loss: 0.2312 - val_accuracy: 0.9016\n",
            "Epoch 15/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.2494 - accuracy: 0.8517 - val_loss: 0.1971 - val_accuracy: 0.8440\n",
            "Epoch 16/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.2368 - accuracy: 0.8338 - val_loss: 0.1790 - val_accuracy: 0.8602\n",
            "Epoch 17/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.2227 - accuracy: 0.8820 - val_loss: 0.2140 - val_accuracy: 0.9034\n",
            "Epoch 18/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.2149 - accuracy: 0.9098 - val_loss: 0.1536 - val_accuracy: 0.9977\n",
            "Epoch 19/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.2210 - accuracy: 0.9174 - val_loss: 0.1648 - val_accuracy: 0.9745\n",
            "Epoch 20/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.2021 - accuracy: 0.9263 - val_loss: 0.1447 - val_accuracy: 0.9888\n",
            "Epoch 21/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.2003 - accuracy: 0.9305 - val_loss: 0.1404 - val_accuracy: 1.0000\n",
            "Epoch 22/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.1965 - accuracy: 0.9331 - val_loss: 0.1240 - val_accuracy: 1.0000\n",
            "Epoch 23/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.1874 - accuracy: 0.9377 - val_loss: 0.1163 - val_accuracy: 1.0000\n",
            "Epoch 24/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.1851 - accuracy: 0.9404 - val_loss: 0.1242 - val_accuracy: 0.9999\n",
            "Epoch 25/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.1814 - accuracy: 0.9431 - val_loss: 0.1104 - val_accuracy: 1.0000\n",
            "Epoch 26/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.1905 - accuracy: 0.9423 - val_loss: 0.1111 - val_accuracy: 0.9999\n",
            "Epoch 27/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.1715 - accuracy: 0.9492 - val_loss: 0.1015 - val_accuracy: 1.0000\n",
            "Epoch 28/30\n",
            "617/617 [==============================] - 2s 3ms/step - loss: 0.1724 - accuracy: 0.9500 - val_loss: 0.1075 - val_accuracy: 0.9958\n",
            "Epoch 29/30\n",
            "617/617 [==============================] - 3s 5ms/step - loss: 0.1688 - accuracy: 0.9507 - val_loss: 0.1254 - val_accuracy: 0.9820\n",
            "Epoch 30/30\n",
            "617/617 [==============================] - 2s 4ms/step - loss: 0.1697 - accuracy: 0.9536 - val_loss: 0.1057 - val_accuracy: 0.9975\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The batch size represents how many examples the algorithm predicts during training before the weights get updated.\n",
        "\n",
        "The parameter does have an impact on training time with larger batches making it faster to train, but large batches have also been shown to make it harder for the algorithm to generalize. A good rule of thumb is to ensure the batch fits into the memory.\n",
        "\n"
      ],
      "metadata": {
        "id": "xsSQntm5-7xe"
      }
    }
  ]
}